Corporate Blogs — Gradations of Value?


First, let me say once again, there has
been no corporate pressure whatsoever to cut short my previous posting
(Broken Windows Theory).  Nobody has
said, or even implied, that I need to change anything about what I said.  So conspiracy theorists, please rest assured
that The Man is not out there monitoring and censoring the blog world.  Seriously.

 

I pulled the content of the posting
because productive discussion wasn’t happening.  Of the 160+ comments, about five have
had any real value from an “open minds, open discussion” point of view.  I also pulled the content (once again,
completely self-initiated with no pressure whatsoever from anybody) because
there is enough internal debate within Microsoft about the value and ethics
of blogging
which I’d like resolved.


[Follow-up:  I’ve restored the original post, after much internal discussion.  Essentially, pulling the content was causing undue attention.]

 

Internal Debate

 

Many perspectives have been voiced to me,
both publicly and privately, debating the value and ethics of employee blogging.  Here, by “employee blogging,” I mean “blog
entries that are openly identified as being written by Microsoft employees.”  The rough gist of internal feedback from
Microsoft employees falls in these categories:

 

  • Thank goodness someone is
    talking about this
    . 
    “Kudos for having the courage to shed light on these critical
    issues.”  “It’s great that we’re
    having open and insightful discussion about this.”
  • You need to put the entire post back up.  Some folks are quite
    concerned that, with Scoble leaving this week and what not, there will be
    increased fervor behind conspiracy theories about how I’ve been silenced, shipped
    to Siberia
    , etc.  This sort of feedback is much more
    concerned about posts staying up from a PR/media perspective, regardless
    of the content of the post. 
    (Let
    me say here once again, for those who have deep-seated theories – my original
    post was shortened unilaterally by me. 
    I was at no time pressured to remove any part of it.)
  • Employee blogs should be an
    extension of the company message
    .  Folks in this category would say that the
    moment I identified myself as a Microsoft employee, my message should be
    on target with the corporation’s message, building a positive image,
    connecting positively with customers, etc.

 

 

Let’s Agree on Goals

 

From my perspective, it’s not a
free speech issue
.  I’m employed by
Microsoft, so there’s a valid discussion as to what sorts of posts are allowed
for me to make as an employee of the company. 
Conditioned in my employment can indeed be restrictions on what I should
and shouldn’t say – I buy off on that 100%. 
(For the last time, though, please remember:  no one has pressured me to change anything.)

 

Second, the simple case that I think
everyone agrees on is that nothing confidential should ever be divulged.  This is where Mini-Microsoft, as entertaining
of a read as it is, crosses a line.  That’s
also the reason it can only remain up as long as it’s anonymous.

 

The more interesting debate I’d like to
have is not whether employees can or can’t post certain things, but should
they.  I have no interest whatsoever in the
set of things that are clearly against company policy to post.  I’m much more interested in the spectrum of
things where people, even internally in Microsoft, disagree.

 

 

So How Does It Net Out?

 

The bulk of the internal feedback I have
gotten falls on the side of encouraging posts like Broken Windows Theory.  The vast majority of emails I’ve received
have to do with how the article has opened up important issues for
discussion
.  Folks in this camp say
that Scoble has given a human face to Microsoft, has made Microsoft more
accessible to the majority of customers. 
The openness, and in some sense, the vulnerability, of both addressing
our strengths and discussing our weaknesses has been refreshing, these folks
would say.

 

Another camp would say that blogs are a
key part of how a company is perceived, and that they act as a megaphone
of both positive and negative opinion.  Part
of an employee’s responsibility, then, is to at all times help build and
reinforce that positive image.

 

What’s most interesting to me is that
even within the company, we don’t quite agree on whether Broken Windows
Theory is a net positive or net negative. 
If I take it purely based on numbers, the overwhelming majority of
employees writing in say that it’s a positive thing.  But I see merits to both sides of the
discussion.

 

Thoughts?

Comments (52)

  1. Anthony Berglas says:

    Pleaes put it back — I suspect that there are good lessons for many software companies, such as my own.

    (This might be a better place for this comment)

  2. Bryan says:

    Yes, please put it back! Sorry of all the lame-itude in the comments, but I feel like I missed something important!

  3. David G. Miller says:

    First off, the original article is availavle from the Google cache at:

    http://64.233.167.104/search?q=cache:OxIU8nS0KaIJ:blogs.msdn.com/philipsu/archive/2006/06/05/617988.aspx+Broken+Windows+Theory+philipsu&hl=en&gl=us&ct=clnk&cd=1

    This article goes to the core of something I am very inerested in: software engineering as the term was originally intended.  In particular, I’m interested in how the nature of software development changes as the scale of the project increases.  

    Based on personal experience (I worked for a large aeorspace/defense contractor earlier in my career where projects entailing several thousand man-years of effort were considered normal), I question that anything all that remarkable happenned as Vista scaled up from the 40M SLOC of XP to 50M SLOC.  What I do see is a systemic lack of recognition on the part of MS management that bigger projects take exponentially longer, especially when the level of complexity of the project increases at least in step with the overall size.  I’m not singling out MS management here since I’ve seen this same head-in-the-sand (I’ll be nice this time) attitude at almost every place I’ve worked.  Management never wants to hear that increased complexity means it will take longer.

    Cheers,

    Dave

    dave(TheAtSign)davenjudy.org

  4. There’s an old ("old" by our industry’s standards) urban legend about irony that actually found its way…

  5. TechBlog says:

    There’s a cliche we’re fond of around there that applies to the permanence of publishing information, either in print or online: You can’t unring the bell. But Microsoft developer Philip Su, who yesterday posted an extensive blog entry on the…

  6. As another fellow software developer staring at Microsoft from afar, I find it somewhat comforting to see the imperfection in a company that can otherwise dictate the course of software development for years to come.  Hopefully, this post and others like it remind developers everywhere that no one has The Plan for writing great software, or to do so ad infinitum without making mistakes.

    Don’t get me wrong, Vista is fraught with mistakes — clearly many that were made at its inception that led to the eventual erosion of its idealistic goals — leaving us with "XP-Pretty", or something far less than it was originally intended to be.

    As a complete outsider, it’s also easy to point to the simplest solution: drop legacy support.  Apple pulled it off, and, yes, you would upset plenty of people… but it serves no purpose to continue on this path that will eventually lead them to alternatives anyway.

    It’s reminiscent (to me) of ASP+… the Faster, Better, Smarter version of ASP that would provide an ideal upgrade path for existing code while eliminating virtually every negative comparison of it against other web/scripted languages.  What happened?  The .NET Hype Machine turned it into ASP.NET, the over-engineered not-quite-ready-for-prime-time "web" version of Windows Forms.  ASP.NET 2.0 was an improvement, but remains helplessly over-engineered.  RAD is dead.

    In the end, the purpose of the platform MUST be to invigorate growth in its child markets, thereby expanding and securing its value for /n/ years to come.  Microsoft has lost sight of that, trying to make Windows the Best Thing, rather than allowing its (intended) transparency to facilitate the Next Great Thing.

    -Matt

  7. Branco Medeiros says:

    Congratulations for putting the original post back (Broken Windows Theory).  It’s a very interesting article (despite the annoying bold sentences). I regreted you having removed it, even more when anyone could get at the original text from Google’s cache (that was my first move when I got here and saw that you had removed it).

    Years ago, when MS employees started blogging, it really amazed me. It gives a human face to the company. I hope you don’t shut such a great thing down.

    MS has such a presence in everyone’s life, blogging comming from inside the company is great to remove that ‘IBM’ feeling about it…

    Regards and good luck.

  8. Rick Schaut says:

    Phillip,

    Why do we blog in the first place?  Your reasons are probably not exactly the same as mine, but I’ll bet they’re not all that much different.  I want to tell our story.  I want to get more direct feedback from customers.  I blog, because I want to improve the relationship we have with customers.

    I can’t see how to achieve that end without credibility, and I don’t know how to establish and maintain credibility unless we tell it like it is.

    Chris Mason hired me to work at Microsoft.  One of my favorite quotes of his comes from one of the original "zero defects" docuiments, "Since human beings themselves are not fully debugged yet, there will be bugs in your code no matter what you do."

    So, yeah, I think "Broken Windows Theory" is exactly the kind of post we should be making.  It does two things.  It acknowledgess that we, as human beings, "are not fully debugged yet," and it shows us debugging ourselves as we go along.

    Rick

  9. Brett Morgan says:

    I’m happy to see you guys actually thinking about the bind you have gotten yourselves into. It gives the rest of us some idea of what scales are do-able, and those which aren’t.

  10. S says:

    (Commenting here since the last post is locked.)

    I left Microsoft in 9/03 after working on a part of Windows Vista which went through the exact same Truthiness situation that Philip described in his post; an expectation from the program office, leadership across the company, etc. that we’d meet a mid-2004 (yes, mid-2004 for Vista) code complete date that everyone working on the program knew was ridiculous for us and assumed was ridiculous for others as well.

    I _could not believe it_ – perhaps I was still too naive – but I could not adjust to a world where I was clearly being asked to lie in order to keep our program from being in trouble. I could not fool myself and I did not want to fool anybody else, and so I had to do something else.

    I would have hoped that after failure and failure and failure again, somebody – anybody – would have stepped up and said enough. Still, no enough.

  11. Kip Kniskern says:

    Why blog for MS?  Yes it puts a human face on the Borg, but more importantly it puts "the owner" of a feature or a product in contact with the end user.  It makes it harder, in fact, to lie if things are going badly, if a dialog is taking place between the owner and the client.  It allows us outsiders to act as watchdogs, and as buffers between managers who want "yes" answers, and workers who, with a connection to their clients, have a stake in the consequences of being caught in the lie.

  12. Jonathan Day says:

    Well, I hate to say the obvious, but given the history of IT – not just of Microsoft but much of the corporate sector – paranoia is not simply a product of one group or another but is an inevitable byproduct of past practices. Sure, those practices may no longer exist, but paranoia has a comparable half-life to plutonium. Once the stuff is produced, it remains lethal for a very long time. It would have been better if it had never occured, but that can’t be helped now. Blogs – this one or any other – are useful in that they help decontaminate, but the process won’t be complete in my lifetime or in the lifetime of anyone on the planet today.

    Paranoia is a luxury that nobody else can afford a person to have.

    With that out of the way, are corporate blogs helpful? Anything that promotes healthy discussion is a Good Thing. Anything that inhibits necessary discussion is a Bad Thing. You could drive a globular cluster through the gap between those extremes, but you’ve got to start somewhere.

    What about Intellectual Property issues? That’s always a tough one. Don’t expect it to be solved by tomorrow. Unfortunately, I’m not convinced ANYone has the skill or experience necessary to understand all aspects of the issue, so I’m not convinced any existing philosophy is worth a whole lot. Economists are still arguing over whether John Nash’s cooperative economics is better/worse than Adam Smith’s notions on competition, and in which cases. And that’s in an area where there are specialists and there is a well-defined problem.

    "Broken Windows" is merely a simplified model of the fight between society as a collective good, society as a collective evil and society as a collevive that exists in every single possible state between these points, simultaneously.. However, it is a situation that exists only because society is a paradox within the framework it is usually defined in. If the type doesn’t support the value you need, change the type.

    Absolutely no non-trivial project in existance will ever have a workable deadline, or a reasonably bug-free product, if the project is pursued stoccastically. In other words, you can’t just define things as you need them and fit them in, because complexity is the factorial of the number of components minus the number of permutations of components that are logically isolated. If these are not well-defined, you have zero means of knowing the impact of any change. This is why it is paramount that design takes place before implementation.

    (If this is not possible – say, you have to be backwards-compatiable with some product or other – then you clean-room. You reverse-engineer a formal spec from what actually exists, and then you implement from the spec, NOT from the original code. The original code will have an unknown number of bugs, whereas a formal spec can – in principle – be proven both complete and correct.)

    Last, but by no means least, anyone who has done work for any Government agency in the US or UK (and I’ve worked on projects for both) knows that managers come from the B-52’s Planet Claire. They do not exist in the same world as engineers. About the best that can be done is to keep the lines of communication as good as possible, but that’s the limit of it. Geeks make for poor managers (usually) and managers make for poor geeks (usually). Arguing that one or the other shouldn’t exist is like arguing for monopoles and one-sided coins. If that’s what you’re into, I’ve a Klein bottle of wine I’d like to sell you. Given that the cultures are irrevocably distant, any error – however small or whatever the origin – WILL snowball out of control and WILL injure not only those involved but all those around them as well. If "failure is not an option", then neither is anything that permits failure.

  13. Henning Pedersen says:

    First of all, I believe that your original post ("Broken Windows Theory") didn’t really bring anything new to the table. There has been rampant discussion around statements by Jim Allchin and others only a few months back, and it’s no secret anymore that Microsoft has had, shall we say, "issues" with the size and complexity of Windows Vista.

    Secondly – it’s really good to see that Microsoft is finally "getting" what blogging is all about. If all you do is make your blogs into corporate-approved news-channels, noone is going to listen we have news.com, microsoft watch and anything ziff-davis makes for that. If you really, truly have a wish to engage and interact with your customers, you need to give the bloggers free reins. A blog is a personal avenue into the huge corporate body that is Microsoft. If you take away the personality, you also cut off the oxygen supply.

  14. shrinkers says:

    It’s nice to get some honesty behind all the corporate bs that covers most things these days. Something that was born in the 80s and is now just accepted as the norm.

    Everyone knows that everyone else is bs’ing about the reason for delays, and the only ones kidding themselves are Microsoft – usually Sales/Marketing. Take a leaf out of Google’s book – I’m not a Google zealot – I can think of numerous other companies who realise that there infinite amount of intelligent people out there who can see past the PR and Marketing crap of a company within seconds.

    Of course, those intelligent people aren’t normally categorised under the group "Microsoft Shareholders", the shareholders need to be re-assured as well.

  15. Wesley Parish says:

    And of course, as "first post" in the "Broken Windows" blog posing, I’m all agog to know if I’m one of the five elect, capable of staying "On Topic", or one of the damned …

    Secondly, whistleblowers like minimicrosoft are vitally important.  You see, "putting a human face" on a company is important, if for no other reason than people will put a "human face" on a company anyway, and if said company has made as many serious missteps as Microsoft, that "human face" will not be a pretty one.  Scoble put a more positive face on Microsoft, and showed it was possible to meet good, worthy people who worked there – though Jason Matusow was the first Microsoftie I actually discussed things with.  The local Microsofties aren’t much chop – the last time I had any dealings with one was 1993, and that was in relationship with my MS-DOS 5, my interest in the MS Windows SDK – which they forgot about, and never got back to me about – and I’ve ignored them steadfastly ever since.

    Where minimsft is important is showing there are ‘softies who care enough to take that sort of risk, in case the problems they see with Microsoft don’t get corrected and just get worse.  If Enron, for example, had had just one or two people with the courage of minimsft, the rot could have been stopped very early on.  Protect and cherish your whistleblowers – they’re a rare species.

    And lastly, I’m not interested in "conspiracy theories" except when used by the likes of Michael Shea and Robert Anton Wilson.

  16. Anton Antich says:

    Sorry for offtopic, but "shipping to Siberia" isn’t that bad, and at Microsoft it is rather good i’d say: last year the region grew >150% and shows the same momentum. And working with the territory that covers some 10 France territories can be lots of fun too. As for bears, I just got a photo of one near Redmond – haven’t seen any for 22 years spent in Siberia though :))

  17. biteydog says:

    Pleased the blog is back.

    I have to admit to being a Linux user for the past 7 years, but rest assured I am not abour to start screeching like a slashdot "fanboi".(We all have them.) Along with others I actually find it reassuring that what has been apparent for some time to the outside world is spelled out (in BOLD and white? – sorry, couldn’t resist it) instead of the corporate spin that flies in the face of observable fact. We in the UK have a government with tendencies towards this, as do other countries – sometimes one just wants the truth, however unpalatable.

    Successive releases of everything seem to experience slippage on a logarithmic scale, the Linux 2.X kernel series being no exception, and I am among those who believe that it is sheer size that causes an inevitably increasing delay. As scale increses a distributed development (such as the differentiation in the Unix world between kernels, X windowing and core apps) does seem to gain the advantage – on a (very) rough line count the two would appear to be of roughly equal size if one includes the same elements, although both ways seem to have their own problems, thoigh different in scale at present, and both methods lead to deadline optimism in public pronouncemets by team leaders and spokespersons.

    I was interested in your "50 layers, of which I know 2" (apologies for the paraphrase) – in the open source Unix model this is generally enough.

    As far as Whisteblower? The whistle has been getting steadily louder for years – Good Luck!

  18. Friday, June 16, 2006 2:17 pm. I applaud you and whoever for restoring your interesting lament — which I had to get off the google cache by the time I showed-up! In particular, the issues you described are exactly the kind of things that transparency, truthfulness, can help fix. Of course the Linux lunatics will show up and that’s too bad, but I would hope your story would get wide circulation where it’s obviously needed — Microsoft!

    — good luck

  19. Bigchris says:

    Polish your resume and sell your stock.  MS is going to diminish in both size and importance.  I’ve been where you are and some huge adjustments in your life are inevitable.  Your problems eminate from the top of the company and the longer those players are in charge, the longer and more painful the downward slide will be.  It’s time for a paradigm shift that only fresh leadership can visualize and implement.  Good luck to you!

  20. Charles says:

    Sincerely, what reaction or response precisely had you expected?

    You seemingly have confused the value of corporate blogging with the value of introspection and review, which need be neither public nor a blog but does need to be honest and informed. Microsoft seemingly avoids either. That lack of honest self-review was precisely your original point in "Broken Windows Theory", was it not?

    What is further overlooked by many is that such introspection and review (even when it does happen) also be "informed", i.e. that there be a comparison against requirements (usually drawn from customers, successful competitors, and investors) and that said comparison be honest in its evaluation. By contrast what seemingly has persisted at Microsoft for well over a decade at least is a studied denial that any comparisons have merit at all and what few strategic decisions are evident (apparent to outsiders such as myself) seem  to have been made in a vacuume.

    I commented on the original thread which you pulled that Microsoft has a great many people who know "how to do" but few or no people who know "what to do". Combine that with a management org in denial that refuses to learn and a work force that arrogantly thinks it has nothing to learn, and Vista is the inevitable outcome.

    The impact is that Microsoft management has pursued an ill-considered Borg-like strategy of attempting to assimmilate any competitor and technology without regard to strategic strengths and weaknesses, and to either entirely implement or (at least tightly couple) the associated technology into the OS so as to lock-in the customer to Microsoft.

    The conventional wisdom (unexamined heretofore) argues that was "successful" by revenue, stock valuation, and market penetration metrics.  Undeniably by those metrics Microsoft has been a very successful company. But not *because* of that Borg-like strategy but rather in spite of it.

    Microsoft clearly was the "killer" OS & app at the right time and right place. Gates rightfully deserves the lion’s share of the credit for exploiting his original strategem that made Microsoft an OS-supplier to IBM, and IBM deserves the blame for not being visionary enough to see what the PC market would become or nimble enough to maintain control of it, and consequently, for ceding it over to Microsoft. But Microsoft’s explosive success is due to explosive demand generated by corporate productivity requirements and the internet build-out and Apple’s early dismissal of IBM compatibility or connectivity, not to any "visionary" strategy pursued by Gates, and certainly not to Windows "quality".  Microsoft would have been just as successful without building IE into the OS and without the myriad of semi-implemented throw-away APIs, for example. That same explosive growth would have been well-served, actually better-served, by a solid OS and suite of core apps based on a well thoughtout architecture and infrastructure. After all, what other OS would customers have bought? And would they not have bought it consistently, in greater quantity, at even higher prices, had there been feature, performance, and reliability incentives to do so?

    But instead, Microsoft has lead itself, needlessly, into some very intractible product development and support problems their successful competitors (e.g. IBM and Oracle) have scrupulously avoided because by experience (their own and that of others) they simply knew better.

    The issue is not "how to" build the Vista OS that Gates envisioned, but rather "what to" have required of Vista (and Longhorn) to begin with – "What" was necesssary from a strategic viewpoint, not "how to" make Vista technologically neat. How to give Vista an "Aero look" pales in significance to why virtualization should have been planned years ago, for example. I suspect at this point that an effective, competitive implementation of virtualization will be found to be  technically unfeasible, and the revenues attributed to the Aero look will hardly compensate for the future support costs and loss of corporate market share, unless of course Microsoft’s future revenue streams are expected to come mainly from gamers.

    And while it is true that "Microsoft bashing" often occurs on blogs such as these, it is equally true that the criticisms have a basis in truth, albeit perhaps harshly and unconstructively expressed. Microsoft is not the "evil empire", but it is very much the "average empire".

    There are not 70,000 software genuises to be hired, and Microsoft (like every large corporation) is built upon average talent with a handful of execptional people. That Microsoft deludes itself in thinking it employs 70,000 genuises and chronically congratulates itself on how talented everyone is, perpetuates the myth and hubris that blinds the technical staff to "what is needed" instead exhorting themselves on "how to" write kewl code. Most of them are neither genius nor stupid, just simply average. But managing 70,000 average programmers requires a lot of process. Processes which exist at all successful large companies but which (like the Borg mentality) seem to have run amuck at Microsoft, ostensibly attempting to compensate for the poorly architected products and infrastructure.

    The problem at Microsoft, in a nutshell, is that large numbers of otherwise average people have over the years risen to their respective levels of mediocrity, that senior management is in denial over its leadership and marketing failures, and the corporate culture has substituted "passion for technology" over solving customer needs.

    The cash cow that Windows has been has, until recently, obscured the need to reflect on these failures. That you (and others) blog about it only makes that long-overdue reflection a bit more public.

  21. gpd209 says:

    Thanks for putting the post back up–I’m glad I got to read it.  Nicely done.

    I’m curious if anybody knows how some of the factors in this analysis of the Windows team compare with Apple’s OS development strategy.  In particular–

    1) 2000 developers in Windows right now.  Anybody know how many on OSX?

    2) 50 million lines of code for Vista.  OSX?

    3) Windows hierarchy 11 deep for Vista.  OSX?

    4) Action control vs. results control.  How does that play out at Apple?  

    I don’t mean to trigger a war among the ideologues.  I’m just curious how the inner-workings of managing OS development compare.  The outer workings are obviously different, with Apple prioritizing more rapid and incremental changes over the past 5 years.  On the whole, it seems that people are generally less critical of Apple’s strategy than Microsoft’s.  So why is Apple capable of this style of output, when it appears that there is nothing MS could possibly do to bring Vista to market more rapidly?  I can think of lots of possible reasons, and I’ve read all sorts of explanations from various partisans, but I’m hoping maybe someone can offer a balanced perspective on this particular contrast between the companies.  

    Thanks!

  22. Greg V says:

    Thank you so much for the great article on Windows Vista development! This article actually gives me MORE confidence that Vista will be a solid product whenever it does ship. A lot of managers tend to forget that QUALITY software development is all about identifying, defining and solving problems. The author "gets it". This is exactly the type of pure unqualified dedication to quality that Microsoft has lost in recent years.

    The author of this article should be given a promotion for his clear thinking and dedication to quality above petty corporate politics! If Microsoft were to put him/her in charge of Vista development I would run to buy stock in Microsoft!

  23. zzz says:

    The author is apparently in Tablet PC group. I’d like to hear why aren’t Tablet PC half or more of the new notebook market? Either the implementation is bad or the upper management hasn’t done as much they could.

    Tablet breakthrough strategy:

    price $800-2000 (the spec suggested below is for higher end 1500-2000 range)

    Many brands: Fujitsu,HP,Acer,Lenovo

    Ensured display at consumer retail outlets.

    Tablet PC 2007 spec expectation:

    Core 2 Duo

    1-1.5 kgram

    11-14" 1600 or better resolution bright, vivid, S-IPS, no glare touch screen with button to turn touch off

    1-2 GB Mem

    80+ GB HDD 1.8"

    ATI X1xxx

    Remove any of the above requirements and I’m not interested.

  24. biteydog says:

    Further to my earlier post, and more thinking, I believe you have become the victims of "biteydog’s law". Well, until I do some more maths (not my best aptitude, but good enough) better call it "biteydog’s theorem".

    BITEYDOG’S LAW

    The time to completion of a software project is proportional to the cube of the number of lines of code.

  25. tom trinko says:

    First a disclaimer, i’m a mac fan.  I think that part of the problem is that Microsoft in it’s march to monopoly control  of the software universe is putting too much stuff into the OS.  They successfully used that tactic to crush Netscape and they probably view it as a winner.

  26. SYnoia says:

    AT&T when it had 1 million employees had 6 levels on management.

    The Military has 7 to 9 management levels and they are huge (don’t could the officer ranks, count the "line positions").

    Wesley Parish says a lot of it when he refers to "The Mythical Man Month", IBM’s near bankrupcy in the early 1960s when developing its first large OS. What Brooks did not document were the political and organizational items that later broke IBM’s product invention and development model. What Brooks did identify was the issue of "Architectural Integrity".

    What’s the reward system at MS, do you get more maney managing more people (up the organization)?

    How does MS identify & promote "High Flyers"? How long do they star in a position? Is it more than 2 years?

    Another (very hard to read) book is "Softwre Engineering Economics" By Barry Boehm, Boehm would demonstrate that MS’ Productiviity is good.

    I’ll bet testing is a bitch, and almost never ending.

    How do you measure progress in testing, bugs discovered or velocity of bugs discovered?

    I’ll bet you have system & stress testeing scheduled late in the cycle.

    Why with OCX and DLL dynamic linking do you have a "Windows Build"?

    How global are you header files? (This could be very bad). OO Concepts especially apply to source code, and are generally never applied (Oh, I’ll just include this header file as well…)

  27. SYnoia says:

    By the way, you need to run system test IN PARALLEL with development…

  28. Brian Clark says:

    Without getting into the "which OS is better" argument, I think that Apple has done one really interesting thing that Microsoft can’t or won’t do:  they’ve shipped a new version of their OS every 24 months or so. The way they’ve done it is by doing their upgrades incrementally rather than stuffing so much new technology into each OS release.

    Why won’t Microsoft do this? I don’t know the answer. They used to provide more functionality to the OS with Service Packs (like in NT 4.0 service packs – remember SP4 and the Option Pack?).  And they seem to have done something similar with respun version of Windows 2003 Server. Why don’t they do this with the Desktop OS? They could have done a lot of different things.

    Maybe they’ve painted themselves into a corner with all of their legal problems? Consider IE 7.0? Why didn’t they work on this separately and ship it a year ago for WinXP SP2? I am sure that it would have been well received. Why not an SP3 for XP? (probably because everyone is so busy on Vista). If MS had a schedule to ship SP’s each year after an OS ships, they could have much more easily slipped in some of the Vista functionality (API’s, etc) for developers to try ahead of time.

    That’s my take on this whole thing. For a project as large as an modern OS, you have to do things incrementally. Otherwise the complexity is overwhelming, even for Microsoft — a company with the deepest OS engineering resources in the world.

    Brian!

  29. Dissapointed Windows Developer says:

    Gee, thanks, Mr. Su – you’re doing wonders for the morale of the 7999 people still busting their ass trying to get Vista done. Looks like you bailed out before the project hit the home stretch  – couldnt deal with the perceived problems, didn’t try to fix them, but just ran away,eh … ? Despite all the ‘commendable’ responses your post seems to be getting, as a loyal and dedicated Microsoft employee committed to shipping the best OS in the world, I’m deeply dissapointed and totally fail to see exactly what your post aimed to achieve? It’s a gripe, it’s a rant, and while you’re totally entitled to express your opinion, all you’re doing right now is airing dirty laundry in public (Do we really, really need to know your underwear is dirty and torn?), giving the press and Microsoft detractors gossip-rich material for sensational headlines, and demoralizing employees and distracting management from their task at hand – shipping Windows *on time*…  

    Yes Windows has problems – so does any project and so does any engineering culture – the problems dont go away overnight – and they’re certainly not fixable from the ‘outside’. And the problems are not even as systemic as you make them sound. Do you really know what it takes to ship an OS on time and to scale a project up to 8000 people? As Microsoft has hired 40000 people in the last 6 years, statistically there is a majority of people in Windows who have not shipped an OS (and I would guess that includes you too) – the development tools we use interally are truly state of the art – but in order to standardize the development process across numerous sub-teams, 2000 developers, and 8000 project memebers, to build a world-class OS that behaves consistently whehter used by the IT geek or the grandmom, and that can serve the needs of 500 million users, you almost have no choice but to turn the process dial to an extreme. 10 years ago, every developer wrote code with minimal process – look what a security and sustainability and nightmare that created in the long run. Yes, there may be too  many layers of maangement but I’d be surprised if putting the internal house in order isn’t already high on Kevin Johnson’s and Sinofsky’s radar. But that will happen after Vista ships – you don’t take the boat apart midstream to re-engineer it.

    Unlike you, I *have* worked on previous versions of Windows and I can tell you from experience that if the VP does not set a date and is not hard-core about the date, the OS will *never* ship – people who haven’t shipped OS seem to be unable to resist the temptation of perpetually tweaking their features till the last minute. When a VP comes in and says can you ship by X date or you must ship by X date, it’s not a multiple-choice question – you better get your act together and whittle your feature are down to whatever fits – or be brave enough to pull the feature. If anyone is to blame for lying here, it’s the project manager who is not able to cut back on his feature to meet the schedule or isn’t willing to stand up to what they beleive in. From the VP’s perspective there may be a thousand different features in Windows and it’s all about making the numbers work – you dont change the ship date to accomodate individual feature areas – you change the features to accomodate the ship date. That’s it.

    All said and done, I’d have respected your opinions if you’d been making any earnest effort to change the system from the inside (talk to Windows VPs, use internal blogs – or write a ThinkWeek paper for Bill Gates – there’s multiple forums for expressing what you beleive). Standing outside Microsoft, jumping up and down and shouting ‘Liar liar, pants on fire’ for the benefit of the tech-papparazi – surely you’re joking Mr. Su if you think that will cause any dramatic changes…

  30. philipsu says:

    To "Disappointed Windows Developer":  I agree that the Broken blog was largely ranting, and furthermore agree that it wouldn’t be the ideal way to go about effecting change.  That, in fact, was why I pulled the blog in the first place.  (But I might venture to suggest that most blog entries, by most bloggers, aren’t put out there with the express purpose of effecting change — most, like mine, are written as perspectives on the world).

    There’s so much to say, but too little to do in public.  Feel free to get in touch with me internally, and we can have a far richer, more productive discussion.  In the meantime, several thoughts:

    1) I have shipped an OS.  Actually I’ve shipped three.  Tablet PC 1.0, Tablet PC Lonestar, and Mobile PC Beta 2.  Every one of those releases were on time, in high quality.  The Mobile PC team was one of the first in all of Windows to get stable for Beta 2.  I’m not the one who deserves the credit for all of that, but I can surely speak to shipping on time.

    2) Yes, I have tried to get things changed from within.  The discussions I’ve had around this area are too many to list here, but it’s incorrect to assume that because something is said publicly, it hasn’t been said much more so privately.  And those discussions continue.  Much to the credit of our senior managers, their doors continue to be open to us (and to me specifically) on this topic.

    3) Setting a hardcore date is only the beginning.  One misses the entire point of the article if one feels the solution to shipping on time is simply to set a firm date.

    4) (Perhaps most interestingly)  Nearly every single employee that has gotten in touch with me about the posting has thanked me for posting it.  While that doesn’t in any way detract from the points you make, it’s just an FYI so that you can put your feelings in perspective with how others in the company might feel.

    So in net:  I make no pretensions that the posting was constructive — there are some things that can’t be unsaid.  You made a few errors and assumptions in your critique that are not true.  I’d love to have a richer, more productive discussion with you in person.

  31. Not that Philip needs rescuing, but Disappointed should also realize that Microsoft’s non-shipping isn’t just a gossip-mill nightmare, but an action that affects hundreds of thousands of outside IT staff, OEMs, and, most importantly, developers.  The first PDC build of Vista was three years ago!

    Now Microsoft is rushing its "partners" to provide Vista-friendly branded products (that leverage WinFX, meet the HW spec, etc.) in time for launch.  Vista has had 5 years and an 8000-strong staff, and is turning out to be an incremental improvement (largely chrome) over XP.  WinFS? Indigo? I won’t even go back to the originally-promised specs.  We (the outside world) have been chasing .NET for these past five years.

    You’re not the only one who’s Disappointed.

    -MattL

  32. Philip, I must give you much credit for trying to discuss Vista’s late schedule. I wonder to myself if Bill G’s slow exit from the company was a result of the Vista mess. I agree with the previous posts that Vista simply tried to take Windows in two many directions at once (OSX eye candy, .NET 3.0 foundation, new IE7, built-in virtualization, and numerous deleted technologies that couldn’t be finished on time). It is one thing to incrementally add these in service packs over time, but to create a complete new integrated whole, and rewrite much of the underlying code at the same time, was just too much. I find that if I get many changes to my products going at once, it takes much longer to finish the job, even if there are advantages in the integration.

    Based on what I see in beta 2, I look forward to the new Windows. The things I don’t like are the lack of backwards compatibility. While most of my existing programs will run without modification, there are some big changes in how drivers work. While I rely on large players for my I/O hardware, I suspect it will be some time after the OS release before new drivers are available. Windows is becoming less suitable for the soft real-time type work that I do all the time due to feature bloat and the conversion to .NET applications (which are just not fast enough for rapid image processing). And don’t try to sell me on one of the mobile versions. I am going to erase XP Embedded off of a small computer I got a few days ago and install W2K or XP just because XPE doesn’t work correctly.

    I find that management can be one of the worst factors in slipping delivery dates and quality. When I first started a major project a few years ago, if I had followed management’s plans to use an internally designed grab board the project would have never been in the market soon enough to take hold. I also had to fight against several people, who due to their excessive egos, tried to move the project into directions that would have created a lower quality product.

    Just my opinion, but I would place most of the blame with management, and we all know that comes down from the top.

  33. Dissapointed Windows Developer says:

    <i>"…I have shipped an OS.  Actually I’ve shipped three.  Tablet PC 1.0, Tablet PC Lonestar, and Mobile PC Beta 2…"</i>

    You’re joking, right? ‘Nuf said…

  34. And look at the news today! It sure looks like those at the top still haven’t gotten the message.

    http://www.channelregister.co.uk/2006/06/22/microsoft_predatory_pricing/

    http://sunbeltblog.blogspot.com/2006/06/microsoft-practices-predatory-pricing.html

    A lot of the delay in Vista was simply trying to get too many things in the OS in order to make others keep trying to play catch-up. Too bad some of the real innovations such as a new file system didn’t make it while the eye-candy (a MS catch-up to OSX and a truely useless feature for business users I might add) did. Hopefully the new security scheme will actually work!

    No wonder MS is such a hated company.

  35. Another disappointed says:

    "…I have shipped an OS.  Actually I’ve shipped three.  Tablet PC 1.0, Tablet PC Lonestar, and Mobile PC Beta 2…"

    Yes that has to be a joke.

    That simply means Phillip got on the XPSP2 wagon. That does not count for "shipping an OS".

  36. hfdh says:

    fzfhghdfhgfbilll sjgfjkjghköfdjh agat6rgatsdjhjhgjhfjgbidll is sdsjkfj der bsdkjfjkjfbeste jk djdern die ksjdkjfkjwelt hghjfhjjgesehen hdfhjhg hat djhfhbill dkjkgjjgatesjkjdgkj

  37. Alok B says:

    philpsu,

       I don’t buy your story. I don’t buy that it is a completely honest write up or that it is such a free (or carefree) and open expression.

       Afterall, how hard can it be to implement blogs.msdn.com to keep entries internal within MS for a few minimum days or until some approval check before they are made public to the world? Afterall people are writing for the company blog, and about the company or its products. I believe such a check system would be in place and hence your dramatization about pressure, blah blah …, seem moot.

      Now for the goodness of me, what I don’t figure out, is the marketing angle to why gobbledegook state of internal affairs dirty linen is being publicly washed? Is it for just more eyeballs, however they come, that you would bare whatever?

    Would you comment on this?

  38. Alok B says:

    Ok. I had some thought about this, and have decided that Billg or whoever up there would have really thought about blogs.msdn.com and must have decided that to choke it would be to allow anonymous leak all over the Internet and more head aches. So my check system argument was weak.

    But still philipsu is presumably writing from within MS, using MS systems and MS logins. Can he really be anonymous? I would’nt bet about it. So why make such posts in the general public???

  39. Vista . The term stirs the imagination to conceive of beautiful possibilities just around the corner.