Tuesday, May 29, 2007

End of May Conversations

Checking in on some recent conversations here... break-out the Welcome Mat for AQNT: okay, Mini-Microsoft is probably not your first preferred Microsoft-focused blog to read:

Thanks for making us feel so welcome!

(An aQuantive employee)

You could drop by MSFTExtremeMakeover and, ah, see if that helps. And helping to provide some perspective comes this comment:

To the poor aQuantive employee feeling unwelcome after reading posts here:

Don't take it too seriously. This is a blog devoted to making Microsoft smaller; we'd complain about an acquisition of 2,000 employees even if they were all super-geniuses carrying big carts of gold bars. People commenting here aren't going to be too happy about it.

Microsoft is so big it varies a lot from team to team. Personally, I've found that joining Microsoft's online services division brought me a substantial salary increase (15%, though my previous employer was admittedly underpaying and bleeding employees like crazy as a result) and a pleasant working environment where performance is expected but the rampant workaholism that plagued Microsoft in years past is not. It's not all bad by any means.

This said, it's not all good, either. The stock's not a good investment (and may never be), the review process is schizophrenic (we can't decide if we're a company of stable, career engineers or genius workaholic alpha males, so we end up pleasing neither), and the fastest way to move up the salary scale is still to quit for a year and come back at a higher level.

I'll go on record with stating I'll welcome anyone coming to Microsoft with big carts of gold bars and that I'll hug 'em long-and-hard right after they drop off my bar. So let's turn to our future co-workers with open arms and --

I'm not looking forward to having to work with Microsoft employees. From what I've seen, all you do is bitch about your jobs.

We ship software, alot of it, and hardly ever have to work more than 40 hour weeks. And our products are highly reliable and actually work.

I don't look forward to having a bunch of Microsofties working on our products, bitching about our processes, because thats all you guys seem to do, I'd put one of our offshore developers against 10 of yours, attitude is everything. Your egos are sickening.

:

What makes our company great is teamwork, and positive attitude, and Microsofties, all I see here is attitude, the bad kind. Its just like in life, if you think negatively, you get what you think, and you think your company is so aweful to work for and broken, guess what, IE7 is broken, office is broken, your OS is unreliable.

One thing good about the aquisition, it will be one if the few pieces of microsoft software that actually works.

Shh-snap!

InsideMS impending restructure: a number of comments about the change around the internal Microsoft discussion blog:

There has been lots and lots of constructive, on-point, substantive commenting on that blog (though why people bother is a good question) and she has systematically ignored just about every bit of it. No mention of people's honest and realistic questions of her when she's posted new entries. No mention of the influence of people's suggestions or concerns when she rolled out MyMicrosoft 2.0 (uh...yeah, well 1.01 or whatever you want to call it).

Now she complains that it's just degenerated into a mess. Well, honey, if you're not going to pay attention to the kids or listen to their concerns after setting them up to expect that, then you get what you get. Take some of the responsibility.

First of all, I hope beyond hope that the recent myMicrosoft post (with comments) never slips outside of Microsoft... it's shaken my confidence in my comfy wardrobe. Second, I expect there will be a common sense re-org for InsideMS. More interaction with the posters, certainly, if the conversation is valued. Comment moderation, perhaps? I think even the threat of moderation would help. Third: I've written about the need for sophisticated comment handling in blog infrastructure and this internal comment flame-out is a classic example of how entropy sets in within a linear, unmoderated, untended world.

I can't imagine the top priority for any blog technology provider beyond innovating in community managed comments (e.g., something like what Slashdot uses). Personally, I would also prefer the case where there's a feed for the most recommended comments for those comment jewels that come in that surpass the original blog post contents. That happens here quite often.

Re-org Intrigue: some speculation regarding what the recent re-org might mean:

Interesting re-org in putting STB under Jeff Raikes. Does anyone else think this looks like a pre-cursor to a break up? I mean, now we can cleanly carve off Windows into its own company, MBD into another and then whatver we do with Live and Entertainment. Honestly, that would be cool. I want the company broken up. I'm in Raikes' world and I'd love that group to be decoupled and given some more freedom to innovate.

As for MSFT insider trading this comment notes:

Robbie Bach dumping a s--t load of shares on the market today...what a shocker!!!!

Everyone pales in comparison to Mr. Gates, but it looks like Mr. Bach and Ms. Brummel are in hot competition for #2.

New Microsoft is Just Fine for Me: here are a couple of positive comments around compensation at Microsoft and the strategy going forward:

Once you factor in the few grand from stock and occasional bonus, even measly ones, the salaries at msft are pretty damned good in the Puget Sound area. I was lucky and bought my house in 1999 when I moved here, so I have zero complaints about the numbers. Coupled with working on software that is run by hundreds of millions of users, I really don't see what all the complaining is about.

New Microsoft:

The new Microsoft is willing to enter new markets, very quickly, and do quick iterations or quick acquisitions. The journey and the learning is important - not duplication of someone's existing product. This generates the insight which enables building something people will want to have versus the old way of rebuilding something they already have. XBOX, ZUNE, Windows Live, MSN, and its incubations (under Ray Ozzie & Gary Flake) have all internalized this. The leadership is willing to quickly make the changes needed to ship current versions and add the right functionality to new versions of projects.

Rank-my-Yank: And here we are on the verge of the major review season. And by season I mean the entire summer, 'cos it sure seems to take that long. Do you know when your stack-rank aka rank-and-yank aka calibration meeting is? Somewhere along the line, we'll figure out that having co-workers compete against one-another for their differentiated compensation lessens our teams, our innovation, and our ability to compete externally (how can we compete in the market place while we're competing internally?). This comment from Simon G notes a study:

If rank-yank is still a hot button with you, here's something for Stanford about it "Business School professor Jeffrey Pfeffer ... faults businesses for spending too much time in rank-and-yank mode, grading and evaluating people instead of developing their skills."

http://www.stanfordalumni.org/news/magazine/2007/marapr/features/dweck.html

I'd expect it to be one of those revelations that changes Microsoft completely for the better. In the meantime, let the ankle kicking begin...


157 comments:

Anonymous said...

Hey Mini,

Can you tell us how many aQNT employees decided not to join MS ranks just out of pride and dignity?

Anonymous said...

"Can you tell us how many aQNT employees decided not to join MS ranks just out of pride and dignity?"

One sour aQNT doesn't sour the whole bunch!

Anonymous said...

"Hey Mini,

Can you tell us how many aQNT employees decided not to join MS ranks just out of pride and dignity?"

I'm not Mini, and no, I can't tell you. But if aQNT employees had options, they just hit the jackpot of their lives. They'll probably all stay long enough for the aQNT options to vest - that's some pretty expensive "pride and dignity" if they don't. But once they vest, they have enough money to have their freedom, and you might see a mass exodus.

Note well that I don't know if all aQNT employees had options or stock, if so how much, or what their vesting schedule is. Just speculating.

MSS

James Jeffers said...

Thanks for the welcome. Not all of us here at AQNT think that Microsoft employees are shiftless, crabby people with bad attitudes. I just hope that we can find a way to work together to make oodles of hot, sweaty cash.

Anonymous said...

From an Atlas Employee (the part of aQuantive that Microsoft wanted, but you had to buy the entire thing to get us.)

Here's a daydream I just had - anyone that cares about our products and have buttons to proove it.

Once the aquisition is final, we'll need to face the challenge of amplifying the power of our Atlas product suite, which will mean adding more developers and teams.

Choice one: do things the microsoft way - add a mountain of warm bodies that have one thing in common: they can solve some obscure bitwise whiteboarding problem with ease, this will indicate that they are perfect fits for any difficult problem they might face in engineering. (tongue in cheek, slap in the face MSoft)

choice two: do things my way. get the usual candidates found by the normal means, then ask in essay form or on team fit what people think about agile, why it can work and why not. If that seems like they have a good attitude and know about the methodology, Then, have them spend an entire Day on a pairing exercise, on a real story card, on the real team, and watch them squirm, or watch them succeed.

Whats success? Do all of these things for a simple customer story, one that is estimated at a day or less. (story cards are only between 1 to 3 days)

1. understand the customer's need
2. write tests that verify what the customer needs
3. write code that gives thhe customer what they need
4. demo to the customer thhat they got what they need
5. deploy that to integration automatically so its tested, and if that works, its ready to ship.

So a few Microsofties would pass my interview process. The ones that didn't, I would move them to a product whos goal was Not to ship any features, to go nowhere, because they need jobs too, and something odd and beautiful might magically pop out of such a group. the IE7 team would be a good place.

The ones that made it, no matter what degrees they have, and no matter how many obscure algorithms they might know, no matter how many years experience they have, would be worth their weight in gold, becausee they would be able to do what we needs to happen to be # in the advertisement world that will change and the software will need to change in real time to keep up with it...

If Microsoft augments our ability to do 1-5 above, rather than dismantle it, then we'll have the momentum to generate 60 billion in five years time, and fundamentally drive how the advertising market works, by letting customers drive it, they are the ones that know where it needs to go.

But in reality, Atlas will end up with countless ego maniac, and brilliant developers who can perform obscure and irrelevant Bitwise math algorithms, can't tie their own shoes, or solve a real engineering problem, and this deal, is just bad.

this will be either
1. Polution and death of something great
2. augmenting something that works for the customer

Anonymous said...

Somewhere along the line, we'll figure out that having co-workers compete against one-another for their differentiated compensation lessens our teams, our innovation, and our ability to compete externally

Please let me know if this happens. I may be interested in coming back if MSFT ever figures this out.

Ihar Filipau said...

fyi

http://www.engadget.com/2007/05/30/steve-ballmer-live-from-d-2007/

Anonymous said...

Robbie Bach sold even more today...prick.

Anonymous said...

I would guess not one aQNT emp will bail. They may talk the talk but when it comes down to it they will come and find a comfy chair and after 1 year will be here discussing the same damn issues we have been discussing for years. MS will destroy them and turn them into robots. 9-5 is the life and times.

Grab a cold one, sit back and enjoy the ride.

Anonymous said...

Could we cut down on the banter from aQNT "employees". If I've ever seen a "holier than thou" attitude - that's the kind of stuff these folks project. They obviously don't know much about MSFT or the kind of people that work here, and their perception of it is mostly influenced by Slashdot it seems. I can tell you right now that with attitude like this they won't be very successful collaborating with the rest of Microsoft. In fact, I'm flipping the bozo bit on them already.

Anonymous said...

"choice two: do things my way. get the usual candidates found by the normal means, then ask in essay form or on team fit what people think about agile, why it can work and why not. If that seems like they have a good attitude and know about the methodology, Then, have them spend an entire Day on a pairing exercise, on a real story card, on the real team, and watch them squirm, or watch them succeed."

interestingly enough, one of my good friends -- a lifelong microsoft hater, btw -- quit atlas a while ago because he considered the devs to be just average and slavishly devoted to agile dogma without using their common sense to adapt it to meet their needs. i believe there was also a comment about being bored out of his skull because his coworkers were lifeless suburban gasbags, but i digress... ;-)

you'll fit in just fine at microsoft, methinks.

Anonymous said...

choice two: do things my way. get the usual candidates found by the normal means, then ask in essay form or on team fit what people think about agile, why it can work and why not. If that seems like they have a good attitude and know about the methodology, Then, have them spend an entire Day on a pairing exercise, on a real story card, on the real team, and watch them squirm, or watch them succeed.

I've got only one thing to say: Mister, sign me up!

Anonymous said...

>>do things my way. get the usual candidates found by the normal means, then ask in essay form or on team fit what people think about agile, why it can work and why not. If that seems like they have a good attitude and know about the methodology, Then, have them spend an entire Day on a pairing exercise, on a real story card, on the real team, and watch them squirm, or watch them succeed.

7 MS interview loops now: 1 when hired and 6 internal - got 4, screwed up 3. Been here... a while. Unfortunately, I've only seen anything remotely like the above in an interview once.

Regardless of what the ACTUAL job would be, if you haven't committed to memory the same algorithms your intereviewers learned in college, you're out.

The suggestion that you make - have the interviewee do the REAL job for a day (as closely as is practicable) makes far more sense.

The problem with the current nonsense is that 98% of your capabilities are never explored - unless your prospective group spends all day doing obscure things to/with linked lists or figuring out the shortest path between 10 cities. Maybe some do. Not that algorithms aren't important - they absolutely are - but determining the requirements for selecting the appropriate engineering approach is far more important than committing it's implementation to memory. You see, we have those things called books, and Wikipedia for that.

I hpe some of the above culture rubs off - but I won't hold my breath.

Oh, never mind me - I just suck.

Anonymous said...

Good for the aQNT folks. Take the money, enjoy the ride for a while at softie-land, until you find something better to do. No doubt that there will be many a bitter msfties, icily glaring at your good fortune. It really sucks being the new ugly, red-headed stepchild, now. But of course, I don't work at MS anymore.;)

Anonymous said...

The sad truth is that the view shared by the sour aQNT employee is very similar to the view that the MSFT Field has of Redmond. On any day you could walk past the water cooler in any field office and hear that view being shared but with much more specifics and probably a little less filter. And this is nothing compared to what our customers think of Redmond. I say welcome to all the aQNT folks, answer your emails, care about the customers, ship something close to the ship date (or year at MSFT), be sure to take advantage of the towels and lunch delivery if you make it to campus, and Good Luck!

Jon Pincus said...

Welcome to our colleagues from aQuantive ... I hope that it's choice #2, or at least something in the range of 1.8-1.9, and that your attitudes and practices help affect the rest of Microsoft as well. Microsoft acquired my (much smaller) startup almost 8 years ago, and we wound up hooking up with a bunch of kindred spirits at Microsoft and continuing to do great stuff.

In terms of the InsideMS blog, well, I just got anonymously trashed there both for what I've worked on professionally and for what I wear ... so yeah, I hope those threads don't leaked too.

One other thing I said there that it's really disappointing to me to see people saying "abandon it". While it's far from perfect, I think there's a lot of value in what's happened so far; the right response is to learn from it and iterate -- not give up.

For those of you who are at Microsoft, we're having a discussion about it here. Please join in!

For those of you who aren't at Microsoft, are there any well-known examples of successful internal blogs run by VPs of HR we should be looking at as a model?

Thanks,

jon

Anonymous said...

Hey, There has been no mention of Ray Ozzie in a while. Does anyone have any idea what the "Great White Knight" is up too?

It's 18months or so since his big ra-ra memo but he still hasn't done anything but talk yet.

Was the PDC his gig? Did we pull the plug on that because Ray could not meet that date? That would have been ~2 years after his memo.

How can MS/Ray be taken seriously on services and agility when Ray can't do anything with unlimited funds and 2+ years of dev time.

Maybe he is just a washed up has-been and should stick with enterprise server software?

Anonymous said...

In terms of the InsideMS blog, well, I just got anonymously trashed there both for what I've worked on professionally and for what I wear


Get over yourself. The complaints about the product you work on are long standing and nothing that hasn't been raised before.

As for the comments about what you wear, if you'd just wear whatever you want, nobody gives a crap. But when you start going on and on justifying it and spew TMI about your gender identity, then people get put off real quick.

Anonymous said...

Former MS partner Sobeski is a SVP at Yahoo, see http://www.linkedin.com/pub/3/200/536

Anonymous said...

:: okay, Mini-Microsoft is probably not your first preferred Microsoft-focused blog to read:

not sure how many other people feel this way, but this is The most accurate blog to read. To find out about Microsoft technology, there's MSDN, its highly accurate. For anything that isn't software, like anything where emotions might appear (everywhere outside the bits), where else would we go? A microsoft site is propoganda, telling us what they would like to be not who they are.

I agree with the Microsoft selection process - I was famous in my 20's for my work in Search and discovery and AI, and shippped alot of products. I got to Microsoft to interview and they give me questions which I answered fairly easily like balancing a BST, whats a Red black tree, etc. but what the heck does that have to do with search? We don't write our own data structures and search algorithms anymore? how about a question like 'how do you index a billion documents'? And no one had the time to actually visit some web sites and use my search tools, I had configurable relevancy and configurable AI, but Microsofties didn't have the time to even look at that.

I'm a proud Atlas employee, and why I chose to work here was what I had heard about their development methodology. It was always my dream to work with other people, in a paired fashion, working through problems. To test drive a problem, and then be able to do massive refactorings without breaking anything! to swap mock contexts with the data driven contexts and then stress test the hell out of the software.

Althought the above poster is talking a bit big, we do have some of the dreams of XP shaking out, but a long way to go. As with anything, getting XP and Agile to work in a Legacy application full of COM+, ASP, C++, VB, and then .net when you're lucky, is pretty darn tough! Its like running in a swimming pool! But the sheer power you get when you get out of the deep end, how rapidly and reliably you can ship software without dependancies....

There are some areas where I don't feel agile and xp can work, in legacy applications, and with huge dependancies, but we've had to make them work.

I love the Mini Microsoft site! its what I'm all about, smaller team, shipping Often, customer focused, and profit making because the customers are getting what they want. I hope they don't try to fix something that's not broke!

Paul - Atlas Rich Media

Anonymous said...

"In terms of the InsideMS blog, well, I just got anonymously trashed there both for what I've worked on professionally and for what I wear ... so yeah, I hope those threads don't leaked too."

Yes, folks are liable to think MS is a dysfunctional mess internally. Oh right, they already do. Nevermind.

Anonymous said...

"Robbie Bach sold even more today..."

Look at Chase: 240,667 shares today. That's half his holdings. But everyone else should be confident in MSFT's future...

http://secfilings.nasdaq.com/filingFrameset.asp?FileName=0001193099%2D07%2D000024%2Etxt&FilePath=%5C2007%5C05%5C30%5C&CoName=MICROSOFT+CORP&FormType=4&RcvdDate=5%2F30%2F2007&pdf=

Anonymous said...

"Hey, There has been no mention of Ray Ozzie in a while. Does anyone have any idea what the "Great White Knight" is up too?

It's 18months or so since his big ra-ra memo but he still hasn't done anything but talk yet.

Was the PDC his gig? Did we pull the plug on that because Ray could not meet that date? That would have been ~2 years after his memo.

How can MS/Ray be taken seriously on services and agility when Ray can't do anything with unlimited funds and 2+ years of dev time.

Maybe he is just a washed up has-been and should stick with enterprise server software?"


hey, i just heard about this thing called "google" and it helps you find information!

http://rayozzie.spaces.live.com/

Bernie said...

Hi Mini,

It was 12 years ago this week that Bill put out the "Internet Tidal Wave" memo that turned the company away from a proprietary MSN and towards the web we know today.

Something as powerful is needed from Ray or Bill now, to try to turn the company to face the platform API challenge that's coming.

The Google Gears announcement today will likely prove to be a major incursion from the cross-platform web app world into the Windows-centric desktop app world.

There's still time for Microsoft to respond, but Silverlight isn't yet enough. There's virtually no way to stay in the game without embracing open source as a key strategic tool to use where required.

Is Ray up to delivering a game-changing message like this?

Anonymous said...

>> Paul - Atlas Rich Media

Your words would carry much more weight if people here knew what the heck is it that you guys actually "ship". I went to your web site and couldn't find anything.

Anonymous said...

hey, i just heard about this thing called "google" and it helps you find information!

http://rayozzie.spaces.live.com/


Did you read the comment? Ray's last post in this blog was April 1, 2006!

The memo I was talking about is is infamous Service Disruption Memo published 10/28/2005:



So my question, remains unchanged... WTF has he been doing? Are we betting the company on a guy that can't code/lead any more? A washed up has been?

Anonymous said...

Atlas ships software that isn't off the shelf like ad sense. Its highly tailored to a very complicated advertiser market, with features that very few understand, but those few spend the big bucks in the advertising world.

One of my dreams is to have more turnkey offerings, wish I could talk about that ;) then we'd need a better and clearer website. I think it tells the perspective customer everything they need to know, and is up there with our competitors. Googleclick has a better site in some areas I believe. I don't like the mohawk guy on the rich media area, no, we don't have mohawks.

I just went to our website and I found tons of information on our products from a publisher and for advertiser... all of our products in enough detail. what more are you interested in learning about? There's a lot there, I only know about rich media in depth.

All the products listed on the website have releases every 2 months or less, we ship all the time. Its one of the most incredible things I've ever experienced and it keeps getting faster and faster and dependancies shrink over time. Alot of the coolest features on our products can be learned about through the atlas institute, we make money because of the numbers we produce.



Paul -

Anonymous said...

not sure how many other people feel this way, but this is The most accurate blog to read.

Total BS. This is largely just a place for disgruntled MS employees to hang out, myself included. It never has reflected the opinions of the MS population at large.

Anonymous said...

I wouldn't link to Ray's blog to support the argument that he's been working on something--he hasn't updated that spaces account in 14 months...

neo man said...

"If Microsoft augments our ability to do 1-5 above, rather than dismantle it, then we'll have the momentum to generate 60 billion in five years time, and fundamentally drive how the advertising market works, by letting customers drive it, they are the ones that know where it needs to go."

This statement shows that the aQNT that wrote doesn't have a clue of what he is talking about. It is basically claiming a compounded grouth of 250% for 5 years! It is also claiming that all other companies in the business area are basically put out of business.

I would deeply discount everything else this guy says because of the extremely poor judgement show in that statement.

Anonymous said...

>> Paul

So you just write software to show ads on the web? Really? What do you need all those "star" developers for?

It's not like you guys are dealing with the same level of complexity as COSD, or SQL, or CLR, or Office, or even .NET Framework or Live Search.

Anonymous said...

:: I would deeply discount everything else this guy says because of the extremely poor judgement show in that statement.

Exaggerated numbers but on the right track, I feel that's what we need to do at microsoft - deliver much faster, the right thing at the right time, something people used to think water fall did! Do you understand how different software development is in comparisson to the manufacturing industry? Do you have any idea that by learning things like Lean, and focusing on time to market, Microsoft can increase profit exponentially?

In 5 years digital advertising will be 100 to 200 billion anually, Microsoft should have a majority of that if it can produce products constantly that follow that growth tightly, and with this aquisition.

60 billion in 5 years is outlandish, but I'm curious why people question Microsofts ability to take a good portion of the 40 billion dollar pie today, which will be much larger in 5 years.

What do people think?

Anonymous said...

Just like Microsoft is to the market, SVP of Human Resourcing Lisa Brummel is to Mini Microsoft. Seems that they're always behind the curve and chasing the last idea to seem relevant.

Mini, confirm for us if you can what some acquaintances within the Window-tinted monstrosity are telling in that the internal free-for-all is now opting for moderation in its comments. Something you opted to perform here at one point. Is by nature Ms Brummel trying with no success to catch the wave, or is this an attempt to protect a thin fragile skin on a less-than-thin-and-fragile person?

The quote sent to me was particularly humorous, and similar to a line in a previous post about rabbits. The precise quotes I received are

"Looks like it's bunnies, kittens, and rainbows from now on. The lunatics are in full control of the Asylum."

"We need an uprising to storm the Bastille. Let's hope our July 14th is just around the corner...."

Anonymous said...

mSoftie Ben Smith made a semi-cross post comment between Pogo and Mini - but since it's over on an old thread, it might get lost or ignored.

Comment:
http://minimsft.blogspot.com/2007/05/ack-as-in-ackquisition.html#comment-2054806330545968919

Anonymous said...

::So you just write software to show ads on the web? Really? What do you need all those "star" developers for?

do people think that we have 2000 developers? Atlas only has 200, is that alot?

our software is no where as near as complex as the ones you listed! But serving 100,000 ads + per second, targeting them to the right person without outages, that's pretty complex.

Not to knock live search, but ad relevancy will be much more complicated, profitable, and performance centric than web search. If we had 2000 developers, and most of those star developers from places like the SQL server team, CLR, search, etc.,

My head would just pop. I've been working on some prototypes and have had some AI dreams since day one at Atlas, but we don't have the personel to come out with some truly sick software that would turn the industry on its head - no one in the industry has even begun to scratch the surface of what well implemeted AI can do in this field.

We'll have to serve a million ads per second, targeted, and then if we start adding sick features on top of that, we'll need a heck of alot of star developers. Thats why I was so excited to hear about the aquisition, because most of my dreams could come true.

Ad serving software will be more complex than the stuff you listed in a matter of years. Today, no where close. But this is the field where AI is actually going to be useful, and with huge amounts of ads per second.

I don't think I understood your question, and I think you believe that we have 2000 developers! Not yet. But I'm confident that Microsoft is going to shift over alot of stars, we'll be needing them to build the future of making money on the web for Microsoft. Unless search will actually make money...? what else is there, web based?

Nice, I'm getting lots of flames, I'm Buddhist, so bring it on! burn me alive, I live for it

Paul - Software Dev at Atlas

Anonymous said...

"It's not like you guys are dealing with the same level of complexity as COSD, or SQL, or CLR, or Office, or even .NET Framework or Live Search."

Hmm ... Paul touches on something that's supposed to be at the core of Microsoft's business -- customer focus. Yet some Microsoft clown decides he/she has time to remind Paul that the Windows operating system, Office, SQL, .NET, and Live Search is more complex?

Who cares? This Microsoft clown completely misses the point. Instead of taking 6 billion dollars and dividing it by the total 80,000 Microsoft clowns, and then handing out each of them "super bonuses", they decided to throw this money at a company that has figured out how to do something that Microsoft has never been able to figure out how to do: generate scads of advertising revenue on the Internet. And to top it all off, aQuantive has figured out how to do it without losing billions in the process (nice job MSN!).

It matters not whether aQuantive uses SQL or Oracle or something else to store its data. It matters not whether they use the .NET Framework or whether they even know it exists. What matters is that it's the company that just got paid $6 billion dollars, which is about 3 times the annual revenue generated by SQL and probably 1000 times anything generated by dot freakin' net.

aQuantive 1, Microsoft 0

Anonymous said...

Hey Mini, new topic. I'd like to hear your take on this:
Microsoft vs. TestDriven.NET

Here's the response from Dan Fernandez of the Visual Studio Express team... followed by page upon page of developers basically kicking the snot out of the bully on the block.

This is good for Microsoft how, exactly?

Anonymous said...

"I've been working on some prototypes and have had some AI dreams since day one at Atlas, but we don't have the personel to come out with some truly sick software that would turn the industry on its head - no one in the industry has even begun to scratch the surface of what well implemeted AI can do in this field."

I imagine this was a typo (and meant "slick" instead) but that's pretty much how I feel about ad-placement software. Sure, it's the way of the world and makes oodles of money, but it doesn't mean I have to like it.

I'll do a lot to avoid ads. Yep, even pay money.

Anonymous said...

Microsoft: Here's VS Express for free, provided that you, among other things don't write extensions for it.
Cansdale: I'm going to take it but not adhere to the license.
Lawyers: Time to face the consequences.

Pretty cut and dried if you ask me.

Anonymous said...

>> generate scads of advertising revenue

Except you forgot one thing - they don't exactly generate "scads" of revenue (much less profit) just yet. :D

As opposed to SQL, Windows, Office and DevDiv.

Anonymous said...

A former girlfriend of mine worked at WebTV and bailed just before she would have had to get a microsoft employee badge.

Anonymous said...

"There has been no mention of Ray Ozzie in a while. "

Ok, here's a little tip: Ozzie is a has-been. Nobody outside of microsoft and the diminishing pool of non-technical corporate IT managers who still swallow your PR bullshit think he matters in any way.

Some interesting things came out of the Plato project back in the day, but it's not clear just what he contributed there.

Lotus Notes is a very poor re-implementation of NetNews.

His Big Ideas about turning software into a service is nothing more than dotcom-era blather.

If MSFT needs a figurehead visionary, Why not hire Alan Kay? I hear HP cut him loose some time ago. Or how about Guy Kawasaki? He's a lightweight too, but he actually was in charge of a decent product at one point in his career.

Anonymous said...

no typo, I actually used the word 'sick'. thats kinda like saying killer app.

I don't like ads either, and I"ve never bought anything because of an ad I saw, of course, I'm not normal :)

The advertising industry is in the stone ages! Advertisers are wasting alot of money on me, on TV and on the web!

But imagine a not too distant future where people like me will actually see ads for things they Want. One reason Atlas is so big is because we have so many ads running on different publishers, based on what sites you've been to, and other factors, and you have cookies on, we know some things about you.

If there was Any option to have some identity about who I was, and when I watch video on demand, web tv, surf the web, and I had to see ads, but I saw ads for stuff I actually cared about, that would help everyone out.

Mindless advertising works for TV and for the web. Just blast a message. But we're going to be ushering a new era with Microsoft where product and consumer find each other much more efficiently, and often!

Double click is just ad serving. aQuantive is an ad agency and ad serving + some other stuff. With Microsoft, we'll be able to help any business put together a message and beat anyone in helping that message reach the people that want it most. Getting the right message on the ad agency side, and then sending that message on the ad serving side.

Microsoft has lots of products to help companies do business - office work, e commerce if they have a website, but now we can offer them the ability to help their product FIND their market.

I spend alot of time thinking about driving through america, looking at all the businesses and figuring out what kind of software they don't have that they need, I think Microsoft is on the right track.

And in the mean time, I'll block ads. Someday I'll be a happy consumer and products for a programmer will be there right in my face. I should start saving up for that day.

Flame away! I love it. I write 'sick' software

Paul - Software engineer @ Atlas

Anonymous said...

Microsoft: Here's VS Express for free, provided that you, among other things don't write extensions for it.
Cansdale: I'm going to take it but not adhere to the license.
Lawyers: Time to face the consequences.

Pretty cut and dried if you ask me.


So why has no one been able to post the offending license section yet? Seems like a lot of eyes must be looking at it by now.

Can you post it here?

Anonymous said...

>Can you post it here?

That's between the lawyers and Cansdale. However, somebody hazzarded a pretty good guess eleven or so comments down in Fernandez's second response.

Anonymous said...


Paul said:

Double click is just ad serving. aQuantive is an ad agency and ad serving + some other stuff. With Microsoft, we'll be able to help any business put together a message and beat anyone in helping that message reach the people that want it most. Getting the right message on the ad agency side, and then sending that message on the ad serving side.


Maybe one of the best uses that MSFT could find for AQNT would be their ad agency portion. We could dump the chumps who've been working on our recent ad campaigns (like about anything from 1997 onwards) and get something better. When I see our ads today, all I can think is that if you put 1000 chimps at 1000 typewriters, you'll get a MS ad campaign. Let's take our People Ready Business and Start the Wow now and get back to some ad campaigns that people will like and will do something for our products.

Maybe the new ad campaign could have the PC guy using the MS Robotics platform to hook some electrods to the nads of the Mac guy and give him the shock of his life...All the while, the guy from Halo would stand watching the PC's 6...

Anonymous said...

>> Can you post it here?

"...you may use the software only as expressly permitted in this agreement. In doing so, you must comply with any technical limitations in the software that only allow you to use it in certain ways... You may not work around any technical limitations in the software."

It's unfortunate that this is turning out this way. All is not as it seems but because we're Microsoft, we're automatically assumed to be the bad guys. There were some 2 1/2 years of exchanges (e-mail & conf. calls) between Jason et. al. (MSFT) and Jamie. In several of the e-mails Jamie strongly implies that making him an MVP will make all of this go away:

In our second teleconference Ben Miller told me this issue would not impact my MVP status or prospects for renewal (baring Microsoft taking legal action). I feel a constructive way forward would be if Microsoft were to make a gesture of good faith by renewing my MVP award for this season. I would then be happy to remove Express SKU integration from my website and engage with Microsoft through the VSIP program as you have suggested.

Jason tells Jamie that his participation level is too low to be considered for an MVP. Jamie's response:

I think you maybe underestimate my participation rate. I admit it has been hampered somewhat by this ongoing situation. Perhaps you could take into account the contributions I didn't make as well as the ones I did. ;o)
(emphasis mine)

Even with the information that Jamie has posted it seems pretty clear that he's attempting to use his violation of the license agreement as a bargaining chip to obtain MVP status or become a VSIP program partner - maybe both. I'd think it would be a relief to all of our MVPs and VSIPs that Jason did not give in to Jamie's "request."

It's also very clear that Jason worked very hard to try to get this thing resolved amicably.

Anonymous said...

Paul - Software Dev at Atlas

I'm with you! Some people don't understand that THE IDEA is much more important than implementation details.
They think "hey, my software is more complex, I'm a better developer."
For me something simple that's a gret solution is better than complexity!
Besides I appreciate your attitude, the way you replied some comments. Here, at MS, we need more people like you.

Anonymous said...

But imagine a not too distant future where people like me will actually see ads for things they Want.

Ok, Paul, so somehow your software is going to know that I need milk today (are you going to check in my frig as I surf?), that my kids need size 3 shoes (today, not yesterday and not three weeks from now), my lawn mower is out of gas, my house needs to be painted in a couple of months so I'll be in a market for a house painter (with local recommendations, of course), etc.?

No, somehow I don't think that's the kind of "targetted" advertising you mean. You mean that you want to target ads for TOYS to me, not ads for things that I actually need to purchase. And I don't want to be targetted and tantalized to buy things that I don't need and are too much for my budget anyway. If I need to buy something, I'll do my own research and find my own bargains, on my own terms. Anything else is simply someone else trying to make money at my expense, NOT in my "interest."

Good try though.

Anonymous said...

Funny thing is, Paul seems to think that Microsoft has bought the software in this case, where in reality they bought a few key folks (a small minority of whom are developers) and AQNT's customer base. There isn't anything in what they do that AdCenter can't do if properly funded and staffed.

If past acquisitions are any indication, the software will be re-done by MSFT folks within 2-3 years, and non-key devs will be gone. I've seen this happen many times.

Anonymous said...

Paul said: "But imagine a not too distant future where people like me will actually see ads for things they Want."

Somehow I can't help but see a wall ahead, coming up fast, for everybody, Google, Microsoft, Aqnt, etc. What you think I want and what I actually want are not the same.

It is about information distribution, how much and how deep. What you can do (via AI, etc.) may not be what you should do and it definitely is not something you have the right to do.

For years and years, advertising moved along, without crossing the line of privacy and now all of a sudden it's all the rage. And to be honest, there is no difference between advertising data mining and data mining by criminals or the government.

Keep in mind that the fourth amendment of the constitution does not just apply to governments, it also finds its meaning and interpretation to describe private entities and individuals. Since Google or Microsoft can now search your private effects with ease. An industry will emerge that blocks such searches if congress does not act first to curtail it. As it is already beginning to do with Google.

What you have the right to know about individuals has to find a strongly defended common line of decency that is defined by the fourth amendment. An amendment, by the way not observed in other nations or governments and is rapidly losing ground in the US as well. Yes Paul, the average citizen will come to understand the 'sick'ness that has pervaded the ad industry

Anonymous said...

>Some people don't understand that THE IDEA is much more important than implementation details.


The truly gifted among us understand that both the idea and the implementation details are important, but far be it from me to burst any PM's bubble.

Anonymous said...

Ok, Paul, so somehow your software is going to know that I need milk today (are you going to check in my frig as I surf?), that my kids need size 3 shoes (today, not yesterday and not three weeks from now), my lawn mower is out of gas

In your hurry to be sarcastic and condescending, you've exposed your complete ignorance of online advertising and Google's business intentions.

The idea is that ads are shown to you according to analysis of your previous web searches. That's where AI comes in. Google is already doing some of this. I'm not sure if Live is or not. If you've searched for things for your kids online before, Google can (should?) infer that you have children and show you products/services for kids. So it's entirely possible that you see an ad for a children's shoe store.

I hope you don't work for Microsoft...

Anonymous said...

So my question, remains unchanged... WTF has he been doing? Are we betting the company on a guy that can't code/lead any more? A washed up has been?

You'll have to check out Ray's keynote at Mix07. He says that the valuable thing he brings to Microsoft is a lot of gray hair. (I think he was alluding to wisdom but in the tech industry - as far as I've seen - gray hair is gray hair.) Best of luck though.

Anonymous said...

You mean that you want to target ads for TOYS to me, not ads for things that I actually need to purchase."

Waaah, waaah, waaah. In case you were completely oblivious to it, there's a multi-billion dollar industry that has been doing just that for longer than either they or Microsoft have been around. Why dump on the Aquantive folks?

Paul, please ignore my whiny and ignorant colleagues. Most of us wish you all every bit of luck at getting a piece of the action. Go Aquantive!

Anonymous said...

My, my - no sense of humor at all. I hope you don't work for Microsoft, but...you probably do.

My point was that Paul said this:

But imagine a not too distant future where people like me will actually see ads for things they Want.

Maybe people like him will see ads for things they want (if they aren't already, as another poster pointed out - certainly targetted ads have been around for eons). But I'm not buying (so to speak ;-) the line that advertising is ever going to pop up ads that I'm going to welcome. That's my point. Sorry you missed it in your rush to jump down my throat.

This poster said it more cleanly, but said exactly the same thing:

Somehow I can't help but see a wall ahead, coming up fast, for everybody, Google, Microsoft, Aqnt, etc. What you think I want and what I actually want are not the same.

Want to jump down his throat too? Go ahead, but it won't change the basic message. We don't like ads!!

Doesn't have much to do with whether this was a good business move or not by Microsoft. Just reflecting on Paul's premise that his work is going to lead to a product that I'll love. Not. Going. To. Happen.

Anonymous said...

If you've searched for things for your kids online before, Google can (should?) infer that you have children and show you products/services for kids. So it's entirely possible that you see an ad for a children's shoe store.

That's my whole point, Sherlock. Google can infer that, of course. But it doesn't mean that I WANT to see an ad for a shoe store whilst I'm surfing the net. I don't buy shoes for my children online (kind of hard to get the right fit).

I'm perfectly aware that this happens all the time (duh). Paul is proposing that the brilliant work he and others do is going to fine-tune this process to the point where I just can't wait for those ads to pop up. My point is that I don't want to see ads of any kind. Period. I don't care how carefully they have been crafted "just for me." So, IMO, he's wrong.

Of course, you could quibble and say that he actually said the ads would be ones that HE wanted to see, not me. But I understood his inference to be that the rest of us would feel likewise.

Anonymous said...

Hey, if you don't like ads, you should say so. And you did. But you also said:

somehow your software is going to know that I need milk today [etc.]

This is the point I was addressing, your ignorance of the goals and methods of targeted advertising.

Anonymous said...

"Hey, if you don't like ads, you should say so. And you did. But you also said:

somehow your software is going to know that I need milk today [etc.]

This is the point I was addressing, your ignorance of the goals and methods of targeted advertising."

So that's not what Paul was implying, when he said that the ad software was going to be able to specifically target just what I wanted to see? I thought he meant that the advertising would be SO targetted that it would be REALLY relevant - not just vaguely relevant ("Oh, he searched for a red car, he must want to see ads about car dealerships"). I was making the point that it was unlikely to ever really, truly be specific to what I actually want to buy. And clearly folks here agree (those that responded) because they wrote off those specific examples as "being sarcastic." I wasn't being sarcastic; I was pointing out how precise that advertising would have to be to be truly relevant to me.

And, secondarily, I don't like ads period. So it wouldn't matter how relevant they were.

Two points, one post. Hope that clears up the confusion.

Anonymous said...

And while you guys argue about ads...the destination..me, is blocking the hell out of them. Everywhere. And I teach others how to do it as well. You should see the delight in a person's eyes when I show them how easy it is. And the amazement as the Page Load Time is cut in half or more. The sigh of relief as articles become legible.

Anonymous said...

how do you block ads?

Anonymous said...

I'm with the poster that is against ads, even though I work for Atlas, and someday soon, Microsoft will be signing my checks! Can I say that I don't like present day ads? Its up to us to make load times low, ads more relevant (and less annoying, 'Snap into a slim jim'? how irritating!) and its up to the ad agencies and publishers not to piss people off! I'm ok with ads taking over the page when I'm trying to read, but someday... someday... on the aggregate we can only take so much! We'll start turning flash and silverlight off in droves, just so we can read the news like we used to.

I get super excited about problems that are left to solve, and thanks for all the people that understand this problem we're trying to solve! There's so much out there I don't know about, the world is huge. Presents to buy, places to see, movies to watch, books to read. And right now, all they give me are options for a society I turned my back on years ago, and NO I will not 'slap into a slim jim', I'm not built Ford tough I guess, and if I take your survey, I'll answer 'definitely not' and 'strongly disagree' to every single stinking question because with you corporate america, no I don't aggree, and not slightly but strongly, no matter what the question!

The future of Advertising with the power of Microsoft will show me just how big the world that I Am Interested In actually is!

At least it better, or someday I'll block ads and enjoy the fast download times ;)

Now Someone better get this targeting right, and quick. Do I have to go to Tibet just to find the things I want? lol

Paul - Atlas

Anonymous said...

how do you block ads?

The simplest method is Mike's ad-blocking hosts file, at:

http://everythingisnt.com/hosts.html

This is a replacement for (or addition to) your regular "hosts" file, located in c:\winnt\system32\drivers\etc. It causes requests made to ad servers, to be redirected to 127.0.0.1, so they fail.

It really does speed up page-loading times, especially on dial-up.

Of course if you have FireFox, there is the Adblock extension too. I use both.

Anonymous said...

Uh oh. Paul from Atlas wrote: "I get super excited about problems that are left to solve."

"Super excited"? Sheesh, he's already a member of the Borg. Or maybe he's really just some Microsoft HR person pretending to be from Atlas. Do people anywhere else than Microsoft use the phrase "super excited"?

Note to Paul: never, EVER, say "super" anything ever again. And try to avoid "stake holders" while you're at it. And don't "socialize" your ideas.

Anonymous said...

thanks for the guidance on ad-blocking. appreciate it

MSFTextrememakeover said...

"how do you block ads?"

This also works quite well:

Ad Muncher

Anonymous said...

In the spirit of improving the coming career discussions, the following article should be required reading for every manager.

Professional Suicide.

It's painful that I can think of four former colleagues who fit this profile exactly.

Anonymous said...

Big changes in Microsoft India around the corner: It is public knowledge that Neelam Dhawan, the current GM, interviewed at Cisco for the Country Manager job but did not get it. The Cisco Country head has just quit and the word is that Neelam is again trying for the job. Meanwhile, Ravi Venkatesan, Chairman of Microsoft India, is lobby hard to become the new Chairman of Nasscom now that Kiran Karnik is retiring.

Anonymous said...

how do you block ads?

The simplest method is Mike's ad-blocking hosts file, at:

http://everythingisnt.com/hosts.html


Doesn't seem to work as well (or at all on my corp machince compared to my home (non corp connected) machine hmm...

Anonymous said...

That's an excellent link on "Professional Suicide" above. Very relevant in this "season".
Here's the money paragraph I read:

A frequent cause of professional suicide is that the behavior required for survival demands unacceptable violations of a person’s values. Not being able to be totally candid with individuals on projects is one example. On projects, the rule is that team effort is expected, but in the end individuals aren’t rewarded—the team is king. If an individual works only in team efforts they will seldom receive individual recognition. This lack of individual recognition begins to erode their egos and self-esteem. Sometimes people are asked to falsify records, or to sign off on unfinished projects. These requests leave them struggling with their ethical values—their responsibility toward colleagues pulls from one side and a sense of personal honesty and integrity pulls from the other.

emphasis added...

Anonymous said...

Great professional suicide link.

Especially the part about clear goals and objectives.

I don't mean those BS line-items that you put on your review.

If you're a developer, your real goal is to do what PM and management tell you. How often is that goal clear and stable, vs. how often does this happen:

* The spec you're given is late, poorly thought out, arbitrary, ambiguous, and will obviously need revision after development has started
* Your PM has such a superficial understanding of his own feature that you have to spend days/weeks in spec reviews raising and resolving issues, i.e., doing his job too
* The spec is changed during or after development in such a way that a large amount of your work is negated, possibly (often?) in response to issues YOU brought up during the initial spec review
* The feature your PM/manager told you to spend months/years on gets cut entirely, due to politics or their lack of foresight (maybe the feature couldn't possibly make the schedule or doesn't "fit" with the product)

Of course this is going to murder your morale. But unfortunately Microsoft culture does not recognize this--quite the opposite. PMs and managers think it's perfectly natural to deliver incomplete specs, change them around during development, do a bunch of DCRs, and cut features. They do not consider these things personal failures but a testament to their flexibility. No wonder so many employees are so irate.

Anonymous said...

Note to Paul: never, EVER, say "super" anything ever again. And try to avoid "stake holders" while you're at it. And don't "socialize" your ideas.

'super' and 'socialize' are so passe. Exec-ass-kissing cognoscenti wandering the corridors of Building 34 can now be heard talking about 'communizing' ideas (linux is the opiate of the masses) and being 'hyper' excited (or hyperglycemic).

Anonymous said...

So the review process starts again. It starts in early June and and the rewards are actually received in September. Thats 4 months. It takes one third of a year to wrap up the annual review process. Wow.
I am so sure that I whatever I do between now and the review meeting will not be counted towards next year. its a psychological thing. I've been here 4 years and i've seen it happen every year.
Time to work on self-development projects and relax.

Anonymous said...

>So the review process starts again. It starts in early June and and the rewards are actually received in September. Thats 4 months. It takes one third of a year to wrap up the annual review process. Wow

We're doing things a bit differently over in Windows this year (Not Windows Live, Windows Dead):

1. "Calibration" meetings - instead of the usual spirited and eloquent horse-trading, managers will meet in a large conference room and throw darts at each other - the last one that can still see gets the pot.

2. As the retrospective (what I did part) is always totally unrelated to the previous year's commitments, we're getting rid of it. Rewards will be based solely on what you expect to have accomplished over the next 16 months.

3. Instead of doing 50 projects with too few people, we are going to devote the available resources to do a few important ones well. So we're only doing 49 projects now (plus the one that got cut as a second job for the members of the cut team).

4. Bonuses will be payable in clothing - top of the range will be thigh-length leather boots (with Windows logo), while if you are an erstwhile "limited", you get one dirty sock left over from Win95. I'm hoping for the matching sock next year.

Anonymous said...

I interviewed at Atlas before coming to MSFT. The people on the loop were nazi xp programmers. They were so excited about pair programming, stand-up 15 minute meetings, "refactoring code," etc, it was hillarious.

The best thing about that experience was meeting the skanky recruiter named **m. In fact if their offer had included weekends with the recruiter, I'd be as super excited as Paul.

Anonymous said...

"super" anything

Heh heh - I made this point in an older comment, that public-facing MS droids seem to be programmed to say super-everything.

As a counterpoint though, attended an MS(DN) event the other day. Very pleasantly surprised. First, great people presenting: sincere, funny, smart (and I didn't hear a single "super"-anything!!). Second, 50% of tech presented was stupid (= "get out into the real world") but the other half was away-blowingly excellent. Nice to see that MS still has geniuses ferreted away doing good things; I walked out already knowing that my development life was different (better) as of right then, and that I'd better stock up on sleep now what with all the 2008 products coming. Figured I'd post this here since some of you MS people seem to be so down on the world; there's goodness too.

Anonymous said...

Mini, three jewels and a rhinestone in a row: either its time for a new post or the scatter effect of bb's bouncing off Lisa's closed door is ending up here; good to see the humor getting through though.

Note, it must be pirate season again. Yup, sure enough Disney just put out another one of those cheezy Johnny Depp flicks for 12 year olds.

Jewel #1:

>4. "Bonuses will be payable in clothing - top of the range will be thigh-length leather boots (with Windows logo),"

(author comment: aaargghh! makes me think of the girl on the chain post from last year!)

Jewel #2:

>"The best thing about that experience was meeting the skanky recruiter named **m. In fact if their offer had included weekends with the recruiter, I'd be as super excited as Paul."

(author comment: :)

Jewel #3:

>"public-facing MS droids seem to be programmed to say super-everything."

(author comment: or at least LOOK like they say super everything, aka reference image: http://dondodge.typepad.com/the_next_big_thing/)

Rhinestone:

>"Exec-ass-kissing cognoscenti wandering the corridors of Building 34 . . ."

cognoscenti
A noun
1 connoisseur, cognoscenti
an expert able to appreciate a field; especially in the fine arts

A couple of new post suggestions:
Talk about WHY Microsoft DRM does not or does work.
Corrolary topic: legacy data and how Micosoft DRM locks out legitimate copyright holders from their own data over time.

Anonymous said...

> Talk about WHY Microsoft DRM does not or does work.

I'm not going to wait for Mini to make a new topic to take this one on.

Basically, there's two reasons why DRM doesn't work - any DRM, not just Microsoft's. One is social; the other is technical.

The social reason why DRM doesn't work is because users hate it. Users have the idea that if they buy something, they can do what they want with what they bought (within legal limits, of course - and sometimes users don't even care about that restriction).

For example, what's up with regional encoding of DVDs? If I buy a player, and I buy a DVD, why can't I play the DVD I bought on the player I bought? "Because the DVD you bought was intended to be sold in Asia" sounds insane to someone who just wants to watch the movie. They cared enough about watching it to pay for the DVD, and saying that they can't watch it because it doesn't match their player just makes them mad.

And it's not just DVD regional encoding - all forms of DRM make users angry, because all forms of DRM wind up blocking activities that any reasonable consumer would expect to be able to do. And until you get a form of DRM that does not block (or even hinder) any activity that an honest, but non-DRM-savvy customer expects that they can do, you're going to have a lot of angry customers.

This brings us to the second, technical, problem. A number of these angry customers are technically savvy. And, as Cory Doctorow pointed out, DRM cannot succeed when determined customers attack it.

See, it's essentially a crypto problem, and it boils down to three parts: the encrypted content, the decryption engine, and the key. So for the customer to be able to view (or listen to) the content, someone has to sell them all three pieces. But if your customer also turns into the attacker, then the attacker of your DRM system has the encrypted content, the decryption engine, and the key. You cannot stop an attacker who has all those pieces.

This does not depend at all on how the system works. It only depends on the attacker being the same person as the customer.

The only way to do it is to try to design a system that irritates customers so little that they don't bother trying to attack it. Nothing else can work.

MSS

Anonymous said...

MSS says:
"And until you get a form of DRM that does not block (or even hinder) any activity that an honest, but non-DRM-savvy customer expects that they can do, you're going to have a lot of angry customers."

So color me completely naive about this stuff, but I expect DRM is the problem my son and I have been having, trying (without any luck) to add a music track to a home video he's been editing. We tried using music he had bought (yes, paid for) from the Internet. We tried ripping music from a commercial CD that I physically own (so clearly I've paid for that). We tried using some kind of tool that claimed to allow purchased music to be used, but it did nothing.

So what's a consumer supposed to do? Put in a support call whenever they want to add music to a soundtrack? This can't be happening for ALL consumers, or the phone banks would be overwhelmed. And I know I've seen slideshows with music - and I don't think ALL of them have been made with Apple computers (certainly not the ones at work).

What to do???

Anonymous said...

Reality check time - I've interviewed lots of places, including Microsoft. I even got an offer at Microsoft (about 4 years ago) which I declined for the oft-argued-about "work/life balance". I will say the Microsoft offer was at least 10% better financially (and up to 20% if you add in stock grants) than the other offer. I just factored in not having to cross the floating bridges and it was an easy decision for me.

The guy who mentioned getting a $180k offer (up from $115k) is clearly the exception. Since accepting my current job I've periodically investigated other offerings and what I've found is that most offers, either at my level (Senior PM+) or slightly above (1st level middle-management) pay about what I make ($100-110k) but might offer slightly better benefits, or an easier commute, or a different kind of challenge. Some pay less! There just aren't that many $125k+ jobs out there (Microsoft or elsewhere). Coupled with the fact that the economy isn't creating jobs (at least in America) like it used to during times of expansion and I think it's easy to see why everyone who had such "easy" success climbing the ladder in the 90s is frustrated now.

Basically, anyone who think Microsoft pays poorly are just nuts. It might be a bad place to work for other reasons (quality of life, stupid decisions, too much political BS), and you might not make zillions like people did in the 80s or 90s, but Microsoft salaries and benefits are generally heads and shoulders above most places. The only exception I hear about these days is Google, but we'll see how long that lasts once their stock plateaus or they falter (and they will).

Everything I've said I think applies to anyone at the PM or GPM level or below. At Director or VP level, all bets are off. I have no idea how those people continue to rake in 25% more each year while the rest of us barely keep up with inflation.

Anonymous said...

MSFT 29.62 -0.67

Just can't seem to beat the trading range of 28-30.

MSS said:
>"The only way to do it is to try to design a system that irritates customers so little that they don't bother trying to attack it. Nothing else can work."

That can't happen until Ben Smith gets religion and realizes the damage legal diligence has done to Microsoft, with all the flubs, not serving your customers has to be the main reason for stock stagnation. Repentance is good. Making products for your customers is better.

Anonymous said...

Reality check time...
at my level (Senior PM+) or slightly above (1st level middle-management)...
Basically, anyone who think Microsoft pays poorly are just nuts...Everything I've said I think applies to anyone at the PM or GPM level or below


Microsoft way, way, WAY overpays PMs. If all you have is a PM skillset, you will have trouble getting a six-figure salary outside of MSFT because most other companies just don't value what you bring to the table as much as they value a Dev skillset, or a (real) management skillset.

neo man said...

"In 5 years digital advertising will be 100 to 200 billion anually ...

... I'm curious why people question Microsofts ability to take a good portion of the 40 billion dollar pie today, which will be much larger in 5 years."


If you use the estimate that Yahoo! and Google control about 50% of the online adveritizing market, you get a current market size of about $20 Billion. Assuming a 35% growth rate, that would make the market about $100B in five years. That seems very, very optomistic, but just ball parking ...

So if Mircosoft could get to 10% market share in 5 years, that would add about $10B/year in revenue and probably about $6B/year in profit. If that happens aQNT will have been pretty good buy!

Let's say the market size is only $60B in 5 years ... still Microsoft has to have less than 20% to get to the same profit level.

Okay ... it is up to aQNT folks to show there stuff! Not amazing growth for Microsoft, but good growth for a company its size. Should add another $10/share to the stock price.

Anonymous said...

RE: DRM software
I had an interesting thought on this last weekend. Seeing all the old cigarette and cigar advertisements, I thought to myself, what's the thing we're doing right now, which our children (or grandchildren) will find crazy/stupid/life-threatening (or shortening)?
Then it occurred to me--using MSFT (well, proprietary) software. If you consider the documents or work we do (or blog posts we write) as our digital "lives", then using software which is proprietary shortens our "digital" lifespans, as it limits the future read-ability of those files. (I'm sure linux zealot has used this argument before, but I haven't read it yet.) (So this is a bit of a stretch, you have to be producing work in software which saves it in a proprietary format. Wake up call to digital media producers here...)
Of course, this analogy leads one to wonder if MSFT (or other proprietary vendors) will be brought into court in the future for selling/promoting/endorsing software which is "damaging" to the consumer's "health". (See the "Digital Domesday" as a reference.)

Anonymous said...

>"That can't happen until Ben Smith gets religion and realizes the damage legal diligence has done to Microsoft. . ."

Erata: I realized I meant Brad Smith. But come to think of it, it's a dual Smith fiasco (if they generate policy re legal specification to drm and wga and MS's security policy and how drm is manifested to erase legacy data, they are both to blame.

Anonymous said...

RE: advertising $$$

I'm not in the ad business, so I could be way off, but I was under the impression that the advertising market as a whole was pretty static, if not cyclical. Some years companies in the aggregate spend more, some years they spend less, but I don't think it's ever been exponential in growth for any sustained period of time.

So, based on that, any growth in ad dollars spent on one medium (e.g. TV, press, internet) invariably come out of some the dollars that are spent in another. So for internet advertising to grow, other ad spending would have to shrink. I'm sure that is happening now (although I don't have any numbers to back it up).

My point is that internet advertising, while highly targetable and somewhat measurable (both good things for advertising), it is still just one avenue and while it should be ridden as long as possible, I can't help but think that once the numbers start to come in, that some (many?) advertisers might start to reconsider their investment in advertising venues. Likewise if some of the benefits of internet advertising begin to be found in other, more traditional media, internet advertising could take a hit.

If internet ad costs go up but return on advertising doesn't, or worse, if there's some sort of backlash and computer users revolt, then that well of money could dry up quickly. When that happens the small fry will be the first to go.

I don't know what tricks the aQNT people have up their sleeves, but I hope they are working on solving the next problem of maintaining revenue as avertising tastes shift (as they always do).

Or (nightmare scenario) they sold out to MSFT at the peak and the online ad market will go south in the near future leaving MSFT stockholders holding the $6b bag.

In that case, I hope we get to keep the aQNT president because he would have demonstrated a very astute sense of timing.

Anonymous said...

Just an FYI...
If you are leaving MS, remember you have 90 days after your last day to
exercise your stock options. Make sure that you carefully read up on the exercising stock options when leaving ms on hrweb. See if you see anything about exercising after 90 days or you lose your options. I do not recall anything like that or anything in the exit interview (as if those hr bozos really give a flying rip). I almost lost out, but this time "almost" is a great word.
Good luck, ex-msfties and enjoy
the Seattle summer! :)

Anonymous said...

Interesting to hear about calibrations in Windows. Can someone speak to calibrations in the other divisions and what us bottom dwellers can expect this year?

Anonymous said...

Why DRM won't ever work
By Jeremy Allison:
http://news.zdnet.com/2100-9588-6189011.html?tag=nl.e539

Remember in Star Trek 1 movie when they beamed those poor souls onto the enterprise and they turned into screaming mush? Its kind of like that.

Anonymous said...

Calibrations...
Wouldn't you just love to have this person reviewing your past yr?
(From MSW Letters re: test apprenticeship program)
I read Casting a Wider Net for Recruits, and as an SDET who went through training in code, CS theory, and projects, I find it sad that we are willing to let low-quality developers into test positions. Do we have a “Dev Apprentice Program” to parallel this “Test Apprentice Program”? If not, then the division between dev and test is perfectly clear to me, and I don’t want to be grouped with these amateurs who can’t code...

Nice. Real professional attitude there. One of the many reasons a good friend of mine recently left our environs after 10+ yrs. Well, there's also the fact that as a test lead, with Proj Mgmt experience she now makes a boatload more money than she ever did here. Oh, and she actually gets to do real work, rather than spend half the work-day in useless meetings.

She also reports a real world out there, where talented people can go home, or even work from home, and still get more done, and get to see their kids during daylight hours. I'm not sure I believe that. No one can leave that early without being labelled a slacker, a limited II, a non-team player, or whatever the latest OCD-driven corporate arrogance has decried them as.

Amateurs.... You mean like someone who is educated in a non-CS discipline, who comes into a CS-related field and does the job that the CS-educated hallowed ones do?

Lessee, Anyone know of a non-CS Softie who seems to be doing OK in this battlefield? OK, excluding Billg (Just a college drop-out he, never a success to be...).

HINT: Look at the person in the next office. Yup. There you are. Go find someone who's at a higher level than you and ask what their education background is. Hey, why don't you ask your boss? I'm sure he/she would love to hear that you're soooo much better than him/her because you have a CS degree...

Meh, what do I know? I only have a degree in some obscure field no one's ever heard of. It's called chemistry. No way I could ever do decent job as a tester, or test lead, or PM. Nope. Those promotions and bonuses over the last 6 yrs must be just because I'm such a nice person. I'll be jumping ship to the other company very soon. Just waiting on this yr's review. I've met all my commitments for the yr, picked up the workload of some people who've left and have very little hope for the new re-org in our group that I've gotten wind of. Looks like I'll be working for someone I'd rather not be. My spouse met this individual once and told me that if I ever turned out like that, she'd kill me. My spouse is a good shot.

Anonymous said...

Make sure that you carefully read up on the exercising stock options when leaving ms on hrweb. See if you see anything about exercising after 90 days or you lose your options.......
I almost lost out, but this time "almost" is a great word.

What about stock awards ?

Does anyone know whether you lose the stock award portions that vest 30 days after you leave Microsoft ?
(the hrweb site seemed to indicate that awards stop vesting the day you leave Microsoft)

I'm wondering whether it is worth staying a few extra weeks to get the portion of my stock award that vests in August :)

Anonymous said...

Anyone know of a non-CS Softie who seems to be doing OK

How about Dave Cutler and Anders Hejlsberg for starters.

Anonymous said...

Rhinestone:

>"Exec-ass-kissing cognoscenti wandering the corridors of Building 34 . . ."

cognoscenti
A noun
1 connoisseur, cognoscenti
an expert able to appreciate a field; especially in the fine arts


Hey, I wrote that comment. And by cognosceni, I meant folks who are experts at -- and able to appreciate the field of -- exec ass-kissing. It's not easy, given some of those SVP posteriors, trust me.

Anonymous said...

What about stock awards ?

Does anyone know whether you lose the stock award portions that vest 30 days after you leave Microsoft ?


Your stock awards are like salary. You get them on day X if you're employed on day X. If you stop being employed, you stop getting them. Similarly, if you go on unpaid leave, your salary and your stock awards are suspended (not paid) for the duration of your leave.

Anonymous said...

Original: If not, then the division between dev and test is perfectly clear to me, and I don’t want to be grouped with these amateurs who can’t code...

Comment: Nice. Real professional attitude there.

Uh, I'm with the original guy. Test !< Dev, and I manage both. A good test developer is harder to find than a good developer, and generally more valuable in the long run. There's a great yin-yang balance between the two disciplines. It's a travesty that some myopic orgs have this ingrained model that relegates test below dev.

Anonymous said...

>"Hey, I wrote that comment. And by cognosceni, I meant folks who are experts at --"

My sentiments exactly. I meant to write:
cognoscenti
A noun
1 connoisseur, cognoscenti
an expert able to appreciate a field; especially in the fine art of ass kissing.

Special Note to Exec VPs from one of your millions of PISSED OFF customers, especially Bill Gates, Steve Ballmer, Ben Smith, Brad Smith:

Five years ago when we started to tell you your OS was so insecure you could drive a Mac Truck through it, we did not intend for you to come into our homes, kill our dogs, flush our goldfish down the toilet and give everybody in the house cancer in order to secure the OS.

Start with DRM, WGA, ActiveX, Registry, yes Registry, Dr. Watson and .dmp files, and on and on and flush those down the toilet, please. While you are at it, start a dialog with Sun and Adobe on the security of Java and Flash or similar products to get them to fix that crap too.

Or to put it succinctly, please, get off your overpaid derriers and solve these problems, not extend them into infinity. And yes, you will have to abandon a lot of code.

neo man said...

Test can be far more difficult than design. I have done both outside of Microsoft. There are asspects to both jobs.

We need products from development, but it is test that that keep the customers buying.

Anonymous said...

Test !< Dev, and I manage both. A good test developer is harder to find than a good developer, and generally more valuable in the long run.

The reason why good test developers are hard to find is simple. If you're good enough to create a product, why would you spend your time testing one instead? This also explains why devs ARE better developers, in general, than test developers.

We need products from development, but it is test that that keep the customers buying.

Yes, yes, everybody's job is very important, but the fact is that Microsoft is a software company, so the people who create the bits that comprise the software are the most critical.

Anonymous said...

Yes, yes, everybody's job is very important, but the fact is that Microsoft is a software company, so the people who create the bits that comprise the software are the most critical.

This depends on your view of the software development process. When you consider that the actual act of designing and writing code takes the less time than testing and maintaining it, the distinction is not so clear that one is more critical than the other.

Anonymous said...

Of course this is going to murder your morale. But unfortunately Microsoft culture does not recognize this--quite the opposite. PMs and managers think it's perfectly natural to deliver incomplete specs, change them around during development, do a bunch of DCRs, and cut features. They do not consider these things personal failures but a testament to their flexibility. No wonder so many employees are so irate.

I interviewed at Atlas before coming to MSFT. The people on the loop were nazi xp programmers. They were so excited about pair programming, stand-up 15 minute meetings, "refactoring code," etc, it was hillarious.

Having done development the, "spec," way and the XP way, I'll take XP any day. XP is perfect for PMs that don't know what they want, and developers that do. Personally, as a developer, I like to write code. XP lets me write code all day, every day. Also, I like test-driven development; it means that I end up with a testable design and generally a better design. (BTW, this pathological split between development and test that has been expressed in some of the posts here is evil.)

Remember the tagline to XP is, "Embrace Change." Every day I improve the structure and maintainability of working code, add new test cases, and add new features or modify old ones. My code skills are kept sharp, and I don't spend months only caring about my bug count. If not all the desired features make it into the next release, they wouldn't have made it anyway, and I didn't waste a lot of time on something that would be DOA because of the schedule.

Anonymous said...

"Every day I improve the structure and maintainability of working code,"

I see, so everyday you introduce risk to the project by changing code that already works and you consider that keeping your coding skills sharp.

I tend to agree that a more agile approach is a good thing, and a software org should use bits of pieces of whatever process works for them. However, the "xp" movement has a certain group of people who breath it like a bad cult religion. So the term nazi xp programmers is right on the money.

Anonymous said...


1. Test !< Dev, and I manage both. A good test developer is harder to find than a good developer, and generally more valuable in the long run.

2. The reason why good test developers are hard to find is simple. If you're good enough to create a product, why would you spend your time testing one instead? This also explains why devs ARE better developers, in general, than test developers.


Attitudes like 2's above are what keep a lot of great people from test development. Also, attitudes like 2's above are the reason Microsoft has to have non-coders come in to do development work that happens to involve testing. I have great development skills, but I prefer test, except for all of the people that give me crap about it like you and tell me I can't develop.

A tester with equal years of experience to you will develop, debug, and design equal to you, WITHOUT having the whole company kissing their ass and telling them that their profession is oh-so-great. Leave your insecurity and your "compensating for something" at the door.

But then we all like a little recognition once in a while, and so continually companies lose Testers who are sick of the BS to dev or pm roles, which are only too happy to have them if they're any good at all. Because of people like you.

*The test apprentice program dilutes the test pool, and so the other testers will get faster promotions. That's a good enough reason to stay with what I'm good at (test) for now!

Anonymous said...

"I see, so everyday you introduce risk to the project by changing code that already works and you consider that keeping your coding skills sharp."

If you're working on an XP project that's not quite true as there is an emphasis on unit testing and regular incremental builds (that's how they manage risk). This would work well at AQNT as the projects are relatively young and have followed the process since the beginning so they should have a full suite of unit tests.

However,introducing XP into a legacy product like Windows,Office etc. at this point would be unrealistic for a variety of reasons.

Anonymous said...

A tester with equal years of experience to you will develop, debug, and design equal to you,

Anyone with dev interview loop exposure knows this is ridiculous.

First, many candidates who don't make the bar for dev are sent on a test interview loop and pass. What are the chances that employee skill/talent is the same if the hiring bar is different?

Second, many testers (usually with years and years of experience) interview for dev positions and I've only seen a small handful pass. The ones I've interviewed personally self destruct on the initial easy question which should only take 5 to 10 minutes: reverse a string, parse a hexadecimal number, etc.

Anonymous said...

>> interview for dev positions and I've only seen a small handful pass

About 60% of success on the internal loop depends on your past history. If you're a dev with a decent yearly review - getting another dev position is a piece of cake. In fact, many don't even go through full loops falling back on a little help from the former managers who are now a bit higher in the food chain.

If you're a tester, however, it's as if you had a really, really bad review. There's an immediate, strong prejudice, and even if you pass the loop, as apppropriate might decide that you're not a good match because, guess what, you're not a dev and therefore you must be stupid.

In my 6+ years at Microsoft I've seen testers who can code, and developers who shouldn't be allowed to. Try to approach cross-discipline interviews without prejudice, and don't generalize from your statistically insignificant sample.

Anonymous said...

>> there is an emphasis on unit testing and regular incremental builds

Are you a n00b or what? There's no such thing as 100% code coverage. Testing does not guarantee absence of bugs! Just because your tests work doesn't mean you're not breaking the code. It just means your test don't exercise the code paths you've broken. This also doesn't mean layers above will exercise the exact same code paths your tests exercise. Ever heard of code coverage metrics?

I mean, I'm all for refactoring things every now and then and writing unit tests, but let's not forget - they're not a replacement for common sense.

Anonymous said...

A test-dev position is much better than a dev position if you are a Core-Platform-Technology kind of a guy who would not like to deal with:

1. PMs/Management crushing a cool idea you were working on.

2. Having to justify your nifty modifications/cleanups to clear a bar.

3. Polishing-up details like internationalization, resources and other pains like that.

As a part of my job I get to implement nifty kernel instrumentation technologies in my relentless pursuit of increasing coverage of our tests and gauzing the reliability and stability of some of the key kernel components.

I find hacking the kernel, specifically MM, FAAAR more interesting and fulfilling than writing C# UI code that many of the "CS educated" SDEs do as their main job.

Sure, I have to deal with test runs but those are mostly automated. Its a routine task and not as interesting as hacking the kernel but a pain just like devs often have to deal with RIs breaking.

Anonymous said...

Not another fucking dev vs test borefest! Mini, I'm tuning out until you implement some common sense moderation.

Anonymous said...

So, since there's this huge prejudice against test -

How do I move internally from test to dev before its way too late? Obviously its already somewhat too late.

Trust me, I know that the bar is lowered for Test constantly, but I believe it wasn't lowered for me. I would like to be respected within the company. Although I think I've been a great tester, I'd rather be a great dev. Putting the disrespect for testers aside, devs get to create stuff that people use outside of candy land.

How do I move? I can parse the hex, reverse the strings...

Anonymous said...

Why DRM won't ever work


DRM should never work for Microsoft. Microsoft liked an open platform as long as it was theirs. Once bug writers started targeting Windows and Microsoft had to lock down, then they became very rights and security concious. Now, along with sniveling to the DOJ about Google's dominance, they are demanding that every industry be locked down, and, by the way, they will provide the keys to content. DRM is a panacea. Want to sell more product? Use the Internet to market and distribute more vigorously.

Anonymous said...

How do I move? I can parse the hex, reverse the strings

Do you realy need to ask how? The process is already defined, just schedule a few informationals for interesting position off the career website, then make the notification/permission to interview and just interview!

Who da'Punk said...

Mini, I'm tuning out until you implement some common sense moderation.

Sorry, too busy enjoying the nice day. I hope you are, too.

Anonymous said...

Realize that the "Test Stigma" is institutionalized within the company, and efforts to change this have been for the most part poorly thought through by TLT's architects and aademics who evidently do not "get it". I was a (mostly) IC SDET, got to 63, and aside from being a lead/manager, there were VERY few opportunities.

I became a developer. And this is a quite common move. Not only is it hard to fill new SDET slots - we continually lose the existing SDETs to development roles. If not for the institutionalized stigma, tis wouldn't happen (so much).

Levels. Take a look at desired level distribution in http://career. Start at 59, and go through to ~64/65. There is ~ 2 level disparity between the SDET/SDE peaks. Having been in stack ranking mtgs for both disciplines, this turns out to true for actual levels. In a frank discussion with one of the TLT test architects, I was told it would take MS 5 - 10 years to "fix" this.

What drove the point home about the institutionalized stigma was the "all testers must code" edict, which suggested that being a kick-ass tester is not now and never was an MS-valued skill. I was NOT a particularly kick-ass tester - but I recognized folks who were, people who could completely destroy something they'd never seen before in ten minutes. Until as a corporation this skill is truly acknowledged, expect the disparity to continue, as well as these childish dev vs. test spats.

Anonymous said...

Take a look at desired level distribution in http://career. Start at 59, and go through to ~64/65. There is ~ 2 level disparity between the SDET/SDE peaks. Having been in stack ranking mtgs for both disciplines, this turns out to true for actual levels. In a frank discussion with one of the TLT test architects, I was told it would take MS 5 - 10 years to "fix" this.

I think I said it was a guesstimate (although somewhat based in reality). I'll assume you're talking about me, but I'll stay anonymous on this public forum.
The reason there are fewer SDETs at higher levels is because people leave the test discipline (and people leave the discipline because they don't think there's a career path).
I've done more research in this area, and if dissipation numbers drop only reasonably, the distribution starts to look normal in as few as 3-4 years. I'm not going to share details on how to lessen the people leaving the discipline with this audience, so I'll just say that I'm working on it.
That's about all I'll share here - I debated posting here at all, but I think I lot of people who care about this information have quit reading Lisa's internal blog, but are still reading here.
Ping me internally if you want more information (yes, I realize that most of you don't know how to find me internally, but I know the poster of the quote above knows, as well as any others aware of the efforts internally).
One final note (for those who know me). I'm in the test discipline, and I'm *definitely* not (nor near) partner level. As long as my name is in the address book, you can safely assume that I still think that testers can reach partner level and beyond. If I ever lose hope, I won't change disciplines...I'll be gone in a heartbeat. Not sure if that's worth anything, but I felt it was worth sharing.

Anonymous said...

thought through by TLT's architects and aademics who evidently do not "get it".

BTW - please tell me what I do not "get". I am honestly working on my own career here along with the rest of the test discipline, so if there's something I'm missing in what needs to be done, I would like to hear your suggestions. Thinking about this problem has dominated my thoughts for years now, and I admit that there could be some obvious things I'm missing, or something I've forgotten to investigate.

So please, just tell me what else you think I should do.

Anonymous said...

which suggested that being a kick-ass tester is not now and never was an MS-valued skill

What in the world is going on with test at Microsoft? The test orgs have gotten so big and powerful with headcount/title/level inflation over the past few years it's crazy. Now every tester thinks he's a dev and you guys are just going off and doing your own things. I can barely get my tester on the phone because she's always busy with a test design spec review or code coverage automation run or some other nonsense. HELLO?? Who is testing the products??

To all the hot-s*** testers saying they're as good as or better at development than devs: why aren't you devs?? Microsoft sells software. We do not sell test harnesses and automation scripts. If you're such an awesome dev, help the company out and develop some products. Leave the testing to people who have test skills and want to be testers.

Anonymous said...

"Are you a n00b or what? There's no such thing as 100% code coverage... .. Ever heard of code coverage metrics?"

I don't think you read what I said.I just explained the risk management methodology of XP processes - no need to jump down my throat and I didn't say anything about 100% code coverage so that's a straw man.

Even on that point there a whole range of variables that need to be considered i.e. whether you want 100% statement,decsion or path coverage,how much you're willing to invest in QA,whether you use formal methods,what language you write the app in and what kind of app it is, number of klocs,what kind of training and experience the staff have etc.

I did say this wouldn't work in Windows or Office and I believe that's correct for the reasons above but at the same time we shouldn't discount new ideas outright.

Anonymous said...

What in the world is going on with test at Microsoft?

Not going to disagree with you, but your anger might be misplaced in most circumstances.

When a someone in test goes 8 years being told they're the most valuable on the team, but still can't get promoted because they're not writing 'test harnesses', then they either leave or start writing test harnesses.

The test orgs have gotten so big and powerful with headcount/title/level inflation over the past few years it's crazy.

I disagree, but I'll speak only to my vantage point in COSD. Level inflation always has and always will be the eminent domain of PM's. Looking around, most test people are below the level of an entry-level dev.

HELLO?? Who is testing the products??

Unfortunately... the customers.

Anonymous said...

BTW - please tell me what I do not "get".

Test as Microsoft wants it done (through automation) is a development function. Unfortunately our culture has resulted in dev orgs where few people understand test cost and even fewer therefore know what it takes to design and write code in a way that reduces test cost by making it amenable for testing through automation.

I do not see any realization of this at any level. Further, efforts to get test teams to do a QA only role by moving automation to junior devs do not have any sponsors at the VP level.

Anonymous said...

>>To all the hot-s*** testers ...

Dude, think with your head and not heart: You made some good points in the first part of your post and then it degenerated into a test vs dev rant which weakens your argument.

Sure,there's plenty wrong with test,but it's a function of the general political disfunction in the company. And yes, the level inflation is annoying but there's plenty of long term testers who aren't impressed with that either.

At the same a lot of people are working to make testing more efficient, and automation is a big part of that, so just because testers aren't manually testing doesn't mean they're doing nothing.

The core problem though is the skills attribution fallacy.At Microsoft that meant testers did manual testing with little emphasis on code analysis and little emphasis on building skills because testing was viewed as some kind of inherent skill.This approach has an impact on quality and also leads to the kind of test org. bloat we have today.

The point about Microsoft selling software and not test harnesses make no sense.Software Engineering is a process and testing/QA is part of that process so your point that testers that can program should develop new products has no validity.Test may not generate revenue directly but it determines product quality which directly impacts the bottom line.

Anonymous said...

>> I can barely get my tester on the phone

Go talk to her. That's what I do and it works really well. And don't just give orders there, also ask what you can do to help her test your feature. Give priority to testability bugs. Show respect. If your tester does good, give good feedback to his/her manager.

Testers will go out of their way to help you if you show respect and help them do their jobs. And as a dev I do know that things can get pretty harsh when you have 30 bugs on you and your tester files a new one every hour, but that's the way it is. Be thankful you have someone to point out the turds you've checked in because the alternatives are all much, much worse.

Anonymous said...

RE: Microsoft is a software company, so the people who create the bits that comprise the software are the most critical.

Just to throw another log on the fire, Microsoft is a company that SELLS software. Sales and marketing are the most critical. If nobody buys it, you'll just have the software version of Betamax: Technically superior but a market failure.

The whole dev v. test "battle" indicates the insular nature of many (thankfully, not all) developers in the company. I'd guess that many of those who are throwing out those "anyone who has any sense would choose to be a dev, if they had the right stuff, of course" have never worked anywhere but Microsoft (College internships, don't count). It also indicates, sadly, how these self-righteous devs would rather battle (and demean) the people who clean up after their holier-than-thou B.S. then work with them as a team to deliver quality products for the MOST important person: the customer (you know those people with the money).

Wake up and fight the right battle, for cryin' out loud!

Anonymous said...

RE: Dev vs Test
(And ignoring the "We all should be salesmen" BS. It's a valid point, but too simple of an answer.)
Thank you to the internal poster who is doing the research and looking into why the Test levels "plateau" at a certain point (lower than Dev and PM). SOMEBODY better be looking into this, because I'm a tester looking at the "new" titles and making some comparisons in my org. Tracking hire date in the two disciplines, guess who consistently makes "II" quicker...mmhhmm...devs.
I also find a lot of truth in the "Test harness" comment earlier. Everyone loves to re-invent the wheel, but nobody gets credit for gluing things together into a workable solution. I've been testing the same product for almost 5 years, and we've made only meager steps in more efficient automation. (Even though our goals is 99% automated tests.)
No one is looking at a product level Testability metric, and its very disappointing.
Lots more to say on this topic, but would also as for a pointer or two from MS-land for discussion groups internally on this topic. Does it come up in EE reviews, or is it just whispered over the water cooler?
A final note to aQNT testers (I'm sure there were at least one or two), make sure you're held to the same level of accountability as your dev counterparts. Any idiot can write a for loop--but it takes a smarter idiot to break it in the wrong place.
(Thats Mini for letting some of this discussion happen...if there are no internal forums, maybe a dedicated post from you to clear the air would help out?)

Anonymous said...

Unfortunately our culture has resulted in dev orgs where few people understand test cost and even fewer therefore know what it takes to design and write code in a way that reduces test cost by making it amenable for testing through automation.

I do not see any realization of this at any level. Further, efforts to get test teams to do a QA only role by moving automation to junior devs do not have any sponsors at the VP level.


Thanks - that gives me a better perspective. These are items on the radar. Technical changes and process changes are easy to see, but culture changes (even the second point would require a big culture change) are hard to measure, and take longer to have any visibility.

I'm 80% done with a paper on the the QA role that I plan to distribute and also submit as a thinkweek paper. Hopefully that will kick off some more thought on the subject.

There's a long way to go, but always feel free to keep me honest (even if an anonymous forum) if you have concerns about the efforts. Use a little discretion, however, in sharing any private conversations we have here :}

Anonymous said...

I've had developers make insulting remarks to my face about being a tester and it doesn't bother me. If they want to claim that they're the most important people at MS, then they have to accept the responsibility for the shoddy, insecure, and bug ridden state of our products. After all, they wrote the code.

To our customers: lynch our developers for Microsoft's bad products, not QA. We just clean up after them as much as we can with whaever people we can get and you can see right here how we're treated for our efforts.

Anonymous said...

> I think I said it was a guesstimate (although somewhat based in reality). I'll assume you're talking about me, but I'll stay anonymous on this public forum.

I'm the OP that sparked this response. I'll add a few things that I should have said earlier:

1. There are folks who do truly care about the business of Test - the above responder is, I believe, one of them, and the response is appreciated.

2. The "not getting" was a generalization for the TLT as a whole, based on the very poor experience suffered as a result of enduring WTT, poorly conceived reporting tools (PHI), and outsourcing to folks who care that they get their next assignment completed fast more than what that assignment actually is. Lower down, it had the feel of a successful snake oil pitch: "Give 'em a mandated automation harness, RIF the STEs, outsource the rest - think of the savings!" Of course, I'm sure that noone said those exact words, but it sure felt like it.
3. Our developers COMPLAINED that there was insufficient test coverage, cause the testers were all too busy buggering about with WTT, and writing automation. Some teams had clandestine manual passes because they didn't trust that the product was really tested by the automation they'd been forced to write in spite of their concerns. Reporting upwards, these fears seemed to get lost or ignored (too many bullet points perhaps).

Well I'm out of that world now but I still care, so in the mode of not "just whining" (although I DO love a good whine):

Push down responsibilites to Test Manager level. The product is tested howsoever makes sense locally - processes, tools. They have a contract to meet is all. Agreed upon common information percolates upwards - how this is garnered is again, up to local management.

Make processes and tools stand on their own merits. If someone writes a process or tool that makes their life easier, it gets used. If it sucks, it gets dropped like a rotting raccoon. Don't force people to hug rotting raccoons - that really stinks.

Make the test balance of methodologies fit the product under test (duh) rather than some global edict about "how it should be done". 90% automated? Why? 80% code coverage? Why? Corporate leaders may not know what the appropriate way is - because they frequently aren't on the front line with the troops that they seek to represent, but are instead, part of a remote committee.

I've ranted enough (yes, I know) but I think that this may just be a slightly different discussion than the "Test sucks!" "No Dev sucks!" style. IMO, that is goodness, as we really have NO OTHER forum like this one.

Anonymous said...

"A test-dev position is much better than a dev position if you are a Core-Platform-Technology kind of a guy who would not like to deal with:" {process and international-installed-base administrative details)

He's so right.

I'm a former Windows {technology that would too easily ID me} dev. I left that track because I was spending too much time on things unrelated to the interesting coding problems that I was accustomed to solving on a more every day basis at other companies. I found that I could explain APIs and write sample code for them, as a programmer writer, without much of the overhead that seems to go with production coding at Microsoft: having to ask five PMs if what I was doing was OK, meeting with other devs on v-teams about standards for coding that never came to anything other than lines on the review of the person who started the initiative, and playing bounce-the-bug back and forth in product studio between multiple orgs until the person whose code didn't work agreed to take responsibility for it. I've heard that it's the same way in many test groups.

Is my salary somewhat less than that of a dev at a comparable experience level? Certainly. Do I enjoy a bit more autonomy in my dev activities than the typical dev? Yes, and for that, it's worth it to me to take the salary and ego hit. I enjoy my work more and feel a more complete sense of ownership of what I produce for our users, than I did as a dev. Oh, and not only is my code replicated in who-knows-how-many third party apps (you've never stolen from sample code before?), but our customers can actually read it and critique it, so writing sample code isn't quite the "free pass" that some SDEs think it is.

For someone technical whose ultimate goal is advancement, dev is a great place to be. For someone who just likes to design and code, whose priorities involve things other than making partner, and who doesn't require constant reinforcement about how great a coder he is, some of the UA and test orgs can be great places to be, too.

Anonymous said...

Looks like the argument between test-dev and dev boils down to this:

If you *LOVE* some technology and want to work on something you are passionate about, go for a test-dev role in the area of your interest. How far you will advance in the levels is dependant on how much of a change the TLT folks can bring about.

If the [somewhat false] honor and prestige of your position is the main thing and you don't mind dealing with Process and PMs go for dev.

Anonymous said...

Review season is here again. Let us do a straw poll on the average raise for partner and non partner employee. We then see how things come out.

Group:( HR, Windows ... )
Expected Average merit raise:
Average stock grant:

Anonymous said...

Ah, the lazy daze of summer are upon us as Whoda's domain settles down into the quiet drone of low risk blogging. Others contemplate discussions that would bore the snot out of a normal person. Test-Dev, me me. Always a sure fire winner for Mini Mi.

Good Idea Mini, enjoy the rainless weather while you can, suck in the sweet smell of dry early summer air, if only for a moment.

Anonymous said...

Review season is here again. Let us do a straw poll on the average raise for partner and non partner employee. We then see how things come out.

Umm...I won't know anything about this until late August, I don't know about anyone else. We start reviews in June but numbers don't come out for almost three months.

It'd be an interesting exercise then, though.

Anonymous said...

Have you gone to http://experiment? They placed some "join the team" ads in our cafeteria. They are quite sad.

A sample from http://exp-platform.com/acmDMSig.aspx:
"Controlled experiments embody the best scientific design for establishing a causal relationship between changes and their influence on user-observable behavior." Riiiiggght.

Anonymous said...

Time for a new mini post?

Maybe you could at least talk about small things like the lunch delivery "benefit" with $1 delivery fee...

Anonymous said...

"Give 'em a mandated automation harness, RIF the STEs, outsource the rest
...
Our developers COMPLAINED that there was insufficient test coverage, cause the testers were all too busy buggering about with WTT, and writing automation."

Sums up why the stuff the TLT guys keep spouting is worse than useless for me too. Maybe it works for really good test groups but it just made a giant mess in the places I've been.

Those eggheads can't be out of touch because they weren't ever in touch in the first place. Lay 'em off and bring back the STEs.

Anonymous said...

Maybe you could at least talk about small things like the lunch delivery "benefit" with $1 delivery fee...

What's to talk about? If you don't like it, don't use it. It's a damn dollar. It's not going to break you. I assume they just tagged that cost onto it so that people at least thought for a moment before placing the order.

I haven't used the service yet (not at my cafe last time I checked), but I probably will. Sometimes I just don't have the 10-20 minutes it takes to pick up lunch and bring it back to my office. That's worth more than a buck to me.

Sums up why the stuff the TLT guys keep spouting is worse than useless for me too. Maybe it works for really good test groups but it just made a giant mess in the places I've been.

Those eggheads can't be out of touch because they weren't ever in touch in the first place. Lay 'em off and bring back the STEs.


I agree entirely! The TLT guys are a little out of touch with the reality of testing in many teams.

Good news on the STE front...fairly reliable sources have confirmed that they will be coming back!

Anonymous said...

"Maybe you could at least talk about small things like the lunch delivery "benefit" with $1 delivery fee..."

Damn straight. It's about time we had a great new important topic to discuss here. It is summer after all.

I want to put my two cents in to suggest they offer a variety of services to go with lunch delivery. I would pay an extra dollar above the one dollar to have my lunch delivered by two nymphettes who also will stay and provide shoulder and finger massage and pop grapes into my mouth as I contemplate important Microsoft business. (Maybe for a little more they could provide a white tunic for me to wear with a headband of flowers to sooth my aching expectations. Et tu Brutus?

Anonymous said...

I would pay an extra dollar above the one dollar to have my lunch delivered by two nymphettes who also will stay and provide shoulder and finger massage and pop grapes into my mouth...

Nice in theory, but it wouldn't work out very well. The company would probably outsource to a low-cost nymphette provider, and the quality of the nymphettes and their shoulder rubs would leave a lot to be desired. To say nothing of the grapes.

The other option would be to bring it in-house and make them blue badges. But then the nymphettes would need their own career path, with CSPs, commitments, etc., and pretty soon, they would never be available for lunch delivery and shoulder rubs because they would be too busy with tasks handed down by the NTL.

Michael A. Banks said...

In the absence of contact info, may I ask you to contact me regarding an anonymous (on your part) interview for a new book?
Thank you,
--Mike
http://www.michaelabanks.com

Anonymous said...

Ouch! First MS management slaps the AdCenter folk across their collective faces with the aQNT purchase...so much for faith in them.

Now this Forbes story comes out...how much longer will the embattled Ballmer hold out? I say one more quarter; end of the year, if he's lucky. Wall Street just isn't going to put up with this.

Link

Anonymous said...

Ouch! That story is a month old. Have you been under a rock or something>

Anonymous said...

"Now this Forbes story comes out...how much longer will the embattled Ballmer hold out? I say one more quarter; end of the year, if he's lucky. Wall Street just isn't going to put up with this."

oh, whatever. ballmer can drive the company into the ground for another decade and still stay exactly where he is. if you think anything short of his heart finally exploding from the constant red-faced rage at [insert whoever is kicking MSFTs ass today] is going to take him out, then there's a bridge i'd love to sell you...

Anonymous said...

City of Seattle may ban microwave popcorn:
http://www.king5.com/topstories/stories/NW_061307WABmicrowavepopcornbanTP.43bf29d9.html

Oh no, mister Bill, what will late night developers do now! Ohh Seattle.efdxz

Anonymous said...

Re: City of Seattle may ban microwave popcorn

We are not in City of Seattle :) Redmond is another municipality...

Anonymous said...

>"We are not in City of Seattle :)"

twue twue. Sorry for the inclusion in such an anal fixated provincial microcosm! Microsoft ought to protest and move all its satellite studios out of Seattle.

Anonymous said...

I didn't know M$ bought the Seattle City Government, it might generate more revenue than aQNT though.

Anonymous said...

>"I didn't know M$ bought the Seattle City Government, . . ."

Mini must be on vacation. He never lets the M$'s through.

Enjoy your popcorn while you can Mini.

Anonymous said...

The company would probably outsource to a low-cost nymphette provider, and the quality of the nymphettes and their shoulder rubs would leave a lot to be desired.

I'm sure the Partner In Charge of Nymphette Quality Asssurance would never allow that to happen, at least for the nymphettes delivering to Partner Level and above.

Anonymous said...

SHOCK AND AWE!

Ask your manager what your group's promo budget is for this year. If you think you've done everything possible to deserve a promotion and there's no way they could possibly screw you out of it, you might be in for a rather large surprise.

Time to start enforcing those 35 hour work weeks. Nobody is going to remember what you did in the first quarter of this FY anyway, so how can it hurt?

Anonymous said...

You can see blogs of people hired by MS but screwed by the H1B, popping up all over:

http://marchfirst.squarespace.com/blog/2007/6/6/its-official-my-microsoft-career-is-over-before-it-even-star.html
http://awesomearpit.blogspot.com/

I am now a US citizen, but came here on an h1B.. I shudder to think about what these people are going through.. any managers out there that can help people like these out?

Anonymous said...

Whoever said that MS really just bought the customer base and a few key people is dead on. I had several rounds of interviews at Atlas last summer (when I was looking to leave MS, actually) - enough to get a very rough idea of the architecture and technologies. Trust me - it will be the first thing to go. It will be placed in the loving hands of some of MS's "Architects" who will promptly declare it the Worst Design Ever and decide to "fix" it. Then it will not work nearly as well, and then they'll do what they always do: throw more hardware at it. Thus making it as well built and easy to maintain as the Incentive Compensation system. Or MIO.

As for leaving MS - it was the best decision I ever made (followed very closely by not taking the job at Atlas because I had a "bad feeling about it"). Not only am I paid more, I am happy at work. It helps to not have 20+ hours of meetings a week to attend...

Anonymous said...

twue twue. Sorry for the inclusion in such an anal fixated provincial microcosm! Microsoft ought to protest and move all its satellite studios out of Seattle.

Why? Any ban would only apply to city government facilities, as the article makes clear (well, except in your case).

Limulus said...
This comment has been removed by a blog administrator.
Anonymous said...

"SHOCK AND AWE!

Ask your manager what your group's promo budget is for this year. If you think you've done everything possible to deserve a promotion and there's no way they could possibly screw you out of it, you might be in for a rather large surprise."

Can you say more? Now I'm worried. My most recent promo was Jan 2006. I've performed at the level I want to be promoted to, since being hired in 2003, yet have had to slowly climb the ladder to get to where I should have been brought in given my true career stage (thank you HR). Last year's review was great, my mid-year was good, and I received a gold star at the end of last year. I don't think there is much more I can do to justify that given my performance, it's time for a promo. Are there any ways I can increase my chances of being one of the more fortunate FTE's?

Anonymous said...

Ask your manager what your group's promo budget is for this year. If you think you've done everything possible to deserve a promotion and there's no way they could possibly screw you out of it, you might be in for a rather large surprise.

Don't trust this. The budget is the same as last year.

Anonymous said...

promo budget ..., you might be in for a rather large surprise.

Don't trust this. The budget is the same as last year.


Same happened to me. Assured of promotion in May, every 1-on-1 it's "yes, you're doing great" and since last week, it's "maybe next year".

Anonymous said...

Same happened to me. Assured of promotion in May, every 1-on-1 it's "yes, you're doing great" and since last week, it's "maybe next year".

I'm in this same boat. I think it's going to be a big boat. Maybe the Titanic?

Isn't it interesting that in many places they made up their minds about promotions last week? Shouldn't they read our reviews to see if we map to the CSP garbage before making this decision?

My manager probably couldn't recall what I did last week. No way (s)he could remember everything I did in the reviews period.

Anonymous said...

>Why?

Pure, unadulterated Schadenfreude.

The good troll.