Of Cogs, Bozos, and Backbones
Okay, time for another post-break of random goodness...
The post The incident at the WinClient All Hands by Drie over at Of Cogs (and other aspects of life inside MS) provides some insight into a BrianV meeting and guest-star Kevin Johnson. Kevin: sounds like Papa needs to get a brand-new shtick before everyone knows the punch-line. Having clarity and enthralling indignation is easy come resignation-letter writing time. You could say this whole blog is one long resignation letter, should Microsoft not re-invent itself into the Next Microsoft.
I think everyone would be a lot more impressed with a contemporary update of that letter, written to Microsofties, our customers, our partners, and our shareholders. Not resigning, but frankly admitting where we were, where we are, and where we need to be to be successful and how we're getting there.
Is Google not all that it appears to be? I'm still waiting for that Mini-Google site to fire itself up. Perhaps being hosted at MSN Spaces for a delightful bit of irony? Anyway, Valley Wag is a fun read, between letting us know that Googlers have pay-raise issues, too, and that Larry Page is just this close from throwing chairs over bad 20% usage.
Mark Shavlik, an ex-Microsoftie and now CEO, has an interesting post: Microsoft employee perspectives. Snippet:
Rapid success is not always good, growth comes fast and internal structural changes come slower creating a company that could be out of balance. Hiring can go too fast, where the quality of available to hire talent is not large enough to meet the needs of the business. That is a tough point for a company. One of the HP founders apparently said a company should only grow as fast as it can find quality people to hire, but if you do that do you miss markets? I am not sure, but I have found its better to grow at the rate at which you can hire great people.
And some people were so desperate to fill positions they'd hire just about any bozo, or bozos to fill up the 3.0 quota to reward their current team. What do we do with the bozo's that have us going around making us flip the bozo-bit all the time? First rule: don't hire anymore bozos! I think we've already had a bozo explosion and now we're dealing with the aftermath from explosive growth and low standards. Anyway, we shouldn't want to do that again. Guy Kawasaki put up the initial post of how to avoid this mediocrity bozo bloomer, followed by Scoble's take from a Microsoft-bozo / anti-bozo point of view.
Though I'm afraid the bozos are going to starting harrumphing "Scrum! We got ourselves some scrum, right here in Salmonburg city!"
Results. Accountable results.
As for the occasional external commentary on the Mini-Microsoft blog, I'm seeing a lot more of this as of late:
[...] Same goes for blogging if the blog is relevant, although I doubt that Mini-Microsoft will be listing his on a resume anytime soon...:) Speaking of Mini, anybody notice lately that his/her blog has basically degenerated into a bitch-fest for MS employees? Mini throws out the raw meat for a few sentences or paragraphs, and the piling on in the comments commences immediately. I read the comments more than his stuff now...lol
This past post was probably a prime-candidate for that observation. I went into low-moderation mode, just about making it appear as if moderation had been turned off. Some folks put up some great discussions, and some folks went off on wild, argumentative tangents. And some folks broke out the "nyah-nyah" factor that at least makes everything else look better by comparison.
Anyway, it's an interesting conundrum that I've typed myself into. How can you be critical, perhaps even constructive in your criticism, and not become the patron saint of the eternal complainer? Let alone a negativity magnet? Perhaps the very people you're trying to hunt down and move out of the company? The whole "bad attitude" post was perhaps too oblique in taking that on.
To quote Jeff Raikes... it's a pickle.
As for this past post, there was a nice thread of comments discussing loosening up the internal hiring process. One commenter waved a yellow flag, though, cautioning about what it was like at Enron from a note put up by Malcolm Gladwell:
Among the most damning facts about Enron, in the end, was something its managers were proudest of. They had what, in McKinsey terminology, is called an "open market" for hiring. In the open-market system--McKinsey's assault on the very idea of a fixed organization--anyone could apply for any job that he or she wanted, and no manager was allowed to hold anyone back. Poaching was encouraged.
Yes, if we start trending towards that you'll see posters slapped up every week for each group with new, open positions, leading to wine and cheese events where you'll hear embolden tales of innovating against all the odds... so I don't think we need to go that far. People shouldn't be burning through the corporation leaping from group to group every quarter. But you shouldn't be inhibiting them, either, worried about the worse case offenders. All this delay of being able to interview represents one of the worse behaviors of fiefdom building... people getting lost inside the corporation once they are hired just isn't productive or worth all the money Microsoft and the shareholders have invested in hiring them and training them.
(Actually, that whole article from Malcolm Gladwell, The Talent Myth, reprinted from the New Yorker, is an excellent read.)
As for financials, Microsoft Nears Xbox Equilibrium has this nice little challenge to Liddell regarding Mr. Liddell's preference of buybacks over dividends:
...a large asset management company who questioned Liddell from the floor (and was thus anonymous under conference ground rules.) "Microsoft is still overcapitalized," he said. "And that's why the stock hasn't moved," he said.
He added that investors don't want to see Microsoft embark on a series of acquisitions. "The market fears that Microsoft will throw a lot of money to buy media or Internet-related companies. "We're strongly opposed," he said.
I hear backbone cracking into firm alignment! What have analysts told us before? A significant dividend on a reliable schedule would boost the stock. And, yes, please, no more acquisitions. How about this week's Microsoft Research Tech Fest show off all of the innovation we're quite capable of without having to go and buy companies? Every acquisition is a failure on our part to innovate and another reason we have to ask, "And why do we have all these people around?"
Finally, looks like BillG and I might agree on reducing the amount of folks working on software development at Microsoft. One commenter notes that LisaB's internal site has a version of that wonderful Ten Crazy Ideas paper that includes feedback, including from BillG and the ever so dreamy AlexGo. One the point about reducing dev team headcounts:
I think Mini would be heartened by one of Bill's comments regarding reducing headcounts on large dev projects:
BillG: “Headcount cuts. This is very tempting.”
Tempting indeed. Give into temptation! Give our bozos the bums rush to B-as-in-Bozo-FE!
172 comments:
What scares me even more is this CNN article I found on how Google and its artificial intelligence capabilities will enslave us all. And no, I'm not some paranoid doomsday sayer - You have respected technologists like Ray Kurzweil predicting this:
http://money.cnn.com/magazines/business2/business2_archive/2006/01/01/8368125/
So think about it - a Microsoft monopoly in the desktop, server, application development and consumer space, or a Google monopoly over life as we know it?
And some people were so desperate to fill positions they'd hire just about any bozo, or bozos to fill up the 3.0 quota to reward their current team. What do we do with the bozo's that have us going around making us flip the bozo-bit all the time? First rule: don't hire anymore bozos! I think we've already had a bozo explosion and now we're dealing with the aftermath from explosive growth and low standards. Anyway, we shouldn't want to do that again.
Hey, maybe it is too late.
As it says in Corporate Confidential on page 17, "Skills and Talent aren't what's most important."
After growing to a certain size, your company starts attracting MBAs and others that think this way.
It is certainly easier to speak the company speak and get promoted.
Someone so consumed with showing how smart they are may not spend enough time talking about how smart their managers are and how everything the company does is so great.
This may have resulted in bozos getting promoted. They're just better at being company cheerleaders.
Oh no! Now they're making hiring decisions - Bozopalooza!
You'll probably have to find a job at a smaller company if you are pining for the meritocracy of an engineering culture. The engineers do not run the company anymore.
Send in the MBAs!
Don't you love farce?
My fault I fear.
I thought that you'd want what I want.
Sorry, my dear.
But where are the clowns?
Quick, send in the clowns.
Don't bother, they're here.
Every acquisition is a failure on our part to innovate and another reason we have to ask, "And why do we have all these people around?"
There are only so many hours in a day.
Most employees are busy fixing bugs and coming out with service packs to support existing products.
What percentage of your time is spent on developing innovative new products?
How much of that time is spent in meetings? If those meetings involve management, you're working on that all important ingredient for a good review score - visibility. Are you going to stop going to meetings and make it look like they don't need you?
How many times do you have to start again because senior management has unrealistic expectations on how much time is required to ship the next release of a product?
How many times do you have to start again because senior management keeps changing dependencies and features in the product?
Not developing innovative new products? In most cases, that is by design.
Wow Mini. When are you going to get some new material?
You have so many readers! Make a difference instead of bitching. Do something worthwhile.
Talk about inovation. If we hire all of these people what can we do with them? We don't need any more people on Vista, so think of a product and get people motivated.
You have the rare opportunity to change Microsoft because of the sheer nubmer of readers you have.
The ability to start a skunk works project staffed by your readers would be amazing. All of the readers who want to change the world would gladly apply.
"With great power comes great responsibility!"
Use Mini! Change Microsoft instead of complaining.
I saw Kevin Johnson at the WinClient meeting, and I was underwhelmed. He came off sounding like the president (of the US) at a state of the union address. Lots of "we need to improve", "we can do better" banter without anything prescriptive. Furthermore, he said that he was on the job for 100 days, yet he failed to describe what he accomplished during those 100 days, other than to "go talk one on one with a bunch of people". We pay the guy on the order of what, $2000 a day and this is what he accomplished on his first 100 days?
Also, his letter didn't come off well, either. After he said that he wrote the letter, my first thought was, "oh, here's a guy that when things aren't going well, he writes a letter and takes off". So, will Microsoft be expecting a personal letter from Mr. Johnson a year from now?
The blog page "Incident at the WinClient All Hands" is gone. Wow, not everyone likes the attention, Mini. Below is what I have some my RSS reader (sorry, not nicely formatted) of the original post...
The incident at the WinClient All Hands
Or, An Initial Look at Propaganda.
propaganda
prop·a·gan·da
noun
Definitions:
1. publicity to promote something: information put out by an organization or government to promote a policy, idea, or cause
2. misleading publicity: deceptive or distorted information that is systematically spread
Many events lately added to the urge of putting up the shingle of my own blog, but none more so than the WinClient meeting last week. The previous such gathering, an ad-hoc meet convened at the behest of BrianV himself, shattered many a winter holiday plan accross main campus. It was proclaimed then that people had essentially a month to be code complete for Vista (with exceptions). Without rehashing that again, suffice to say it had a strong enough impact of the lives (and outlooks) of Windows workers that, when the new meeting announced Brian as a featured speaker, probably more than a few wondered what major reshuffle will he do for his next trick. HR was certainly worried, since BrianV's exuberance was toned down, he didn't mention again the empty parking lots and so on.
Not long later, during Q&A, a dude comes up and launches his interpretation of what Brian had actually meant to say. That we have to make August, or the company will lose 200+ mil. daily, and that we should skip the bloody Survivor and spend some more time in the office, fixing bugs. He qualified his remarks with the fact that's he's been an employee for 19 years. I respect that, but still his comments riled me up. We've been through the management screw-up that was called Alpha LH; we went through the restart, the fire drills of B1 last spring, the fire drills of code complete last December, will be in for a push in the early summer and still, some guy decides we need to be shamed into working later into the night, so that the shareholders don't lose too much.
I have a deep, serious problem with this. Many times I think of what I do, and I can't say I'm particularly impressed with my contribution to the society. I am gainfully employed by one of the world's (still) most profitable companies and I write software that allows the 10% of the population of the Globe who can afford a computer to have a better experience. In the process, I work (like many others) 50+ hrs per week, and I bring my work, fatigue, worries and stress with me at home. Nothing noble in there, as far as I can see, except for the work itself. Who and what grants someone the ability of making judgment calls about the efforts of an exhausted and morally depleted mass of workers? If we were saving lives - yes, by all means, call me out and shame me into giving my best, I know my effort might make a difference. But this? Please don't try this manner of rallying the ranks, it's more poisonous than it is efficient. Improperly used, propaganda is transparent to most people and the effect is a heightened awareness of their insignificant roles as cogs. Must row faster.
(Truth be told, the intensity is low. While not two months ago there was some urgency in the air (and few empty parking spots), lately you can see the people arriving at 10am and trickling out before 6pm. The top level of the parking garage tells the most compelling story: it is the last to fill up, and the first that gets emptied in the afternoon. )
Back to the meeting; finally, Kevin Johnson was introduced to Windows Client. He spoke at length, but I can't remember a single word of his speech - I'm quite sure it's a requirement of higher management. Towards the end, he read the now famous letter (Mini speaks about it here: http://minimsft.blogspot.com/2005/12/comment-repost-new-hope-part-ii.html), of a great, customer-dedicated gentleman who lived the shame of being the only accountable fellow in his group, during a meeting with a customer. He then folded the letter carefully, placed it in his pocket, (pause for emphasis) and announced that he was the gentleman in question, and the letter was his resignation from IBM.
If this hadn't been so scripted, if it didn't reek of self-promotion, of "watch this, Bucky, this is how I got here", it might have been a compelling statement. But in the end, coupled with the event I mentioned earlier, it felt as if the audience came in, allowed the management to try and instill some patriotic pride in them and went back, unfazed and unimpressed, thanks much for the drinks.
BrianV's message was loud and clear; from here on out, the focus is on quality. I believe him, and, had the meeting ended then, the audience would have been more energized, for that is a proper challenge for technical people. Hopefully he meant it, too, and not only as propaganda.
1. Netdocs was almost a good idea. Brian stepped on toes and pissed off the office franchise but he was on to something. Live hosted services on the web displayed in a rich and dynamic interface (AJAX?) Maybe it was just a bit before its time. Lets make Live really cool
2. Voice.net was sort of thinking about maybe doing something like Skype. Lets get a real team on the real time opportunity.
3. XBOX is cool. Lets double or tripple down while Sony is screwing around with cell chips.
4. Commerce is happening on the web. Really happening. Biztalk or something like it could be run as a hosted platform for B2B and B2C services. Think "AmaBay".
5. A complete P2P and RSS media publishing infrastructure. Client ships in Live. Lets make it run on other platforms... all of them..
Another approach to getting lean is to rip the big projects into their component parts. Apple is not perfect but they seem to be doing pretty well in this regard. iLife has a release cycle that is annual. Each of the components has a release cycle of its own.. Significant innovations go into these itterations. Each year there is enough planned for the big release that people will buy upgrades.
OSX is on a slightly longer release cycle but there again we see a number of very exciting enhancements coming between the major releases. A 5 year release cycle would never be approved by Jobs.
Windows Live could be a great playground for this kind of innovation if it does not get netdoc'd off the planet. Let people interview freely. Maybe allow 4.0 performers to take 20% of their time to build Live projects. Shake it up a bit. Let the Live team have access to source and drive user feedback and new features into a highly iterative "experimental" version of Office and Vista (like the unstable branch of an open source project).
Lets start writing software again !
The ability to start a skunk works project staffed by your readers would be amazing. All of the readers who want to change the world would gladly apply.
In other works, work some more for the company.
24 hours x 7 days = 168 hours/week
- 50 hours/week working
- 8 hours x 7 days sleeping = 56 hours/week
- 0.5 hours x 5 days commuting = 2.5 hours/week
- 0.25 hours x 3 x 7 days eating = 5.25 hours/week
- 1 hour x 7 days = 7 hours/week personal hygiene
= 47.25 hours/week "free time" (assuming absolutely no life outside of work, don't shop for food and somebody else cooks it, don't go to the doctor, etc.)
6.75 hours/day with absolutely "nothing" to do.
Given the same number of people that work on a project in their day jobs, it would only take 5.8% longer to develop a commercial software product. How long does that take at Microsoft?
Given what Corporate Confidential says about the importance of not saying anything negative about management or the company's policies (e.g. stack ranking), perhaps LisaB's listening tour is a way of smoking out all of the negative people in the company.
Microsoft may have its first layoffs in the near future to remove all the disruptive people saying negative things about management. It will take a while since they have to be real tricky about it to avoid lawsuits.
Then, they'll spread a rumor about why all those people really got laid off and that will shut everybody else up.
Problem solved.
How about this week's Microsoft Research Tech Fest show off all of the innovation we're quite capable of without having to go and buy companies?
Whenever you suggest technology transfer from Microsoft Research, you'll get some manager telling you their code is crap like they don't realize that their job is to write code and research's job is to come up with ideas. It is a very pervasive attitude among managers.
You have to think of research's implementation as a 'proof of concept'.
If the idea is good and the code is crap, can't Microsoft developers fix it or rewrite it?
Or, maybe those managers also think Microsoft Research's ideas are crap.
If the people in management don't change, do you really believe the amount of technology transfer is going to change?
The blog page "Incident at the WinClient All Hands" is gone. Wow, not everyone likes the attention, Mini. Below is what I have some my RSS reader (sorry, not nicely formatted) of the original post...
He used his real name on his blog.
He may have realized someone in management could be reading it which could adversely affect his career.
You can read his 'damage control' (in my opinion) post on his blog.
Big Brother Is Reading Your Blog
Corporations are investing in text-recognition software from vendors such as SAP (SAP) and IBM (IBM) to monitor blogs by employees and job candidates.
Corporate bloggers are also coping with increased vigilance by bosses. Getting fired for blog entries is so common now that it's come to be characterized by the term "dooced." Dooce.com, a blog kept by one of the dooced, has seen its traffic more than double over the past year, according to Web site ranker Alexa.
people getting lost inside the corporation once they are hired just isn't productive or worth all the money Microsoft and the shareholders have invested in hiring them and training them.
Delaying someone's ability to change jobs for an extended period of time when they really don't want to work for a particular manager or have lost interest in a project, creates unmotivated employees with ill will towards the company's policies where none may have existed before.
Management is creating the very people they don't want working at their company - people who don't agree with their policies.
After they create them, now they have all the work associated with finding them and getting rid of them.
If I was a manager and one of my reports expressed a desire to transfer, I would make it happen as soon as possible to avoid having someone around saying negative things about me behind my back. The rumors don't have to be true; just repeating them does damage because not everyone checks if something is true when they hear it.
Speaking of Mini, anybody notice lately that his/her blog has basically degenerated into a bitch-fest for MS employees? Mini throws out the raw meat for a few sentences or paragraphs, and the piling on in the comments commences immediately.
If you believe your compensation by the company is fair and you believe in the company's policies, why would you spend your time on a blog talking about it?
The only reason for talking about what you like about the company on a blog is having management read it and notice what a loyal employee you are.
On this blog, you end up with people who are unhappy with the company and who probably don't know how to dig themselves out of being marginalized and/or being managed out.
If you want the short version of Corporate Confidential, here it is:
1) Never say anything negative about the company to anyone who works at the company or who has friends who work there.
2) Never reveal anything about yourself that would cause a co-worker, a friend of a co-worker, HR or management to think you are a risk (e.g. illness, stress, gossip, etc.). If any of those people thinks you are a risk because of what you revealed about yourself it could result in you being managed out or being passed over for promotion.
A co-worker can use what you reveal about yourself to sabotage your career and advance their own. That includes co-workers you consider to be friends.
3) Stay away from anyone who is unhappy with the company.
4) Assume any communication can be passed on to anyone in the company (e.g. emails, instant messaging conversations, conversations, 'anonymous' surveys, etc.). Anything you say can and will be held against you if it is negative in any way.
The complaining people are doing on this blog is the very thing that is holding them back in their careers because they are likely doing the same thing at work.
Let management discover the company's problems and the solutions to those problems on their own unless you are asked.
If you are asked what could be improved, do it in the most positive way possible. The very thing you are complaining about could have been implemented by the person you are talking to.
You may still be marking yourself for being managed out no matter how positive you may think you were when discussing what could be improved.
It would be better to not say anything at all. Just do your job and go home.
How can you be critical, perhaps even constructive in your criticism, and not become the patron saint of the eternal complainer? Let alone a negativity magnet?
When moderating comments, you could insist that anyone posting a comment or a reply to a previous comment propose a solution to any problem they point out.
When proposing a solution, discuss why management would want to implement it (e.g. increased profit, productivity). The very filtered view of the company most executives get together with their compensation and the company's balance sheet makes it unlikely they will want to implement any of your suggestions so it better be compelling.
People shouldn't be burning through the corporation leaping from group to group every quarter. But you shouldn't be inhibiting them, either, worried about the worse case offenders.
A lot of the proposed solutions to problems are 'all or nothing'.
1) Fit the reviews scores to a curve for a few years and then don't for a few years after you believe you have cleared most of the deadwood. Call it a 'layoff' if you want to avoid the risk of litigation but quickly get rid of employees who have no hope of advancement. You aren't doing them any favors by letting them think they have a chance when they don't.
2) Every few years, let anyone change jobs a few times through the year without permission to allow them to find a good fit and to get away from managers and projects that are sapping their motivation.
No offense to Liddell, but could he be lower key? I'm sure he's a capable guy, but listening to him address that audience was like watching paint dry. Maybe that's the plan - by being super dull, he's makes even this stock look dynamic? :-) But it hardly seems like the right fit to get an obviously disinterested street excited. I do give him kudos for coming the closest of any MSFT exec to calling the one-time a failure - which is was and a massive one at that (though of course he didn't use those words). I also like how finally that disconnect between client growth and PC growth really is a problem after all. Ditto trends at Retail and Commercial. Gee, that only took two years of Liddell's predecessors denying it and giving excuses about "mix-shift", or commercial renewals being "about the same" to get to the actual truth. On the downside, he's still spreading this "money returned to shareholders" bs w/o acknowledging that shares+equivalents have still never gone down YOY despite $40B of buybacks since 00 (in other words, it all went to insiders not shareholders). True, since he's taken over, at least the buybacks have been accelerated. So maybe even this mgt team won't be able to dilute the stock fast enough for that to have no positive effect. Still, it's TBD. His answer on the overcapitalization question was pure fantasy/drivel. MSFT is hoarding far too much cash as the questioner correctly pointed out. It's simply ridiculous esp in the context of a company that doesn't even provide a market dividend despite horrendous stock performance. Finally and related, you still get the sense that this mgt team is either in complete denial about the stock having underperformed the market EVERY year since 02 (for a net underperformance of some 60%), or that understanding that, there is no perceived urgency in addressing it. They're the Enrst & Jullio Gallo of stock appreciation. We "will serve no wine until it's time". Translation for shareholders? "Later, dude". Ballmer/Gates need to go.
"So think about it - a Microsoft monopoly in the desktop, server, application development and consumer space, or a Google monopoly over life as we know it?"
Easy. I'll take the MS monopoly - I'm a shareholder.
["With great power comes great responsibility!"
Use Mini! Change Microsoft instead of complaining.]
Folks like you crack me up. Nothing prevents you from forming your own blog and populating it 7/24 with Ballmer worship, general "atta boys" and eternal sunshine. Instead, you criticize Mini for his efforts despite him likely having done more than ANYONE to shine a spotlight on what's not working at MSFT and needs to change. And along the way, lo and behold, there are even some bright, positive ideas expressed by him and others - you just have to look. As a shareholder, I'm happy Mini exists. MSFT has no shortage of smart, positive people capable of making a difference (Mini likely one of them). They also have no shortage of ridiculously highly paid senior executives whose job it is to supposedly create an environment where those folks can realize that potential. Unfortunately, in too many cases, that's not happening and it's adversely impacting both the top and bottom line. It would be nice if senior most management - the folks officially tasked with the responsibility - were dealing with this issue head-on and aggressively. Instead, we get deafening silence and/or lip service. That is until Mini started shining a spotlight on it and developed a following. Now, albeit still too slowly, we at least see some positive action from senior management. In my mind, you should applaud his efforts - not criticize them. If he didn't care and want to make things better, he'd wouldn't have started this blog and committed the personal cycles to it that he has - he'd just rest and vest (like too many others) or quietly leave.
Mini, what's up with hating on agile/SCRUM?
Mini, what's up with hating on agile/SCRUM?
Ooo, I'm not being an agile/scrum hater here. No no.
I'm saying as soon as these so called Bozos find out that agile/scrum is the in thing to save their Bozo-bacon they'll start touting their esquisite scrum-deli-icious abilities.
Let's hope that scrum is self-correcting to those that would abuse it.
Mini.
The complaining people are doing on this blog is the very thing that is holding them back in their careers because they are likely doing the same thing at work.
Let management discover the company's problems and the solutions to those problems on their own unless you are asked.
If you are asked what could be improved, do it in the most positive way possible. The very thing you are complaining about could have been implemented by the person you are talking to.
You may still be marking yourself for being managed out no matter how positive you may think you were when discussing what could be improved.
It would be better to not say anything at all. Just do your job and go home.
SO this isn't a free country?
I had an idea tonight that I'd like to bounce off this august company.
As Microsoft matured, our businesses grew to the point where an increasingly small percentage of people were doing the work that really (at least in the minds of developers) deserves the big payoffs in stock, bonus, and raise. We now have a ton of maintenance projects keeping customers happy and our products serviced.
I wonder if one way to handle the so-called bozo issue is to create several different tracks of employees. For example, the hot devs working on the leading edge features of our cashcow products (or entries into new markets) would remain on the current competitive, curve-based system. On the other hand, other employees could choose to work on less sexy but necessary efforts such as sustained engineering. These folks might not share in the big rewards on the big bets. They might be perfectly competent to fix bugs, or create patches, and perfectly happy to do it in an environment less pressured than the trenches of product development. They might not even be eligible for stock, or they might be permitted to stack up years of 3.0s without ever risking their job. We definitely need people to do this work, and people will do it better if we create an atmosphere in which they can feel perfectly happy doing it.
I remember hearing when I was applying for college that Harvard and other colleges of that ilk liked to accept a certain number of C students, people who might be happy being inferior to the crop of coffee achievers all wrestling to be valedictorian.
As to anyone who might suggest this would lead to a caste system at Microsoft, I would say it only refines the existing caste system and admits that it is there. We know that devs rule; we know that PSS gets posterior lactation. QA and PM can rank themselves as they see fit, and don't even get me started on recruiting or HR.
Certainly, these castes don't need to be rigid. Motivated folks might prove themselves. Conversely, some former superstars might grow tired of the intensity and take a role that has less pressure, and fewer potential for rewards.
We are now a big company, and we don't need to pretend any longer that all we hire are superstars. We need to take a long look at ourselves and make sure that we find ways for the superstars to create maximum value, and the lesser lights to perform the valuable work they can still contribute.
Of course, I have no problem with the incompetent and the truly indolent and stupid being fired. (You make your own snide remark about executives.)
Accountability is nice on paper, fiendishly difficult to achieve in practice. If there is one lesson to be learned from observing bozos, it is that any management metric can be gamed, and the bozos are generally masters at the art of gaming because that's all they do, unlike those distracted by making products.
I'm afraid once the rot sets in, it is almost impossible to expunge.
>>It would be better to not say anything at all. Just do your job and go home.
>SO this isn't a free country?
What crack are you smoking, commenter? Free Speech is only guaranteed by the government (and even then there are restrictions, like joking about bombs at the airport). When you swipe your badge and enter the door, you don't have the right to free speech anymore.
That's the kind of bizarre thinking that I see when people gripe about MS IT shutting off the ports of "my computer." It's not -your computer-, it's Microsoft's computer.
>On the downside, he's still spreading this "money returned to shareholders" bs w/o acknowledging that shares+equivalents have still never gone down YOY despite $40B of buybacks since 00 (in other words, it all went to insiders not shareholders).
It sounds like you're saying that buybacks are a bad idea. Bill and Steve actually agree. There was a Town Hall meeting a long time ago where some blue badge got up and argued with Bill saying that buybacks would drive up the stock price, and Bill patiently explained why that wouldn't work. It went back and forth and got beyond my comprehension. The blue badge just got louder and louder, until I think it was Steve who cut in and said "We'll have to agree to disagree." Very surprising.
SO this isn't a free country?
Not during working hours.
During that time, it is more like the Cold War era Soviet Union.
If you read that Corporate Confidential book, you would know that employers can easily get around most employment law.
Your co-workers will inform on you to get ahead.
If you complain, your career can easily be over.
You have to be considered loyal to be safe.
Wow Mini. When are you going to get some new material?
You have so many readers! Make a difference instead of bitching. Do something worthwhile.
[..]
The ability to start a skunk works project staffed by your readers would be amazing. All of the readers who want to change the world would gladly apply.
I think you're missing a large part of the audience here. You're missing people that love what they do, don't want to switch groups but are helplessly witnessing the bureaucratization of what used to be a dynamic process - writing software. People that see some half-assed designed passed as the next great platform, make a career out of it and grow into "leadership" roles, leaving to others the fixing of horrendous, fundamental problems. People that have been reorged 3 times in less than 5 years and witnessed the leaders at various levels describe how the new scheme will empower us to greater efficiency.
I don't want to go elsewhere; I'd rather stay put and do what I do, as there are very few other places in the world where I'd have the same impact, working on the same technology. Elsewhere I'd use it, here I help making it. I just want to be able to respect whomever currently hovers 3 levels above me.
Another category you're likely missing are the very capable, but not very ambitious guys. (Big paranthesis here: I believe there are three basic qualities that are important in having success at MS:
- ambitiousness
- conscientiousness
- intelligence.
In my mind, any other traits are either combinations of the primitives above or are not relevant (for instance, honesty or fairness). Any of those 3 will do nicely; having just 1 is (in my mind) what separates the haves (4.0+) from have nots (3 and 3.5). Of course, one can always achieve much more through just ambition that through just intelligence. End of big paranthesis.)
So your noble call to action is not going to touch the intelligent contributors who just don't have the aggressiveness of those more ambitious around them. They don't work well when not challenged at their potential (say, when they're fixing bugs for months on end (devs), or when they're writing API tests (testers) or ordering meals and spending hours in conference calls (PMs)). They don't want to leave, either. They simply want a challenge better suited to their abilities, and perhaps a bit more fairness/recognition of their contributions.
Those that excel at accomplishing tasks (the conscientious category) are also being punished by the current system. They keep getting 3.0s in a row until they're being let go. You can't really argue that the company doesn't need this type of worker.(I saw a comment above that deals with this further - implementing a caste system).
All these people come here to find a place to vent, a story similar to theirs and the reassuring feeling that comes from anonymous solidarity. Many might even hope their concerns will make a difference (for the better, and not for the unemployment, that is). Oh, and some come for the juicy partner posts - we all want to see how the other half lives, don't we..
This should also address Mini's rhetorical question: How can you be critical, perhaps even constructive in your criticism, and not become the patron saint of the eternal complainer? Let alone a negativity magnet?
Well, there's not much you can do, can you? You have your goal, it's clearly stated, the comments provide examples and reasoning, pro and contra. Just know there are many more that silently read and think than there are useless flame-baiters and negativists. There is noise everywhere.
He used his real name on his blog.
He may have realized someone in management could be reading it which could adversely affect his career.
You can read his 'damage control' (in my opinion) post on his blog.
Mmm no. That is not my real name, obviously, and, in my opinion, the post was written not to control damage, but to explain why the previous posts were gone. I felt I owed Mini that courtesy.
(For those interested in this soap opera, some commenters suggested that I used Mini's blog to promote my own. While I believe their indignation paled in contrast with my own, it was obvious to me that the only proof to the contrary was to obliterate my blog altogether. No blog == no motive, right? It just so happened that while I was deleting the posts, Mini was kindly referring to one in his most recent post. Not 30 minutes separated the two publishings. Later on, I reloaded MiniMSFT, saw the new post etc etc. (end of soap))
Needless to say, you haven't missed much. (Thanks to whomever reposted it here.)
I wonder if one way to handle the so-called bozo issue is to create several different tracks of employees. For example, the hot devs working on the leading edge features of our cashcow products (or entries into new markets) would remain on the current competitive, curve-based system. On the other hand, other employees could choose to work on less sexy but necessary efforts such as sustained engineering. These folks might not share in the big rewards on the big bets. They might be perfectly competent to fix bugs, or create patches, and perfectly happy to do it in an environment less pressured than the trenches of product development.
I see your point, and it makes sense as it introduces another evaluation criterion: ambitiousness.
I also see two reasons for which this might not work:
- often times the hot devs are those who have a clearer, deeper view of a component/product/technology. Their help is much needed on hot fixes, SPs and any other legacy code, one-off, important issues. So they have impact everywhere, and they're still competing, only this time against a much tougher competition - against their ilk only. They used to be automatic 4.xs until now, and their status quo is gone. Where is the incentive?
- the other category (the conscientious class from my earlier post) has now a shred of hope that, once in a few years, the time comes for their 4/promotion. Doing their job dilligently is how they compete. If you're granting them forever 3. (in other words shelter and food) but no more, their incentive disappears as well, and their efficiency (or worse, the quality of their work) drops.
(Oh wait, there's another one, what do you do about the former performers, who are now bearers of distinguished titles and do not care much about the dogfight down the hallway? If they would be content with straight 3s and their current pay grade, does it make good business sense to maintain their statu quo?)
We know that devs rule; we know that PSS gets posterior lactation. QA and PM can rank themselves as they see fit, and don't even get me started on recruiting or HR.
--
Not right. In terms of max money for least IQ HR rules. PM comes next. Dress well and talk well promotions will follow. Dev and QA are up next. Recruiting and PSS -wish they didnt exist.
"It would be better to not say anything at all. Just do your job and go home."
With the attitude you expressed in this and the preceeding comments, you sir are a born bureaucrat. Suspect you're on the fast track to VP at this very moment - god help us.
"...QA and PM can rank ..."
To be clear, Microsoft has plenty of Testers, but little, or no QA. There's a big difference between the two, and the lack of "real" QA is part of the root of Microsoft's quality issues.
And to keep this somewhat on topic, I think the commenters who have suggested ripping the big projects into their component parts are right on track. Little pieces released more often will not only be better for the customer, but will end up making a lot of the useless middle management jobs that mini (and so many of us) hate obsolete.
Has anyone at Microsoft EVER heard the term "underpromise and overdeliver"? I ask because in light of that and especially given the track record of MS Search in particular, I find the following asinine:
Microsoft says better than Google soon
"I believe there are three basic qualities that are important in having success at MS:
- ambitiousness
- conscientiousness
- intelligence."
I was there 5 years with a lifetime 4.0. Your list might be accurate if you define success as moving ahead. Otherwise, I saw numerous examples of folks who demonstrated little ambition or conscientiousness and frankly weren't overly intelligent that did just fine. What they excelled at - and a major item missing from your list for those who want to get ahead - is politics or at least flying under the radar. MS imo was as much about being perceived to do a good job as actually doing a good job. The latter wouldn't normally get overlooked, but the former was often sufficient even in the complete absence of actual results. Unfortunately, the collective impact of that sort of system is starting to show up in the company's overall performance or, more specifically, lack thereof.
"It would be better to not say anything at all. Just do your job and go home."
With the attitude you expressed in this and the preceeding comments, you sir are a born bureaucrat. Suspect you're on the fast track to VP at this very moment - god help us.
You are surrounded by bureaucrats.
If you don't do it their way, once they find you, they will easily circumvent every employment law designed to protect you and quietly shuffle you out the door.
Take some good advice about not touching the 'hot stove' or get burned and learn the hard way.
MSN Search News
So why is the president of EMEA making statements to the press about our MSN search strategy and execution. Are we just completely lost at this point where we've relegated comments on our product strategy to field sales management?
Sometimes I hate us...
Oh wait, there's another one, what do you do about the former performers, who are now bearers of distinguished titles and do not care much about the dogfight down the hallway?
Nothing. They'll cash out and end up going to some other company to cash in again.
Well, there's not much you can do, can you? You have your goal, it's clearly stated, the comments provide examples and reasoning, pro and contra. Just know there are many more that silently read and think than there are useless flame-baiters and negativists. There is noise everywhere.
You have a real mixed vibe.
On one hand you claim you want change but, on the other hand, you dismiss others that do not share your particular flavor of complaining as negative.
If they would be content with straight 3s and their current pay grade, does it make good business sense to maintain their statu quo?
Microsoft does not want people who are content with what they have.
They want employees who will, through hard work and knowing how things are done, eventually get promoted while the lower levels are replenished by new graduates who are more cost effective.
Anybody who doesn't agree with this eventually has their salary decreased and their stapler taken away.
If your name is Bill and you're trying to rule the world of technology, you can't have a company full of people satisfied with what they have.
It would be nice if senior most management - the folks officially tasked with the responsibility - were dealing with this issue head-on and aggressively. Instead, we get deafening silence and/or lip service.
You can get promoted to make those changes yourself.
However, at that point, through what you had to think and do to get promoted to executive, you will be indoctrinated to think the same way as those currently in charge.
You will think of those below you as not seeing the big picture and ignore their negative noise.
I'm saying as soon as these so called Bozos find out that agile/scrum is the in thing to save their Bozo-bacon they'll start touting their esquisite scrum-deli-icious abilities.
Speaking of bacon (perhaps Sir Francis Bacon), Hamlet seems to be the management style of Microsoft managers.
To take a dependency or not take a dependency?
To cut or not to cut?
To fix or not to fix?
To ship or not to ship?
To promote or not to promote? Rosencrantz and Guildenstern repeat 3.0's for you!
>>It would be better to not say anything at all. Just do your job and go home.
>SO this isn't a free country?
What crack are you smoking, commenter? Free Speech is only guaranteed by the government (and even then there are restrictions, like joking about bombs at the airport). When you swipe your badge and enter the door, you don't have the right to free speech anymore.
This is the common misconception about free speech. In fact, you do have the right to free speech in this country, and within the hallowed halls of Microsoft. As the constitution guarantees, you do have the right to say or publish whatever you wish. Nobody can legally prevent you from doing this.
You do not, however, have the right to be free of accountability for the effect of your speaking out. The constitution does NOT guarantee you that everyone will still like you, hire you, admire you after you have exercised your guaranteed rights. You only have the right to say. This is so often confused by, for example, the wackos in Hollywood that insist that any repercussions in their popularity or public image because of their extreme left positions is an assault on their right to free speech. It is not, they were able to speak their piece, just as I am free to react to what they say.
At Microsoft you remain accountable for what you say and to whom you say it. You retain the right to say what you want and to whom you want, and 'they' retain the right to take whatever action 'they' deem necessary to protect themselves, once your position is known.
The HR advice from the oft-quoted book is sound. Use your right to free speech carefully. You and only you are fully accountable for the damage it may cause to your career.
While there are some useless VPs (such as the ones without direct reports) there are some that remind me of what Microsoft used to be in the old days - specifically Amitabh Srivastava. Without his research into developing applications to automatically catch software bugs, we might be knee-deep in testing hell today.
Which makes me wonder - Do any of you know people from MS Research that are as discontented with the way things are going as the rest of us devs?
>>It would be better to not say anything at all. Just do your job and go home.
>SO this isn't a free country?
What crack are you smoking, commenter? Free Speech is only guaranteed by the government (and even then there are restrictions, like joking about bombs at the airport). When you swipe your badge and enter the door, you don't have the right to free speech anymore.
That's the kind of bizarre thinking that I see when people gripe about MS IT shutting off the ports of "my computer." It's not -your computer-, it's Microsoft's computer.
Like I said, it's not REALLY a free country.
The ability to start a skunk works project staffed by your readers would be amazing. All of the readers who want to change the world would gladly apply.
To the anonymous employee listing all the work hrs. You are forgetting 1 group that recieves no paycheck for their "volunteer" work hrs. These are the people who test your software before release to the customer and provide feedback to help improve in a test real world environment.
Jeez, get a real life or get a real job!
Customer
Did anybody attended the techfest today? Things look hopeful to me (personally). What everybody else thought about it. Do you think this set of geeks, eating fat salary, called researchers could inject some innovation in our blood stream.
[..]there are some that remind me of what Microsoft used to be in the old days - specifically Amitabh Srivastava. Without his research into developing applications to automatically catch software bugs, we might be knee-deep in testing hell today.
You have got to be joking. What we're knee deep in today is:
- pref* bugs - thousands of them, and they spring from the ground like daisies!
- policheck
- the utter crap monstrosity that is WTT and its awful logic and logging
Our architect is fixing PREF* bugs because the rest of the devs are either allotted to side pet projects of the higher management or picking up the slack of the shell team. Test devs fix dev bugs and I half expect to see PMs starting fixing test bugs any given day.
I've heard recently of an fxcopp-raised bug, which complained that a function name was spelled incorrectly. This kind of "automated bug catching tool" is a mandated check-in criterion. The top mgmt is having a silent crisis about releasing Vista on time (won't say more here), and we have to fix freaking spelling mistakes, go through stupid automated check-in systems, all of which cost time, add process, add more people that oversee the process etc etc. For what, a 1 in 100 legitimate bug?
We need to step back, see the shades of gray and stop thinking in binary. There is a time for trying various high-potential tools, but the crunch-time of the most important product is not it. I wish someone would estimate the time and energy lost to disruptions caused by crappy analysis tools and half-baked frameworks.
Yeah, next time I'll tell you how I really feel.
Amitabh Srivastava. Without his research into developing applications to automatically catch software bugs, we might be knee-deep in testing hell today.
The truth is, we're in bureaucratic hell thanks to process loving idiots like him.
Do you think this set of geeks, eating fat salary, called researchers could inject some innovation in our blood stream.
Go to http://research.microsoft.com and find out! They have some pretty interesting new products emanating from MS Research. If only there was a good process in place to coordinate the researchers' efforts with the product devs.
You have a real mixed vibe.
On one hand you claim you want change but, on the other hand, you dismiss others that do not share your particular flavor of complaining as negative.
Sorry, I didn't mean to say that which you inferred. I meant to say that, in addition to all sorts of rebuttals, examples, arguments, complaints of any nature, there is always some noise. You know, trolls ("Suckers!"), one-liners ("You suck") or unsubstantiated negative remarks ("This sucks").
Did anybody attended the techfest today? Things look hopeful to me (personally). What everybody else thought about it. Do you think this set of geeks, eating fat salary, called researchers could inject some innovation in our blood stream.
Ask the researchers about the amount of technology transfer from their group to the product group.
If you dare, if you see some technology that could be used in your group, ask your manager what they think about using it. If your manager dismisses it as not usable, you probably just identified yourself as a 'trouble maker' and discovered why 'innovation' sits on the shelf at Microsoft.
I am leaving Miocrosoft next week after more then 11 years working as a Software Engineer and a Development Lead.
I found an outside job from the first try that pays 80% more then I used to earn at Microsoft. I would have gone, even if it paid 20% less. The management drove away the last thinking people with mindless pay cuts and promoting meritocracy. I found really no reason to stay. Projects are poorly managed. Nobody thinks long term. The whole company has turned into a Dilbertesque circus.
You are forgetting 1 group that recieves no paycheck for their "volunteer" work hrs. These are the people who test your software before release to the customer and provide feedback to help improve in a test real world environment.
Jeez, get a real life or get a real job!
Well, that increases the estimate of the time to complete a 'skunk works' project doesn't it? You'll have to spend a lot more time finding volunteers inside the company to make up the difference.
A lot of those hours are spent during your day job if you have any business reason to install the software.
BrianV, years ago, brought some of you into our team meetings to show us who was going to lose their jobs if we didn't ship soon because they recommended our software over something already shipping and their company spent a lot of time and money testing our software.
If you want to keep your 'real job', next time recommend something that has already shipped and volunteer your time elsewhere (e.g. Habitat for Humanity).
If you are actually testing Microsoft's software on your own time, again, you can probably find something more worthwhile to volunteer for.
Like I said, it's not REALLY a free country.
The closest you're going to get to a free country is a small community where strangers looking to cause trouble end up as barbeque served in a tasty sauce - like the community in the movie Fried Green Tomatoes.
The HR advice from the oft-quoted book is sound. Use your right to free speech carefully. You and only you are fully accountable for the damage it may cause to your career.
If you're young and are bursting with good ideas, you would be better off using them at a small company instead of just telling them to some manager at Microsoft and being shot down because they know better than you.
At a small company, if they really are good ideas, you would have a better chance of benefiting from those ideas instead of some manager taking credit for them.
While there are some useless VPs (such as the ones without direct reports) there are some that remind me of what Microsoft used to be in the old days - specifically Amitabh Srivastava. Without his research into developing applications to automatically catch software bugs, we might be knee-deep in testing hell today.
Which makes me wonder - Do any of you know people from MS Research that are as discontented with the way things are going as the rest of us devs?
There was someone from Microsoft Research that posted here a few months ago about the level of frustration there and some of the reasons why. If Google fixed 'Blog Search', I could find it.
And to keep this somewhat on topic, I think the commenters who have suggested ripping the big projects into their component parts are right on track. Little pieces released more often will not only be better for the customer, but will end up making a lot of the useless middle management jobs that mini (and so many of us) hate obsolete.
Amitabh Srivastava sounds like someone who could make this happen. You would need an executive like him to push for this since the middle management you complain about surely will not make it happen.
The tools used by the architectural layering team could be used.
Send him an email if you think it is a good idea.
So, I'd love to see KJ's letter. As a former services employee - I can say this. The first I heard of KJ was through a "chain e-mail" that talked about him reading this book:
The Tipping Point: How Little Things Can Make a Big Difference by Malcolm Gladwell.
This is one thing that frustrates myself more than anything else - MSFTies, concerned with the company's direction, post their frustrations. But what are they doing about the broken windows? (quote from the book). Sure there is a system - you can work with the system, outside the system, doesn't matter, just fix that broken window. Critical mass will be reached!
"Lets start writing software again !"
Some very good ideas in this post (just typed the last line to save space). Maybe not all the right ones but at least the right mindset. Kudos to you.
"It sounds like you're saying that buybacks are a bad idea. Bill and Steve actually agree."
No, buybacks would be reasonable if they were doing what buybacks should, namely reducing shares outstanding. Unfortunately, MS has historically been diluting the stock faster than they're buying it back. So while some $40B has been spent on buybacks since 00, shares+equivalents has NEVER gone down year over year. Unless you're reducing the share count, there's no positive effect - just the avoidance of a negative (further dilution). To then turn around and call that "money returned to shareholders" imo is an outright lie and totally self-serving.
Amitabh as our savior? Oh please.
I'm sure he's a fine guy and a smart researcher. Some of the tools his group has created are really usefull. But the soul-crushing bureacracy he's created is going to kill WinDiv.
He's created an environment where adding a SAL annotation (which has zero change is code operation) is as important as fixing a crash and way, way, way more important than fixing broken behavior impacting millions of customers. Forget about long-term engineering improvements like refactoring or re-architecting. Everything has to get done by the 15th of the next month.
The top mgmt is having a silent crisis about releasing Vista on time (won't say more here), and we have to fix freaking spelling mistakes, go through stupid automated check-in systems, all of which cost time, add process, add more people that oversee the process etc etc. For what, a 1 in 100 legitimate bug?
BrianV's mantra is Don't trust developers. Long ago, he used to be a tester which is where he got that perspective.
Given the trouble they've had getting a grip on getting Vista out the door, is it really that surprising?
Spelling mistakes don't look good if you ever have to license the source code to a competitor or, if you have to show the source code to a communist government with trust issues or some other government concerned about security.
For ten millon customers, a 1 in a 100 bug works out to 100000 customers. If they're at a large corporation, you'll be in on the weekend figuring out what is causing the bug.
Headcount cuts. This is very tempting.
Everything spouting from management has a hidden message. This one just translates into 2-3 new hires in India/China for every headcount cut in Redmond.
Our architect is fixing PREF* bugs because the rest of the devs are either allotted to side pet projects of the higher management or picking up the slack of the shell team. Test devs fix dev bugs and I half expect to see PMs starting fixing test bugs any given day.
--
Let me guess your org. You are in the shell team. Isnt that the bozo central group?
Architect/PMs/test fixing bugs - isnt that a good thing?
"If you are actually testing Microsoft's software on your own time, again, you can probably find something more worthwhile to volunteer for."
Maybe I did not make myself clear in my response;
We are an actual customer that has used your software since we opened our store 12+ years ago. During that time we not only learned how to run our business succesfully, we also learned how to use your software to manage our business for a decent profit to ourselves and family.
Sometime ago a new software offering by your company came along that was "right up our alley" so to speak. This software was offered to us in beta to "test" in a real world environment along with your competitors version. We spent several months of double entry of data for comparison purposes for us as well to provide feedback to the Team we work with as per our NDA agreement.
The time we spent testing your software has paid us back in actual $ value to our company. I see this as time well spent continuing to make MY software more valuable to our company.
Habitat for Humanity is a very worthwhile volunteer organization. With our increased profits from your software we were able to give more money this year to our local charities, school districts and organizations.
Don't know BrianV but do know some of the Team that I work with and they all (to date)are people that I would love to have working for me.
To add, our business has nothing to do with yours technically speaking. We just use your product(s).
Customer
"- the utter crap monstrosity that is WTT and its awful logic and logging"
Does *ANYONE* think that WTT is anything other than complete crap?
I know devs who were asked to give a few hours to help maintain it and ran away screaming.
I remember one time someone showed me a performance problem in one of the SQL queries. The damn query took up half of a printed page and was composed of a couple dozen joins.
WTF???
Worse yet, it gives a very false impression of quality. It gives pretty numbers for upper management that don't map 1:1 to quality.
It is yet another example of something that would have died a quick death if it wasn't forced on testers by management, and it is just another reason why I left MS last year.
Amitabh as our savior? Oh please.
I'm sure he's a fine guy and a smart researcher. Some of the tools his group has created are really usefull. But the soul-crushing bureacracy he's created is going to kill WinDiv.
He has done more for MS than any other Corporate VP ever has. He is also the only Corporate VP who is also a DE.
Of all the bureaucrats listed on the exec press pages, Amitabh is one of the few VPs that deserves his bonuses!
We know that devs rule; we know that PSS gets posterior lactation. QA and PM can rank themselves as they see fit, and don't even get me started on recruiting or HR.
Ooh, don't forget about the marketing weasels! For expertly obscuring information that might actually be useful to the consumer and spouting half-truths and trite marketing-drivel that the most naive would have trouble swallowing, they can't be beat.
Ooh, don't forget about the marketing weasels!
I totally agree with you! Sometimes I wonder what the marketing PMs even do! Seriously, what's the purpose of having a "Group" PM at all!
Let me guess your org. You are in the shell team. Isnt that the bozo central group?
Mmmm no, I'm not in shell. I don't know whether they are bozo central or not, although I do know of a few people that might take offense at your remark. They have a reputation for being snotty, but I wouldn't kick them out of my group (if I had a say in it, that is).
Architect/PMs/test fixing bugs - isnt that a good thing?
Yes, but you probably skipped over the rest of the image. Nobody is doing their job - what they're best at - everyone is doing someone else's job. Clearly, this is sub-optimal.
And if it just couldn't be helped, that would be fine. But in this group, some (of the better) ICs (dev and test) work on a project parallel with Vista, which helped creating this shift of workforce. Hopefully both projects will come out fine.
He has done more for MS than any other Corporate VP ever has. He is also the only Corporate VP who is also a DE.
Thanks for the laugh! So corp VP (VP) and technical fellow aka distinguished engineer (DE) mean the same thing: level 80+. DE was created to allow senior ICs a way to break past level 70 and have an official HR track and titles. We then went back and bastardized the system by making some new 80s DEs and not VPs (Peter Spiro, etc. Check out the 'engineer' alias).
So your comment is like saying 'not only is Sally a PM but she is also a Principal PM'. She is still just a level 65 PM.
And all of Amitabh's process improvements have resulted in a product that still isn't out, it has more DVDs than the first season of the O.C., takes a miracle (and a day) to build, and has no perf. The tools didn't see that coming?! And he pisses off every dev in the world. How many layers of management did he introduce in places like build?
He is the best we have? Dog save us all...
sigh!
The company grew to a size where engineers no longer run the company.
MBAs and other business types value loyalty more than engineering talent and as a result promote based upon those criteria.
Engineers call 'crap' what it is when they see it which sounds so negative. Let's get rid of all the negative people! Oops!
People who are better at politics rise faster at Microsoft than those who aren't. Those better at politics are not necessarily the best people to be making engineering decisions.
Over time, the company has now reached a point where the people who are more talented at politics than engineering have saturated upper management and other senior positions.
As a result, more and more people with real engineering talent are being shut out and leaving.
HR may be able to side step employment law at will but people who are expert at politics and high enough in management are staying put and, as a result, so are the problems. They deserve each other.
Karma on a slow burn.
>Yes, but you probably skipped over the rest of the image. Nobody is doing their job - what they're best at - everyone is doing someone else's job.
Surely some people in the group arent doing their job. If you have to help shell, shell group management isnt doing its job.
If your group is checking in half baked features with low quality, the group pays the price. Your GPM, PUM and GM etc. should be held accountable for this.
For ten millon customers, a 1 in a 100 bug works out to 100000 customers. If they're at a large corporation, you'll be in on the weekend figuring out what is causing the bug.
Your logic is wrong. How many customers the 1 in a 100 bug works out to depends on how many of the customers use the feature that exposes the bug. It does not depend on the efficiency of bug-detecting tools.
Your logic is wrong. How many customers the 1 in a 100 bug works out to depends on how many of the customers use the feature that exposes the bug. It does not depend on the efficiency of bug-detecting tools.
Are you saying that Microsoft is wasting time on features that very few people use?
So, because very few people use the features you are talking about, it is not worth fixing a 1 in 100 bug.
Isn't the only way a feature that very few people use is going to get into the product is if a very important client requests it? If so, shouldn't you fix the bugs associated with feature containing the 1 in 100 bug?
Sure, every bug in the product should be fixed, but it's a trade-off. Maybe automated tools find a legitimate bug once in a while, but is it worth the hundreds (if not thousands) of man hours spent per group per cycle making the tools happy? This time could be spent finding and fixing more significant bugs. It's impossible to say for sure but my gut feeling is that the tools are decreasing our net stability. (And lowering morale.)
So why is the president of EMEA making statements to the press about our MSN search strategy and execution. Are we just completely lost at this point where we've relegated comments on our product strategy to field sales management?
Microsoft says better than Google soon
You mean Microsoft has finally come up against a product it can't kill by bundling it with Windows?
But bundling would find little purchase against Google because it lives insulated from Microsoft on the Web, unlike other applications that were easy game for the software giant as they perched directly on Windows.
Hey wait, they found a way to bundle it at least partially. Integrate Microsoft's new search engine into Windows Messenger which is bundled with Windows. Tricky! I'm sure the EU won't notice. Can you uninstall Windows Messenger?
Microsoft will put its search engine into its widely used communications tools Windows Messenger and Hotmail.
"Integrating search into those other applications ... makes it very seamless for people," he said. Timing in Europe will be pegged to that in the United States.
"The UK will probably be at the same time, France maybe three months behind, Germany maybe three months behind. It's not two years behind."
Are you saying that Microsoft is wasting time on features that very few people use?
So, because very few people use the features you are talking about, it is not worth fixing a 1 in 100 bug.
Nonono, I was simply not (crystal) clear on my initial point (that is why context matters). What I meant to say was that a code parsing tool spews out 100 different warnings, all of which need to be investigated, and of which only one is valid (ie a bug). If you have a sizeable component and have real work to do/bugs to fix, pref* bugs are an absolute nuissance: they come in bursts, pile up quickly and all are pri 1, sev 1. Your entire group is now in bug jail, halting real work and fixing pref* for 1-2 weeks. By the time the next round comes, the pref* tools have been "improved" to find more obscure warnings, and so the pref* bugs don't trend down as quickly as they should.
"but people who are expert at politics and high enough in management are staying put and, as a result, so are the problems."
Of course they're staying put. Have you seen how much these clowns are getting paid? The sad fact is that senior mgt can do well regardless of whether MSFT, general employees, customers/partners/shareholders do well. They've been proving that for 5 years now and NOTHING suggests that a change is coming any time soon.
Regarding prefast tools- I think they they are a good investment and do help in finding some real bugs - especially ones that could to security issues- So we do need them. However I do accpet that we have to fine tune them and remove certain plugins that produce a lot of noise and find very few real bugs...
Finally if your team *always* gets a lot of prefast bugs on every run, then it means that the devs are not writing good code.
Of course they're staying put. Have you seen how much these clowns are getting paid? The sad fact is that senior mgt can do well regardless of whether MSFT, general employees, customers/partners/shareholders do well. They've been proving that for 5 years now and NOTHING suggests that a change is coming any time soon.
That's right. The higher the level the more compensation you have as an incentive to stay (presumably because you're more valuable to the organization). The attrition rate is lower so these very political people accumulate at the top of the organizational hierarchy (which has been happening for many years now). The more of these people you have around the more adept at politics you have to be.
HR doesn't even know how to sort everyone out at the IC level when there is a dispute so they just take management's side every time.
HR is totally lost when it comes to correctly problems with senior management.
If you do what the bozos say and let them accumulate a string of failures, it will be more likely that someone will notice.
If you feel that you should let them know when they do something wrong, go ahead. Chances are they will not listen to you. You've done your duty and possibly pissed them off enough to think of you as someone to get rid of (especially if you do it repeatedly).
Propping them up only prolongs the agony of having to work for them.
Can somebody post the top/bottom comp ratio i.e. what CEO earns/entry level earns and similarly for GMs.
In ideal company top/bottom ratio should never exceed 40.
My Gawd. As a Dev I am really ashamed at the putrid junk being spewed on this blog by Devs. Yes 85% of the whining here come from Devs. I am a Dev. I know my ilk when I see them.
Consider these facts:
1) Devs have an over-inflated sense of self worth. We write the code so we must be the smartest. WRONG.
2) Devs have a superiority complex that is actually an inferiority complex. PMs are worthless. Testers are useless. Marketing is stupid. HR should be shot. WRONG
3) Devs are very jealous. How come the PMs get to do customer visits or the conferences? Well maybe if you brush up your social skills you may get a shot
4) Still on the jealousy tip, have ever seen Devs resisting designs from a deeply technical PM just because the PM in better versed in the new area than the Devs? WRONG
5) Like some directors badly want to be actors, some Devs badly want to be PMs just because the "Manager" in PM sounds cool. Thus they do every thing to sabotage their component PM. WRONG
6) Devs can't understand why somebody else is being paid the big bucks. Yet they want to be paid big bucks. WRONG.
7) Devs show their immaturity by complaining about Process, forgetting that process is critical to achieving software engineering perfection. They complain about every thing but the truth is that Process exposes their flaws for all to see, like those Pref* bugs. WRONG.
8) Have you seen Devs in a group with high IQ SDETs react to bugs or code reviews from these SDETs? They feel very threatened by these SDETs who have a more complete understanding of the component and therefore constantly challenge the parochial fixes that these Devs do. Again, WRONG
9) Devs continously point fingers at things that don't go perfectly right like the auto bug tools but will not pause for a second to consider the similarity to the bugs in their code. WRONG.
Before you say it, I know that not all Devs are bad and there are very good ones. I manage some and work with a few. But it is a shame that all most Devs do these days is to complain and complain, and to denigrate the efforts of others.
To my peer Dev Lead who has been posting his complaints here (I know because he voices the same to anyone who'd listen), I say quit complaining about how everything is messed up and contribute your own quota to make things better. Yes, certain areas of MS need improvement, and Mini's blog has provided the jolt our leaders need. But complaining perpetually will not solve any problem, especially since these Devs will not take the option of quitting MS. That's right, RESIGN. Of course they won't quit but will discourage people from joining MS. Pathetic!!
And to everyone working on Vista, thank you for the solid Feb CTP release. If you have been waiting to self-host, just get yourself a taste of the Feb CTP.
Microsoft will most certainly bundle MSN search with Vista. It will be the default search engine in any web search field and might have the same effect setting MSN.com as the default Internet Explorer homepage has had on site visits.
That will certainly make the DOJ's job easier, won't it?
Regarding prefast tools- I think they they are a good investment and do help in finding some real bugs - especially ones that could to security issues
Yeah, the process wonks always show how Pre* would have caught some past security exploit. No one mentions of course that the tool was built after that exploit was already known. Static analysis is harmless, but how it gets done in Windows is far too wasteful - those code brahmins should be forced to review any issues their tools find before they get filed as pri 1 sev 1.
To my peer Dev Lead who has been posting his complaints here (I know because he voices the same to anyone who'd listen), I say quit complaining about how everything is messed up and contribute your own quota to make things better. Yes, certain areas of MS need improvement, and Mini's blog has provided the jolt our leaders need. But complaining perpetually will not solve any problem, especially since these Devs will not take the option of quitting MS. That's right, RESIGN. Of course they won't quit but will discourage people from joining MS. Pathetic!!
Why would you have a problem with everyone doing exactly what you want?
Dev does their job of writing code and they leave PM to write their own specs and leave test to write their own test cases. Succeed in your job on your own merits.
If an SDET or anyone else finds a bug in a code review, file a bug, create a test case and make it a BVT if it is critical. No arguments necessary just write a test case.
If you're right, everything gets done.
If you're wrong, we lose a bozo.
Everybody who isn't a bozo wins.
Devs show their immaturity by complaining about Process, forgetting that process is critical to achieving software engineering perfection.
Really? Then how did Linux achieve worldwide popularity in less than 2 years, with programmers scattered across the 7 seas? Was it through "process"? I think not!
Of all the bureaucrats listed on the exec press pages, Amitabh is one of the few VPs that deserves his bonuses!
Thanks for joining us, Amitabh. If you don't mind, would you help me fix some prefast bugs though, instead of commenting on this blog? We could really use the help, because we're all idiots in my team too focused on shipping a product instead of smart guys like you and the quality gates tools "devs" who strive for engineering excellence.
Despite LisaB's listening tour, I like others that post on Mini feel that she is at the same "political" livel as senior management.
Why don't we send an email to Tanya Clemons - Corporate VP, People & Organizational Capability?
Maybe she might do more than LisaB?
5) Like some directors badly want to be actors, some Devs badly want to be PMs just because the "Manager" in PM sounds cool. Thus they do every thing to sabotage their component PM. WRONG
Quiet on the set! Action!
PM's are spread across projects and are responsible for multiple features.
They are graded on a curve like everyone else and, with that in mind, based upon how they prioritize features, they decide where to spend their time.
If you're working on a feature they don't consider important, what's to sabotage? They don't do any work to sabotage.
Why don't they cut the feature if it doesn't deserve any of their time?
Cut!
Despite LisaB's listening tour, I like others that post on Mini feel that she is at the same "political" livel as senior management.
Why don't we send an email to Tanya Clemons - Corporate VP, People & Organizational Capability?
Maybe she might do more than LisaB?
Tanya Clemons works for LisaB.
HR is going to figure out how to hide forced ranking or rename it. They are not going to fix it.
When it comes down to it, compensation is the real ranking.
Interview with some other companies and see if they give you the same 'score'.
8) Have you seen Devs in a group with high IQ SDETs react to bugs or code reviews from these SDETs? They feel very threatened by these SDETs who have a more complete understanding of the component and therefore constantly challenge the parochial fixes that these Devs do. Again, WRONG
Sounds like yet another hiring manager mistake.
They have a developer who understands a component and then they hire someone else. Maybe they got confused by the 'T' on the end of 'SDET'.
Management needs to do a better job of hiring hiring managers.
Managers who manage managers shouldn't manage drunk.
If you manage to fix your management problems, drop by again. We'll be here.
4) Still on the jealousy tip, have ever seen Devs resisting designs from a deeply technical PM just because the PM in better versed in the new area than the Devs? WRONG
Never seen that happen.
A deeply technical PM? I saw two in over ten years.
A deeply technical PM? I saw two in over ten years.
I wouldn't have thought there was any harm in naming them, in that case. No one would complain at being named as a deeply technical PM - especially if, by extension, it meant that all the otherwere idiots.
I wouldn't have thought there was any harm in naming them, in that case. No one would complain at being named as a deeply technical PM - especially if, by extension, it meant that all the otherwere idiots.
Naming someone in that way sets them up to be knocked down. If you've ever come across a teacher doing that to a student in elementary school, you would know what I mean.
PM's have these dual roles of having to write specs and having to deal with customers.
Most seem to prefer the role that lets them travel and see the world.
Specs are relatively boring to most of the ones that I have come across. The specs are rarely kept up to date as the project progresses.
A lot of PM's are technical in other areas (e.g. mathematics, nuclear physics, biology, etc.) but really don't have an interest in computer science and it shows in what they come up with for the software features they own.
Given what they know, they aren't idiots. However, only looking at what they produce would give you that impression.
Are you saying that Microsoft is wasting time on features that very few people use?
No. I'm saying the logic used is wrong. "1 in 100" as it was used above means that out of 100 bugs the tools detect, only 1 is valid. Whether or not the estimate "1 in 100" is accurate, it is not the frequency with which the bug is encountered by our customers, except possibly by some strange coincidence.
A dev, SDET and PM met in the hall one day. Here is how the conversation must have gone:
DEV: points to the PM -> Your fault!
PM: points to the SDET -> Your fault!
SDET: points to both -> Your fault!
N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! (Meanwhile, little gets done)
Let's use the latest release of Microsoft CRM 3.0 as an example. My company is struggling trying to use this thing. Clearly, the conversation outlined above was taking place, because very little effort was being made creating a usable product. Where CRM should integrate with Outlook Calendar and Task, it doesn't and can't. Where it does integrate makes little sense. The areas we want to modify and customize, we can't because it won't. The areas we can modify and customize don't make any sense. Simple things like auto-filling fields when creating new activites don't happen, and apparently can't. Reports don't. Scripting doesn't. If you have Outlook open, the IE interface behaves differently and in odd ways than if Outlook is closed, which it never really is. You send an email from CRM and it appends a very ugly tag TO THE SUBJECT LINE where your potential customer can see that he is not important enough to get a non-mass generated email, if he gets the email at all because most spam filters pick it up and discard it because of the ugly numerical tag inserted on the subject line identifying the email as coming from a CRM system. Organization and workflow are non-existant. What we have is a highly modifiable product that doesn't do anything well, and some things not at all.
(Meanwhile: N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! )
Meanwhile, the Dev, SDET and PM continue their conversation, finger pointing, barrier building, turf protecting, ass covering and general whining. Their managers are to busy trying to look good to their managers to pay attention to the fact that their teams are shipping unusable software to a captive audience.
I left Microsoft over a year ago as a L67 GM. My background is startups, which is how I ended up at Microsoft a number of years ago.
The Microsoft inbred problem pointed out earlier is real. The people making the key decisions have little or no experience in the real world, hired into Microsoft right out of college 15 years or so ago. Their children (the trio having the hallway conversation) are flawed as a result of the very small gene pool.
(N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! )
Microsoft is in dire need of a dose of reality, brought in with new people that have had to live and survive outside of Microsoft. The company has lost sight of the fact that there are customers at the end of the food chain that have to use the CRM 3.0 products being released to run companies, make decisions and keep the economy growing in the right direction. This lost sight is apparent in the products being released these days and cannot be argued with any conviction.
The product matters to people other than yourselves. It has a life after you decide it should be released into the wild, people and companies have to deal with the bad design decisions, the flawed usability and the bugs you left in because you got tired of the process and wanted to work on something new. Once you let it go, it doesn't just fall off the edge of the table, never to be heard from again. We end up with it, and you are making our professional lives miserable.
Get a clue Microsoft. Get a clue Bill and Steve. Get a clue Sinofski, Valentine, Tatarenof, Thompson, Mundie and all the other VP's responsible for allowing this unusable flawed software to ship to paying customers (don't even get me started on MOM, the most flawed piece of enterprise software known to modern man. Thanks Debbie, Quentin and Kiril for that little blessing). You are not going to get another chance once you break this one. Companies like Microsoft come along once a generation if we are lucky. You, the leaders of this once great empire, are allowing it to stumble and fall. If you are tired, leave. Lead, follow or get out of the way because you are destroying the company otherwise.
Meanwhile, back in the hall:
N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! N0, you! ...
A deeply technical PM? I saw two in over ten years.
Thanks for proving correct, my earlier assertion that most of my fellow devs are psychologically immature and afflicted by inferiority complex. Go and find out the meaning of the word "Program" as it relates to Projects and then you may have an idea of the role of a program manager.
Two words for you "Social Skills". Every Dev manager/lead should be harping on the need for their Devs to improve their interpersonal skills and learn the structure behind the business unit of PM, Dev, Test & Product Marketing
"Get a clue Microsoft. Get a clue Bill and Steve. Get a clue Sinofski, Valentine, Tatarenof, Thompson, Mundie and all the other VP's responsible for allowing this unusable flawed software to ship to paying customers (don't even get me started on MOM, the most flawed piece of enterprise software known to modern man. Thanks Debbie, Quentin and Kiril for that little blessing). You are not going to get another chance once you break this one. Companies like Microsoft come along once a generation if we are lucky. You, the leaders of this once great empire, are allowing it to stumble and fall. If you are tired, leave. Lead, follow or get out of the way because you are destroying the company otherwise."
I'm a former employee and while I can't comment on the CRM situation per se, your other comments would be absolutely consistent with what I saw when I was there and what I'm seeing externally now. I especially like the "If you're tired, leave" although sadly, the folks who most need to go in my experience where the ones least likely to be self-critical enough to note their execution challenges and/or too wedded to chowing down at the options trough to consider it even if they have the appropriate self-awareness.
>> A deeply technical PM? I saw two in over ten years.
I saw two in 6 years. The rest just talk smooth (and not all of them, mind you), pull requirements out of their asses, "check on status", and travel abroad "to study customer interest" on company dime. Quite frankly, I think if we fire 50% of PMs in our org and make the other 50% work, productivity will skyrocket.
>> A deeply technical PM? I saw two in over ten years.
I know some teams like C# team have technical PMs out of necessity. But once a few non-technical PMs get hired, they hire more of their ilk. And then social skills, presentation skills, "interpersonal awareness", (insert your own non-technical skill here), become more important.
Tell me this. What good are all these skills when you do not install your own product! Yes, I have seen PMs who do not install the product, do not know what it does exactly, but have all those other skills listed above. If any issue comes up that is techincal in nature they run to the Devs and SDETs (by taking turns so that no one dev or sdet takes too many questions).
Sad thing is, it is now accepted in many teams that PMs need not be technical. "Technical stuff can be learnt. But you cannot learn social skills". This will ultimately reflect in the product's usability.
But then again, given MSFT's stock price, pay, cost of living in Remond-Bellevue area, this is the quality of employees we are able to attract and we have to work with what we have. (applies to all employees - PMs, Dev, test etc). Don't blame hiring managers. They can only choose among those that apply.
I'm getting quite sick and tired of HR emphasizing on these "soft skills" and how "technical skills can always be learnt." - NO Goddamit! It's NOT as easy to learn technical skills as it is to learn social skills! Why the hell did we go through 4 years of hell in college doing our CS/CE/EE programs??? So that we would learn how to swirl wine while scooping out oyesters thereby showcasing our "interpersonal skills"?
How many people in marketing really even know about what our core technology does? Hell, BMO, EPG, and HR wouldn't even exist if our forefathers at MS hadn't made OS/2 a success!
In Google, engineers ARE the prima donnas and all the other MBAs HAVE to take a second seat - they realize their place, as they should, that without engineers, there IS no company! HAH!
Things started to go bad as soon as SteveB took control as CEO. He's an MBA (well he dropped out but whatever) and as such has instituted MBA-like bureaucracy - all those useless "executive reviews" were just so that his underlings could have something to do.
BillG, as arrogant as he may be, is still a smart technologist - I think he realized that SteveB is taking the company down the gutter but there's really nothing he can do about it. I mean, if he fires Ballmer, it's SUCH bad PR. The stock would tank to a level that's WORSE than Sun!
It's time to redistribute the company profits to those that REALLY do all the work here and that's us Product Devs!
Let's use the latest release of Microsoft CRM 3.0 as an example. My company is struggling trying to use this thing. Clearly, the conversation outlined above was taking place, because very little effort was being made creating a usable product. Where CRM should integrate with Outlook Calendar and Task, it doesn't and can't.
Some of the people working there ended up there after being forced out of the previous product they worked on for many years after the product reached $1 billion in revenue per year.
So, several of the people working on CRM did not have a lot of experience with that type of product.
The type of market research that lets you know what customers want seems to happen slowly over several years at Microsoft.
As for integration, when your product is not making much revenue for Microsoft, other product groups tend to put you at the bottom of their list in terms of priorities.
Thanks for proving correct, my earlier assertion that most of my fellow devs are psychologically immature and afflicted by inferiority complex. Go and find out the meaning of the word "Program" as it relates to Projects and then you may have an idea of the role of a program manager.
Two words for you "Social Skills". Every Dev manager/lead should be harping on the need for their Devs to improve their interpersonal skills and learn the structure behind the business unit of PM, Dev, Test & Product Marketing
Thank you for providing an example of a manager that does not listen in spite of a problem being identified.
This is the type of person you should let fail.
Don't prop them up by fixing that which they don't recognize as a problem.
If they're right, the job gets done by doing what they say and, if they're wrong, they rack up another failure which brings them one more step closer to being shown the door.
As a Dev I am really ashamed at the putrid junk being spewed on this blog by Devs. Yes 85% of the whining here come from Devs. I am a Dev. I know my ilk when I see them.
And yet 98% of generalizations are incorrect. Just because we talk about tools and bugs doesn't mean we're all devs, does it? (By the way, I like your capitalization of the D in devs, it shows respect. ;-))
Consider these facts:
Another generalization. You meant to say "I consider these as facts".
If you don't mind, I'll skip the more ridiculous ones.
4) Still on the jealousy tip, have ever seen Devs resisting designs from a deeply technical PM just because the PM in better versed in the new area than the Devs? WRONG
Which makes me think you're not a dev. Granted, quite a few PMs are deeply technical but not in the field of software engineering. They have expertise in the domain they're probably working in, and their role is to model a component/technology based on the customer requirements that they gathered. The dev's role is to design the software construct(s) that implement the PMs 'roles and interactions' model. That is not always trivial, is not always feasible and, when egos are involved, I can see why this could be seen as a clash of wills. I could give you plenty of such examples when a pretty simple and clean PM spec translates into headaches for the dev/tester, because of the constraints met at implementation time. Think security gaps, a diverse hw ecosystem, available computation power, availability of specific OS primitives (such as a privileged process type) etc etc. (Sorry for being vague, this could identify the group I'm in.)
Call me naive, but I believe that most of the time (minus possibly 2 periods/year), most people at MS actually try to do their (best) job. People aren't thinking "I hate my PM, I'll key his car and grenade his spec", or "I hate my dev, I'll call him stupid and give him a sh!tty spec." I really think people try to solve problems, negotiate the spec and plain talk about their differences.
To get back on track - yes, the PMs might know better the field, from their point of view. The dev/tester also know their parts better than the others. (I'm not saying exceptions do not exist, but they're hardly the norm). It takes all three of them to develop the component and make it work.
7) Devs show their immaturity by complaining about Process, forgetting that process is critical to achieving software engineering perfection. They complain about every thing but the truth is that Process exposes their flaws for all to see, like those Pref* bugs. WRONG.
I take exception here, and my conviction that you're most definitely not a dev is strengthened. There is good process (spec/design/code review, triaging, team meetings) and there is bad process (automated check-in systems that delay even the most trivial of changes for days/weeks, test passes on unsupported configurations that must always pass 100%, throwing a team in bug jail to investigate/fix spurious code parsing-generated bugs, going to war personally to get approval for a pref* fix etc). If you'd like more details on why the latter category is bad, kindly let me know.
If you firmly believe pref* bugs "expose their flaws for all to see", allow me a condescendent snicker. Example:
Foo* pFoo = NULL;
FunctionWhichAllocatesFoo(&pFoo);
pFoo->Blah = Something;
Plenty of these in old code (not SAL annotated) that are filed as pri 1/sev 1. It doesn't take long to identify these, obviously, but they do add up over hundreds of thousands of lines of code. Yes, we should fix these, but not "drop everything else in the new code, just before the RI" fix these. Couple this with a check-in system that is more on the floor than it up and running, and your frustrations of not getting real work done would probably increase, too.
And lest you think it's always the same - of course not. I'm not the type to focus on what works well, but quite the opposite - just like everybody else.
8) Have you seen Devs in a group with high IQ SDETs react to bugs or code reviews from these SDETs? They feel very threatened by these SDETs who have a more complete understanding of the component and therefore constantly challenge the parochial fixes that these Devs do. Again, WRONG
I happen to be an SDET, actually. Leaving the IQ aside, it is my job to find bugs in dev code, through any means. I do code reviews (and I find bugs), I work on specs and on occasion, even write some code. The reaction I've seen from devs was consistently one of appreciation - wouldn't you thank someone who helped?
Now, if I were thinking I know the code better than they do, and if I were suggesting my superior fixes without understanding side effects (and do so frequently), I'd probably get dismissed quickly. In my distorted reality (powered by my unbelievable IQ), I'd probably think the devs feel threatened by my mad codin' skillz.
9) Devs continously point fingers at things that don't go perfectly right like the auto bug tools but will not pause for a second to consider the similarity to the bugs in their code. WRONG.
You missed the point of the whining - it's not that the tool is flawed, it's that the timing and the heavy-handed enforcing of running it impacts work that needs to be done. And on top of it, the tool is flawed. Those types of bugs can be fixed after the cut-off date, but the incomplete feature is punted into oblivion.
To my peer Dev Lead who has been posting his complaints here (I know because he voices the same to anyone who'd listen), I say quit complaining about how everything is messed up and contribute your own quota to make things better.
Not sure if you're referring to me, do we know each other? It seems that criticism which you can identify with is considered "constructive"; all the rest is called "whining". I'm not going to come here and wax lyrical about the great job everyone is doing - we have top mgmt for that. I come here to vent my frustrations (and poison others with my negative energy ;-)). That way I can go back relieved, and do all the little things that I can do a bit more efficiently. It's not all bad all the time, but I'm not getting that frustrated when things are going well.
Looks like I've filled Mini's inbox - sorry 'bout that.
Let's use the latest release of Microsoft CRM 3.0 as an example. My company is struggling trying to use this thing. Clearly, the conversation outlined above was taking place, because very little effort was being made creating a usable product. Where CRM should integrate with Outlook Calendar and Task, it doesn't and can't.
EXCUSE #1:Some of the people working there ended up there after being forced out of the previous product they worked on for many years after the product reached $1 billion in revenue per year.
Excuse #2:So, several of the people working on CRM did not have a lot of experience with that type of product.
Excuse #3:The type of market research that lets you know what customers want seems to happen slowly over several years at Microsoft.
Excuse #4:As for integration, when your product is not making much revenue for Microsoft, other product groups tend to put you at the bottom of their list in terms of priorities.
In the words of Bill O'Reily, "this sounds like spin to me."
To me, the customer, I could give a rat's ass why it is happening, I am only affected because it is happening because I am being asked by an over-anxious Microsoft sales person to pony up $10 grand or more for the CALs I need to see if I can actually make this thing work for us.
The fact that within hours of my posting a real-life customer position I am presented with not one but FOUR excuses clearly points to the problem. The customer is on Microsoft's priority list somewhere just below making sure the Bldg 43 cafeteria doesn't run out of trays at lunch.
Until the response from within Microsoft to a customer's pain and anger over what is being passed as usable enterprise software is closer to "You are right, this situation sucks and we are to blame", Microsoft will continue sliding down the dark path to market failure.
By the way, for you Microsoft inbreds listening to this blog: What lies at the bottom of that dark path to market failure is the following:
1. Missed numbers
2. Stock price failure (you have no idea what you are in for, the current five-year pain is nothing compared to what lies at the end of this path)
3. Massive layoffs, resulting in property values crashing and you, the Microsoft inbred, left on the sidewalk holding your patent cubes and wondering just what the heck happened and who is to blame.
WAKE THE F**K UP MICROSOFT, INERTIA CANNOT LAST. THE CUSTOMER MATTERS! YOU SHOULD ALL BE EMBARASSED AT WHAT YOU HAVE TURNED MICROSOFT INTO.
Mr "Ex-L67 GM CRM stinks", it feels good to complain now that you are on the other side huh? What did you achieve while in MS? Weren't you one the people that screwed up Exchange?
Mr "Ex-L67 GM CRM stinks", it feels good to complain now that you are on the other side huh? What did you achieve while in MS? Weren't you one the people that screwed up Exchange?
Nice. No acknowledgement of the issue, no understanding that I am using CRM as a single example (there are plenty of others) because I am fighting that pain right now, just a finger fully extended, first pointing at someone else, then proudly uplifted into the air telling the very customer that pays your salary that he is just sour grapes.
No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you!
And no, you have me confused with someone else, I was never involved with Exchange. I am proud to say that I recognized that I was unable to get anything productive done at Microsoft due to the infighting, finger pointing and back stabbing by the more seasoned Microsoft inbred, and I left, rather than staying (which I easily could have) drawing north of $200 grand a year and not really doing anything. I left over 250,000 options on the table when I left.
My pride and dignity are worth more than that.
You, on the other hand, are still there....
As far as PMs go, from my experience, it all seems to fall apart at the specification stage which is what I see as a developer and, despite the job description of program management, I have not come across of many technical PMs.
writing "WOW" emails
Emails have become the official means of recorded conversations in corporations. At Microsoft, a typical program manager writes an average of 30-40 emails every day and reads many more. It would certainly help improve productivity, if the emails are crafted in an optimally sound way. Guy Kawasaki has done a great job in delineating the craft of email writing in a recent post.
Let's change the world!
PM at Microsoft
While at Stanford this week I was asked by a number of PM (program manager) candidates to talk about the PM role at Microsoft. The PM role is unique to Microsoft and was actually created in response to developing software that is more usable and at the same time pushes the state of the art of technology.
The job title "program manager" is a bit of a misnomer, since program managers do not program nor do they manage--go figure.
So at Microsoft a new role was created in Program Management with the explicit goal of partnering with development and working through the entire product cycle as the advocate for end-users and customers.
Where developers were focused on code, architecture, performance, and engineering, the PM would focus on the big picture of "what are we trying to do" and on the details of the user experience, the feature set, the way the product will get used.
A PM serves as the coordinating point where all the disciplines come together--this is why you can think of the PM as being the hub. What is critical about the Microsoft PM role is that all the different people serve as a "virtual team" building the product--the PM does not have to go pull them together but rather has to help get everyone aligned.
So what does a program manager do?
Learn --> Output of learning is a prototype
So often we will spend days at a customer's place of business learning from them and watching them trying to get their work done.
Convince --> Output of convince phase is a plan and goals
The output of this phase is the set of objectives for the project area.
Spec --> Output of spec phase are a series of written specifications
So in writing a specification, the PM that worked on the learn phase now has to turn that learning into an experience that customers will have.
So if your spec omits a lot of details and developers need to guess, then there is a good chance you will end up not being able to realize all of your dreams because too many things needed to get filled in a the last minute, plus your developers will not be very happy working with you. So doing a good job on the spec is important.
Refine --> Output is the product
During the development of the product, PMs are constantly refining the details, working with development and test to determine what needs more clarity and what is working better than expected (that can happen) and what is going less well (that happens more). PMs are on call to answer questions and "unblock" development.
Of course our senior program managers are also responsible for working with the press and product management.
PM Cheat Sheet
-A clear communication plan should imply additional informal relationship management strategy (luncheons, one-one's, after-hour events
-The Event-based task follow up gives you accurate measure of progress (based on reporting of events via feedback plan), whereas the new time time-based task follow up forces you to assess (via the perception of task owners adjusted with estimated vs. actual time) and manage the schedule.
Where's The Program Manager At?
Market research plays an important role in aiding his efforts, but in the end, it is the PM's call. A well thought out specifications list makes the development teams life a lot easier.
To me, the customer, I could give a rat's ass why it is happening, I am only affected because it is happening because I am being asked by an over-anxious Microsoft sales person to pony up $10 grand or more for the CALs I need to see if I can actually make this thing work for us.
I would call what was outlined 'reasons' and not 'excuses'.
For those 'reasons', I would not 'hold my breath' if you're looking for a quick resolution to your problems.
I would use those management skills of yours to convince your customer to try a different product.
Give them a demostration of how to do the tasks they want to do in a several different products and see which one they choose based upon a better understanding of what is out there.
If you don't get rid of this source of stress in your life, you will have more problems than just an unusable product.
I was reading this and I thought to myself, WOW.
One of MS' biggest problems is that they chose perhaps the worse climate in the WORLD to base their hq.
Nearly EVERY employee has to be "imported" because WA has NO real CSc programs. This combination means that THERE WILL BE bozos hired just to get a body in a test seat.
It also makes me wonder why the SDET initiative was undertaken. If you can barely get process contained with a separate test team how can you manage the beast called "testing with your own code." This will mean that SDETs are spending more time switching gears than doing either job.
I am leary of Vista.
"One of MS' biggest problems is that they chose perhaps the worse climate in the WORLD to base their hq."
Nops. One of MS' biggest problem is too much money in the bank. This provide too big a safety net which allows execs to act silly. Further, this cash pile allows other companies, governments to claim a part of it. What EU is thinking? By colluding with complainants, EU showed that they are only interested in the cash pile which MS has.
I think MS should have got rid of this cash pile few yeara earlier. RealNetwork settle was gross. Giving them more than their market cap? Market cap of realnetwork represented what its owner (shareholders) estimate its business value (which included any and every potential settlement with MS). Look at the recent settlement of Blackberry. NTP patent is the lifeline of Blackberry. Shareholders of Blackberry were expecting to give more than billion dollars to NTP but still Blackberry executive's did the settlement at just over six hundred million dollars.
Why Microsoft executives always overpaid to settle past cases? Because they have so much contempt of hard earned money? Because the huge cash pile allows them to put their fist in the pot of gold and settle all the trouble they were facing. Let us see how they resolve the recent EU game.
One of MS' biggest problems is that they chose perhaps the worse climate in the WORLD to base their hq.
Nearly EVERY employee has to be "imported" because WA has NO real CSc programs.
Excuse me! The University of Washington (Seattle) Computer Science Department has consistently been ranked in the top 10 C.S. departments over the past decade. In fact the following even ranks it as #1 in the nation: http://www.cs.ucf.edu/newsletter/vol1/issue_one/ranking.html
A UW Ph.D. alumnus who chose not to work for Microsoft. And whose 1981 interview schedule included Paul Allen, Steve (we have free cokes here) Ballmer, Charles Simonyi and Gordon Letwin.
Excuse me! The University of Washington (Seattle) Computer Science Department has consistently been ranked in the top 10 C.S. departments over the past decade. In fact the following even ranks it as #1 in the nation: http://www.cs.ucf.edu/newsletter/vol1/issue_one/ranking.html
There are ONLY 5 million people in all of WA State. It is in the middle of nowhere and is the most dreary place I've ever been. That in and of itself will cause lowered output whether you want to admit it or not. Add the process and stack and MS is in real trouble.
It's time to redistribute the company profits to those that REALLY do all the work here and that's us Product Devs!
Sounds good as long as there's a $500 dollar fine for each bug that goes out the door. (Think about it: Regardless of the roles of the other disciplines, the bugs were written in by the devs.) That would help clean out devs who would be "good attrition" real fast.
Lose the Marie Antoinette-type "Let them eat cake" attitude, Drei. Wherever the heck you work, it sure isn't like my group or where ever the heck Mr. Capital-D Dev works.
Over here, PMs are technically illiterate nearly across the board. Devs here do, in fact, whine about process being adopted in a place where there was essentially NO process. (Usage of code analysis and code coverage tools are just now being adopted.) They also openly mock testers too.
It's arrogant and foolish to dismiss his rant just because you work in a place that's not dysfunctional.
RE: CRM 3.0 guy
Loosely quoting the guy who was PUM there in 1.0 timeframe, "when you eat a sausage you don't know what kind of meat it was made out of". He said this to a room full of employees implying that it's OK to ship a poorly architected system with known, horrible bugs. This was after several schedule resets (each with massive feature cuts) and a revolt in dev team when he attempted to force (!) everyone work weekends.
Believe it or not, things have improved dramatically since then. They've changed that PUM (promoted, now he's at almost VP level), hired some more senior devs, demoted their V1 architect to an IC (because he was a pompous, smooth-talking ass who didn't even know how to architect a DB schema properly). Those senior devs rewrote most of the product the way it should have been written to begin with. To reiterate, it is now MUCH better than it was in V1.
Having said that, after they shipped 3.0 lots of good people there threw in the towel and moved on to greener pastures. Which kinda tells me CRM is still a dump for an IC not interested in brown nosing and politics.
Turns out customers do care about the taste of the sausage after all, especially after they pay dearly for it.
It's time to redistribute the company profits to those that REALLY do all the work here and that's us Product Devs!
Sounds good as long as there's a $500 dollar fine for each bug that goes out the door.
THANK YOU!!! The so called smart, technical 'better than everyone else' developers forget that without their crappy coding there will be no bugs. Yes no product but NO BUGS.
As a dev, if you want me to write code without bugs (which is something I make every effort to do right now), I'm going to require the following things:
1. Quadruple the dev cost estimates, add generous buffers to the schedule on top of that to let Devs properly integrate the pieces
2. Spec is completely frozen before coding even begins
3. PMs who came up with the specs resulting in code rewrite within a product cycle are fired on the spot.
4. Schedule is put together taking the sequencing of work into account. No more coding in vacuum. No more "parallel development". If a dependency is required to develop a piece of code, work does not start until work on that dependency is finished.
5. Management makes every effort to NOT disturb/disrupt/randomize dev process. Each dev concentrates on only one task at a time.
6. Bug bar is abolished. Schedule deadline moves for as long as there are bugs to fix.
7. There are big payouts come ship time. And I do mean big, a Tivo or MP3 player won't do it.
Make this happen and I'll accept your $500 per bug thing for bugs found in my code. I'm confident in my ability to write extremely high quality code. I just can't do it in the environment that exists at MSFT right now.
I am an SDET. Thanks to all SDE's whose poor code is helping us keep our jobs.
"Make this happen and I'll accept your $500 per bug thing for bugs found in my code."
So you as a dev accepts that there are in the company who do or are expected to do some important work. For an example satisfy a your requirement so that you do not write buggy code. That's the reason not all the money should go to the devs unlike one dev proposed that profit sharing should be re-balanced in favor of devs because all the work is done by devs. When devs decline to take responsibility of writing a buggy software and make somebody else responsible for it. Then that means in re-balancing of profit sharing should be done in favor of those who are responsible. Salaries are proportional to responsibility.
I am an SDET. Thanks to all SDE's whose poor code is helping us keep our jobs.
Thank the GMs who value making arbitray deadlines over quality. Thank them for not letting piddly irrelevancies like Trustworthy Computing get in the way of getting the next rev out the door. Thank them too for buying out companines and bringing their lowest staff in at levels higher than you'll reach in a decade of service. Thank the PUMs, dev managers, and test managers for setting an aggresive schedule, creating a farcially long list of asinine team committments for reviews created solely for the purpose of making them look like they are Doing Something, and have the cheek to add, "Oh, but we want you to have work-life balance too, though we won't reward you one jot if you give it up to make everything succeed". Thank the leads who reward people for going through the motions to meet said asinine commitments instead of getting the job done. Thank the horde of PMs who think mauve DBs do indeed have the most RAM, the significant number of devs who really shouldn't have been allowed to touch anything other than LOGO, and the career STEs who got bumped to SDETs only because the team needed warm meat to keep things from hitting the fan.
Thank you all for making MS the place it is today.
>> in favor of those who are responsible
You must be new here. NO ONE takes responsibility. There are whole bureaucratic constructs (such as war team and triage) that insulate those in power from the responsibility for their actions.
"NO ONE takes responsibility. "
Exactly my point. Devs are not taking responsibility either. On one side dev are saying they should control the company, have higher salaries because they do all the work. If that is the case then why are they complaining if devs mess up because some other set of people, like GMs and PMs, mess up. This logically implies that devs are not doing/or not assigned all the work. There are some other roles who have equally important work to do. I am not saying what quality of works they do. The average quality of employees, whether at PM level, or dev Level, or at test level is what is called 65 percentile. If you point an incompetent PM, I can point to an incompetent dev.
There's a critical flaw in your reasoning, though. If tomorrow all PMs and testers are fired the company will still exist. Devs will just write their own specs (as they in part do now) and write their own test cases (as again they do now). If devs are fired, PMs and testers (and PUMs, GMs, VPs, CVPs, CxOs) are useless.
I actually have nothing against testers who take time to maybe attach a debugger and see what the problem is on a deeper level. 20% of testers that do that - those testers I'd keep. But PMs are just such a waste of shareholders money in 99% of the cases - I'd shrink their headcount dramatically, so that there's at most one PM per five devs. And that PM not only should have social skills, but he must be technical as well. I'd be OK with him earning twice as much as I do if he doesn't fuck things up and really knows what the customers want.
THANK YOU!!! The so called smart, technical 'better than everyone else' developers forget that without their crappy coding there will be no bugs. Yes no product but NO BUGS.
Don't make me laugh!
Without the last minte changes and code rewrite requests coming in from the PMs which flowed down from management and the 'marketing gurus' who have discovered a sudden new 'customer trend', there would BE no 'crappy coding' as you call it.
The main problem here is that us devs take our directions from those that know crap all about technololgy.
How about:
1) $1000 deducted from the GM-level above EACH time a change is made at the PM level
2) $1500 for EACH day that is delayed beyond the actual ship date.
So, in this case, BrianV would have to accept deductions from ALL the PM-induced changes that roll up to him - thus, the higher up you are, the more the penalty. Guess what's gonna happen? BillG would rapidly drop like a rock from the world's richest people...
Devs will just write their own specs (as they in part do now) and write their own test cases (as again they do now).
For any project of significant size, yeah, it's still possible for devs to write their own specs, although only in the sense that it's possible for them to refine their own gasoline. And, yeah, it's also still possible for devs to test their own code, although only in the sense it's possible for them to do their own dentistry. Anything's possible, but not everything's practical.
Crafting a successful piece of multi-MLOC software requires a team of specialists in various areas to shepherd all phases of its creation. That quite a number of devs still believe in the lone hacker stereotype instead of realizing this is just another example of the egotism and, dare I say it, social ineptitude that seems to be an occupational hazzard of being a dev.
Seriously, if you're really so gosh-darned special that you can do the job of three people, why are you still here? Why aren't you out starting the next Microsoft?
That quite a number of devs still believe in the lone hacker stereotype instead of realizing this is just another example of the egotism and, dare I say it, social ineptitude that seems to be an occupational hazzard of being a dev.
No, it's the 'Where the f*** is everybody?!' reality of it all.
Where's the planning? What do you mean we're changing dependencies again?
Where's the PM? He's at a conference or working on a higher priority feature? Write your own spec. Don't want to. Just have to because no one else has written it and the schedule says I have to start coding soon.
"For any project of significant size, yeah, it's still possible for devs to write their own specs, ... ... Why aren't you out starting the next Microsoft?"
Well said. But we dev still like to treat all others like lower castes animals for the argument sake. Finally we do not understand is what really gives power and money is the leadership skill. Do you know a guy called George Bush? He does not know how to develop softwares. But he knows what dev do not called leadership skill. There are very few George Bush's - though every dick and harry thinks himself better than George Bush. It is a paradox.
Devs, your job is to do code development provided all the high level design stuff is done. If you think you could do design stuff and the job of a designer is easier and better paid then why did not you interviewed for PM or similar job? If you could do all kinds of job then you must have chosen the most rewarding job. Then why complain? If you could do all kinds of jobs but still did not choose an optimal job then in that case either you do not understand the world or you are lacking basic judgement. Both these properties makes you a good target to satisfy Mini's goal of having a mini microsoft.
What scares me even more is this CNN article I found on how Google and its artificial intelligence capabilities will enslave us all.
Can we please ensure that this AI thing will enslave management instead?
I would say from my experience at MS, that the mgmt needs to stop setting dtaes 2 years in advance.
It would be better to just get the job done and then offer a release date. No one expects Windows to come out sooner than 2-4 years, especially server.
If all these superhero devs and managers can't get Windows out in 4 years..........
Thank the PUMs, dev managers, and test managers for setting an aggresive schedule, creating a farcially long list of asinine team committments for reviews created solely for the purpose of making them look like they are Doing Something
In fact if you think about it, a lot of their time IS spent on reviews, reviews, reviews! These overly bureaucratic review processes are instituted by HR which makes me think we'll never get rid of the stack rank system - why would they? HR and these PMS/PUMs/GMs help each other keep their jobs!
Devs, your job is to do code development provided all the high level design stuff is done.
All the high level design stuff is not done by the people responsible for doing that. That is the complaint.
If you think you could do design stuff and the job of a designer is easier and better paid then why did not you interviewed for PM or similar job?
Some SDE's to move into PM roles at some point. Others have no interest in doing so.
Again, the complaint is that specs are not being written by technically capable people or not being written at all by people who are and are apparently too busy to do that part of their job. The customer requirements are not getting translated into a document that developers can use. You then get complaints that Microsoft is providing the wrong features. Slowly, over years, the right ones eventually do get added.
If you could do all kinds of job then you must have chosen the most rewarding job. Then why complain? If you could do all kinds of jobs but still did not choose an optimal job then in that case either you do not understand the world or you are lacking basic judgement. Both these properties makes you a good target to satisfy Mini's goal of having a mini microsoft.
Maybe they are on an H1B and cannot easily change jobs. They're getting paid less and, because someone else is not doing their job, they are picking up the slack in addition to doing their job.
"There are ONLY 5 million people in all of WA State. It is in the middle of nowhere and is the most dreary place I've ever been. That in and of itself will cause lowered output whether you want to admit it or not. Add the process and stack and MS is in real trouble."
Total b.s. Lots of people love the Pacific Northwest and the outdoor lifestyle it affords. The cost of living is also substantially lower than many other high-tech hubs (i.e. SF, Silicon Valley, etc.). MSFT's problem isn't attracting good people, it's providing an environment where those people aren't hamstrung.
"The fact that within hours of my posting a real-life customer position I am presented with not one but FOUR excuses clearly points to the problem. The customer is on Microsoft's priority list somewhere just below making sure the Bldg 43 cafeteria doesn't run out of trays at lunch."
As a former employee, I'm embarassed by the answers you're getting as well but consider the source. As a suggestion, have you escalated this to the GM of MBS and/or Burgum/Ballmer directly? If I were in your shoes and not getting the answers I needed, I would be doing that as well as letting them know that if the situation doesn't get resolved in a timely manner, you're taking the entire experience to CRM magazine or equivalent. I'd like to think that Ballmer at least would respond because it's the right thing to do. But I KNOW they'll respond if they think not doing so will result in more bad PR. Just a thought.
Devs are complaining that they write or contribute to specs, and make statments like:
All the high level design stuff is not done by the people responsible for doing that. That is the complaint.
My heart weeps. I am ashamed. This kind of thinking is coming from devs in MS? So you don't know the difference between Functional, Design and Test specs? You don't know that developers own writing design specs and should and must contribute to Functional specs? You don't know that Functional specs are simplistic and should represent blackbox (input & output) design while the Design spec owned by dev should go into the technical details of implementation?
You want all the design to be completed by the "useless" PM and then handed over to the code monkeys to start? And the design will be frozen never to be changed until the uber smart code monkeys finish and then hand over to Test? Waterfall software engineering in the 21st century? Did you pay attention at your NEO when the iterative Product Cycle Model was taught?
You know, I was one of those that thought this whole MiniMsft thing was a crazy idea but given the level of thinking shown by the developers on this forum, I say let the riffing start. The layoffs should start from these types of developers, before they become something more than SDEs. There should be no place for people who are too daft to react to changing requirements and circumstances. Gawd!!!
Somebody console me and tell me these developers here only represent the minority, please.
This thread degenerated in another dev-pm-test love fest. I'm sure there will be a few shocked to hear there is a lot of bad blood in MS. It's unproductive and worse, it may not even have legitimate roots.
I think we should apply the old adage that discourages passing judgment until you've walked a mile in their shoes. After Vista ships, let's spend a week walking in the tasks of our most hated local guy. Let's put the devs in a meeting with customers, give the testers a bug list that need to be fixed, and let's have the PM diff the current matrix vs the test plan and work on the uncovered part.
I think a lot of the frustrations we have with each other come from judging on appearance. We think "they" only travel/smooth talk/hide bugs in the code/file bugs for spelling mistakes. We don't talk to each other, and when we do, it's to do what the CRM-unhappy customer accuses us of doing: "No, you! No, you! No, you!" Let's see if it's true.
I'm not saying it would have to be a formal "love your dev" training week; I'm not saying there is a metric to it, or that you should accomplish a task and demonstrate everyone how versatile you are. What I'm aiming at is to have everyone delve into the little, menial, daily tasks that each one of us have on our "to do" list, understand the unseen complications - the language barrier (have you seen the legal documentation PMs have to hammer out with 3 different vendors and the LCA?), or the knowledge barrier (pick your own example), and honestly evaluate if you could/would do that job day after day. It would probably help most people find the lost respect for their team-mates and get rid of a few misconceptions.
And if, after that, you still find yourself just as displeased (or, worse, even more) as before with the people you have around you, do the honourable (logical?) gesture and go elsewhere. It's unlikely you'll be able to change those around you and, even if you are the star of your component and love dearly what you do, you can't work efficiently filled with disgust, and working with people you can barely salute.
And before I'm accused again of grand-standing, I find myself in the same conundrum. I enjoy working in my group, but I find myself losing my respect for most of my team-mates, some of which I'd categorize as epitomes of bozoness. Too often I find myself grumbling I'd have done a better job (and quicker) than a PM or another, or one of the devs I work with. (And I'm sure they thought the same about me). I simply thought the only solution was to leave the group, which is a rather extreme (and difficult) step to make. (I've spent all but one of my seven MS years in this group.) But all these public battles here made me think mine, or others' are not simple, isolated cases - it seems the malediction between castes is indeed widespread.
Obviously, not everyone will react pleasantly to being scrutizined, even if with the best intentions. Still, talking with them face it face is bound to be better than talking about them. But I think the majority will be more than pleased (and flattered) when someone else takes interest in their trade - we're all proud professionals here.
So you don't know the difference between Functional, Design and Test specs? You don't know that developers own writing design specs and should and must contribute to Functional specs? You don't know that Functional specs are simplistic and should represent blackbox (input & output) design while the Design spec owned by dev should go into the technical details of implementation?
Yes, I know the difference between functional and design specs.
The PM is the one who is supposed to have visited the customer and would have the information to write the functional specs.
They were 'too busy' so I had to write the functional specs and write the design specs.
If functional specs are 'simplistic' and the PM visited the customer, they should have been able to produce that document very easily. How come they didn't? Please explain.
“Again, the complaint is that specs are not being written by technically capable people or not being written at all by people who are and are apparently too busy to do that part of their job. The customer requirements are not getting translated into a document that developers can use. You then get complaints that Microsoft is providing the wrong features. Slowly, over years, the right ones eventually do get added.”
As a long time customer that uses MSFT’s development tools in a professional services environment, I can understand this person’s viewpoint. I struggle getting customer requirements as well, but I think this is just indicative of software development in general. My view and 15 years developers experience is that software development is trial and error at best, despite Agile, Scrum and the rest of the method madness.
While not immediately apparent I can draw my own comparison for a leaner meaner MSFT to the developer products/technologies that Microsoft provides. How about leaner and meaner developer tools and technologies? There are at least 5 technologies (7 if you include InfoPath and SharePoint Designer) for developing user interfaces. And even more tools to support these technologies. ASP.NET, Atlas, Windows Forms and WPF
It’s ridiculous! Working in a pro services world, our customers are expecting us to quickly develop user interfaces on the MS platform, but when I have to learn a half dozen (complex) tools and technologies just to develop a UI, I get pissed.
Innovation? What would be "innovative" is one tool/technology for developing user interfaces, not 7. This would also greatly reduce the number of MSFT staff required to develop the product. And with only one UI dev product to focus on, the quality might increase along with the ship dates. From my limited customer view, I would then be “pleased customer”. Right now, on a Sunday afternoon learning SharePoint Designer as a Beta 1 (for a customer RDP program) where half the features don’t work and with no shred of documentation, I am a pissed off MSFT customer (with no life!).
I think we should apply the old adage that discourages passing judgment until you've walked a mile in their shoes.
I don't need to walk a mile in their shoes to know they are not producing that which is required by their job description to do my job.
I don't care why something is not getting done and only care that it is not getting done.
Despite having brought this to the attention of management several times, it is still not getting done.
That means management, who should be looking at the reasons why it is not getting done and fixing it, has not solved this problem which has been going on for several years in what appears to be many projects across the company.
Who is responsible for the job descriptions at Microsoft? If you give the benefit of the doubt to people who aren't getting things done, the reason is most likely to be that they don't have time.
That is also a problem.
Instead of fixing these problems, you get a bunch of managers on this blog saying developers are just making noise. Given what has been said about middle management on this blog, that is not surprising.
They were 'too busy' so I had to write the functional specs and write the design specs.
Okay, time for a ceasefire. I didn't know that all this time we were dealing with a bumbling idiot who was bent out of shape because he wrote a functional spec. You need to apologize to the dev community for almost bringing them to disrepute. Imbecile.
Okay, time for a ceasefire. I didn't know that all this time we were dealing with a bumbling idiot who was bent out of shape because he wrote a functional spec. You need to apologize to the dev community for almost bringing them to disrepute. Imbecile.
Ceasefire? More like a misfire on your part.
This is many specs over several years and this is not just happening to one person on one project.
You don't know that Functional specs are simplistic and should represent blackbox (input & output) design while the Design spec owned by dev should go into the technical details of implementation?
If functional specs are so simple, why is the customer on this blog complaining about the features of CRM?
Where did it all go wrong?
Waterfall software engineering in the 21st century? Did you pay attention at your NEO when the iterative Product Cycle Model was taught?
If by iterative you mean restarting the entire project again and again, I've seen that happen and heard about it happening elsewhere.
When I started over a decade ago, we just got some brief speech about diversity at NEO.
I did learn about Waterfall software engineering in school in the 20th Century though (you must be a recent graduate to think of it as a product of the 21st Century) and used it over several releases at Microsoft.
Dear Dev,Test,PM
Can't we all just get along. We are on the same team. Well, you are.
While this comments does not fit in the dev/pm/test fight it is an evidence of my argument in the last posting. iTune+iPod is a monopolistic abuse.
Here are some excerpts from recent news articles.
Reuters reports, "Plenty of press scrutiny may also result from a claim by Thomas Slattery in the U.S. District Court of Northern California that Apple engages in anticompetitive practices by not licensing its Fair Play DRM (digital rights management) technology to third parties."
Further excerpts from another news article:
"With respect to the "tying" allegations, the judge said Slattery was claiming Apple forces people who own iPods to buy music online only from iTunes and also forces iTunes customers to buy iPods to play the music they purchase."
It is a clear cut argument.
You PM-dev-testers are lucky to have each other to blame. In field services, the only one to blame is MCS - those guys get stuck (like a pig) for every misplanned project that comes from their sales and managers.
>>Waterfall software engineering in the 21st century? Did you pay attention at your NEO when the iterative Product Cycle Model was taught?
>I did learn about Waterfall software engineering in school in the 20th Century though (you must be a recent graduate to think of it as a product of the 21st Century) and used it over several releases at Microsoft.
Amusing. Perhaps you should re-read the original poster's comment and try again.
Amusing. Perhaps you should re-read the original poster's comment and try again.
Maybe you should read the other posters' comments regarding PM.
In the Waterfall method, the assumption is you have all the requirements up front.
Well, I didn't even get a spec most of the time from the people who were collecting the requirements from customers.
Iterate on no information?
Instead of us devs banding against the people that REALLY make our lives a living hell (i.e. management), we have turned this into a mudfest against each other's technical skills.
Management must be laughing their asses off at the irony of this situation and will only rest their case that without them, us devs can't handle ourselves and work harmoniously together.
If we want to make MS lean and mean, we first have to show management that we can do our work WITHOUT them !
we have turned this into a mudfest against each other's technical skills.
That's just it though. Working at Microsoft is a lot like mud wrestling .... a lot of slippery characters.
Unlike this blog, at work, because the company cares so much about diversity, the mud usually hits you in the back of the head. PLOP!
On here, they aim for your face.
As a suggestion, have you escalated this to the GM of MBS and/or Burgum/Ballmer directly? If I were in your shoes and not getting the answers I needed, I would be doing that as well as letting them know that if the situation doesn't get resolved in a timely manner, you're taking the entire experience to CRM magazine or equivalent. I'd like to think that Ballmer at least would respond because it's the right thing to do. But I KNOW they'll respond if they think not doing so will result in more bad PR. Just a thought.
I've seen that approach work before.
If Ballmer thinks he's going to lose a lot of money, he pushes to get a bug fixed.
BTW, who wrote the specs for Live Mail UI? There are basic problems.
For an example, there is no link back to inbox from calendar view.
The UI seems to be a copy of Outlook or OWA. Can't the person even copy something properly?
I think the testosterone-driven nature of much of the argument, here, neatly indentifies a key problem in Microsoft culture. It doesn't matter who's right, or who's the real bozo, when the point is being made in an air of defensive, self-righteous, arrogance. This seems to be a default setting, for everyone who works at Redmond: 'me and my big willy'.
Well, get over yourselves, because there's no point being right, if nobody likes you for it.
I have a suggestion for all dev\pm\test. Yes most act like everyone else is an idiot. Maybe before accusing people of being incompetent when you have a problem, simply say,
Hey, if you've got a minute I wanted to talk to you about something. I was working with "component\spec\test" X and noticed that maybe it would work better if we did this. Since you're familiar with it I thought I'd find out what you thought.
Sure some people can't be dealt with but this will show you who is just an a-hole and who is truly trying to give the customer the best experience.
Just my $.02
This thread is a classic case of why I left Microsoft. All of the whining and finger pointing happens in real life too. Not just on the pages of this blog. Microsoft is constipated, and this is just one of the many examples.
I wanted to write software, deliver it in a friction free environment, and have a good time doing it.
For me, the decision to leave Microsoft and join Google was a no brainer.
For you, you can either bitch and moan about who should write which spec, you can suck it up and enjoy your free soda pop and the other benefits of monopolistic existence, or you can check out of Microsoft and get into another situation that will hopefully deliver all that you are looking for.
Terrific blog, I was just told about it by a 'softie comrade who's been there the whole time while I've come and gone four times. I'm currently CSG.
I just wanted to echo the fact of leaving through the front door and coming in the back as a great way to get around the moribund ratings system. I quit DMD after almost three years because working there really could not be much different from Hell. I wanted to note one aspect of FTE employment I have not run across here yet: LYING.
One manager, peer to my own, lied on seven occasions that I know of, every one of them to create an impression that things were better managed than they are. DMD under the Vista project is the Seventh Circle, ten lines of code in two weeks would be the leading edge of productivity, all the rest of the time spent wrestling with BVTs, Autosmoke, the checkin system, lousy Vista builds, quality gates .. nobody is getting anything done and one project after another is getting "cut." License migration from XP to Longhorn just got cut, for example.
The system is, in a word, broken, but some junior managers think this is best addressed in the same fashion as the US planting stories in Iraqi newspapers.
I came back in less than two weeks as a CSG, in a much better group, with a much better (and HONEST) manager, even a better dev box. And if when I go blue in this group I expect the promotion that was not available in DMD at all.
"For me, the decision to leave Microsoft and join Google was a no brainer."
Google is going with even more rapid growth than what Microsoft went through. Google would saturate at much smaller scale than Microsoft has now. All in all you would see the same in Google if not worse. Neither Google nor Microsoft, neither Google employees nor Microsoft employees have come from heavens. Statistically all belong the same culture. Happier when the pot of gold is growing and fighting when it becomes a zero sum game. Happy coding at Google! Oh ya. Protect your ass because a Microsoft kick won't even leave you constipated.
Well, get over yourselves, because there's no point being right, if nobody likes you for it.
This is why performance management at Microsoft is a popularity contest.
If you bring up a problem, you're more likely to piss off someone who can affect your review score.
It is better to play it safe and just go along.
It is like that scene in 'Shaun of the Dead' ... walk like them and talk like them so they won't eat you.
Bill Hill makes an interesting comment about most users not knowing about features that are not turned on by default.
Microsoft's IE7 Ignores Windows Setting
"It certainly is unusual that IE is not respecting the system ClearType setting, but the problem is that most users do not know about features that are not turned on by default," Hill said. "We've found that a large percentage of Windows XP users don't know about ClearType, or how to turn it on.
ClearType in Internet Explorer 7
When we shipped WindowsXP in 2001, we were very conservative with ClearType. We were certain ClearType would improve the user’s reading experience, especially users with LCD displays. But we didn’t have enough experience with ClearType to know either its drawbacks or the full extent of the benefits. After 5 years of real world experience and research, we’re now quite confident that the benefits for using ClearType are significant, and it was a mistake that we didn’t turn it on by default in XP.
Neither Google nor Microsoft, neither Google employees nor Microsoft employees have come from heavens. Statistically all belong the same culture. Happier when the pot of gold is growing and fighting when it becomes a zero sum game. Happy coding at Google! Oh ya. Protect your ass because a Microsoft kick won't even leave you constipated.
The kick has been delayed on account of brain.
Seek help elsewhere to improve regularity.
Microsoft Backtracks On Timetable To Beat Google
Microsoft Corp. has tempered the chest-beating of its European president who said last week that the software giant's online search engine would be "more relevant" in the U.S. market than Google.
At a tech and media summit sponsored by news agency Reuters, Holloway, Microsoft president for Europe, Middle East and Africa, said the company's search engine would surpass Google's this year.
""What we're saying is that in six months' time we'll be more relevant in the U.S. market place than Google," Holloway said. "The quality of our search and the relevance of our search from a solution perspective to the consumer will be more relevant."
Can a Microsoft version of this game be far behind?
Microsoft never passes up an opportunity to launch another advertising campaign with someone wearing a sweater.
The elderly gamers
TOKYO: Like many retirees, Isamu Shishido sometimes forgets names and even his own telephone number. But now the 67-year-old says he has found a product that could sharpen his thinking: a new brain-training game from Nintendo Co. The ailing maker of Mario and Pokemon games has scored a smash hit by courting Japan's burgeoning gray market with "Brain Training for Adults" — a number and puzzles game that Nintendo says can stimulate the brain and ward off dementia.
Here's why testing is important.
Just because somebody said they are going to fix something doesn't mean they fixed it or that the fix solved the problem.
FDA finds benzene in soft drinks
WASHINGTON - When small amounts of benzene, a known cancer-causing chemical, were found in some soft drinks 16 years ago, the Food and Drug Administration never told the public.
That's because the beverage industry told the government it would handle the problem, and the FDA thought the problem was solved.
A decade and a half later, benzene has turned up again. The FDA has found levels in some soft drinks higher than what it found in 1990, and two to four times higher than what is considered safe for drinking water.
No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you! No, you!
Any questions?
iTune+iPod is a monopolistic abuse.
Please don't start this argument again.
And please don't expect any more Apple v. Microsoft discussions, unless super-relevant to the post, to approved while moderation is turned on. Blog to your own page because otherwise your passionate comment is at risk of falling into the bit-bucket.
Mini.
"Happy coding at Google! Oh ya."
Absolutely. How about you? Happy bitching and moaning about life at Microsoft? You really think I give a rats ass about Microsoft's ambition?
Microsft's goal seems to be complete control and domination of the software industry. Funny thing though is that when they get to the top, they turn clueless sitting around on a pile of cash waiting for another challenger to emerge.
You seem to like that "win at all costs" mentality? You like to place blame on your pm's, tester's, your management team, your ceo, etc. If this is what makes you happy in your career, stick with Microsoft.
Like I said, for me, I wanted something different. I am just a computer scientist who loves to create, code, and deploy. I am not a political beast. I hate stack ranking. I hate screwing over my comrades. I hate fighting over which 1000 man army should own this feature or that.
I went to Google because it offered an environment where I could excel at what I enjoy. If/When Google can no longer offer that environment, I will leave and move on to the next thing. Until then, I am happy and productive, and don't have to deal with 90% of the bullshit you guys seem to love dealing with in your day to day lives.
Note, I am not saying Google is right for you. My guess is that it IS NOT right for you. You need to decide what it is about the software industry that excites you. Be honest about this decision, and then go find a match.
Microsoft might be the right place, but who knows, you might find more of what you are looking for at IBM, at Borland, at BEA, at Zillow, at Adobe, at SAP, at Oracle, at MySpace, at Yahoo, at Nokia, etc.
You are not a slave to Microsoft. All you are is a soldier in their war. If you die on the battlefield, steve and bill don't really give a shit. You are helping them achieve their goals and they understand that in a war, some of their soldiers will die along the way. Don't beleive me? Go ahead and ask steveb this exact question. I know his answer, I know how he thinks. You are replaceable and he won't shed a tear when you die young, burn out early, when your family leaves you, etc.
You think this blog is going to change anything? You think you have the answers? You think you can help Microsoft "win its war"? You are clueless. Like this post says, you are a "Cog". You are equivalent to the foot soldier in Iraq. You don't think those boys over there have ideas, bitch about strategey, etc? You think any one of them, or even a large number of them have ANY power to change bush's mind, alter in a material way the course of the war? Nope. They too are "Cogs". They are fighting the wars choosen by their generals just like you are fighting the wars of yours.
Protect your ass because a Microsoft kick won't even leave you (Google) constipated.
LOL! You tell 'em! Everyone knows that well over 90% of Google's revenues comes from AdSense. Go to http://www.businessweek.com/technology/content/feb2006/tc20060227_930506.htm and see how click fraud will eventually finish Google.
Your stock is already tanking. This will be your year of purgatory. Be afraid Google, be VERY afraid!
>No, you! No, you! ... Any questions?
Just one: if you're not aware of how common finger pointing between disciplines is to the software industry, why should we believe in the validity of anything you have to say? I really don't think it's any worse here than anywhere else.
Try to come up with a more intelligent response than "No, you!" this time.
"iTune+iPod is a monopolistic abuse.
Please don't start this argument again."
Mini, I respect your wish for not writing more on this topic here. But this is really a part of the problem you are trying to solve here now. All the bitching against management, and against each other and need for smaller Microsoft is because stock price is not growing. Well, one reason is the chain of law suites against Microsoft including worldwide anti-trust punsihments. My argument is that Microsoft did
not do anything which other companies are not doing. These cases are just a way to loot Microsoft's hard earned money. Since Apple behavior is much worse in everything Microsoft is blamed for, so Apple is a good example to understand the reason behind this stock price glut.
No, you! No, you! ... Any questions?
Just one: if you're not aware of how common finger pointing between disciplines is to the software industry, why should we believe in the validity of anything you have to say? I really don't think it's any worse here than anywhere else.
Try to come up with a more intelligent response than "No, you!" this time.
Hmmm. I am not really sure how to respond to this one. I am now running my 4th startup, if you need my street creds. I left Microsoft because the only solution I could come up with was to split the company into four pieces (servers, operating systems, office/exchange and 'other' if you must know), use the process to dramatically thin the ranks in each, bring in professional management at the top of each and retrain from the top down. I knew this wasn't going to happen, and I left. It is as simple as that. I was there as a result of an acquisition, I stayed because the economy went to hell, I left when I figured there was no way to make even the slightest difference. Along the way I had several shouting matches with Bill and Steve and left each shaking my head at the amazing intelligence and thick-headedness wrapped in a single package. Bill and Steve are the same in that regard.
I am more than aware of the constant finger pointing in the industry. I just don't accept it as a necessary result of being successful. It is allowed at Microsoft, encouraged perhaps, and now it is what it is.
It will end up killing the company and there is nothing anyone can do about it, other than my solution above.
The fact that my bringing the finger pointing to your attention and getting an emotional reaction means I am pretty close to the truth.
Everyone at Microsoft falls into one of threee camps:
1. Knows what I say is true and is just going to ride it out as long as possible.
2. Knows what I say is true and is leaving at the first opportunity.
3. Is a Microsoft inbred (said in the most respectful way) and is in complete denial.
Which one are you?
Along the way I had several shouting matches with Bill and Steve and left each shaking my head at the amazing intelligence and thick-headedness wrapped in a single package. Bill and Steve are the same in that regard.
Joe,
How the hell did you manage to keep your job after having 'several' shouting matches with Bill and Steve! Quite curious since I'm a SDET almost completing my first year here at MS!
"Does *ANYONE* think that WTT is anything other than complete crap?
I know devs who were asked to give a few hours to help maintain it and ran away screaming.
I remember one time someone showed me a performance problem in one of the SQL queries. The damn query took up half of a printed page and was composed of a couple dozen joins.
WTF???
Worse yet, it gives a very false impression of quality. It gives pretty numbers for upper management that don't map 1:1 to quality.
It is yet another example of something that would have died a quick death if it wasn't forced on testers by management, and it is just another reason why I left MS last year."
Let me say this: I feel for you, but you are not alone. at my company we have developed similar tools (CHECKMAN anyone???), and similar bureaucratic processes (here, known as a process named PIL).
Some of the readers might guess which company I work for :-)
The effect on developer morale is the same as at Microsoft. Seems to be a natural law once a company reaches a certain size: suits and bureaucrats take over, sapping the soul out of the grunts who do the real work.
If your name is Bill and you're trying to rule the world of technology, you can't have a company full of people satisfied with what they have.
Um... except it gets harder and harder to find little mini-Bills.
And, filling a company with hyper-aggressive types tends to work against you when the attention turns inwards, as is happening right now.
Post a Comment