The rise and fall of Microsoft’s UX platform – Part 3


It looks bad, I mean it really just looks bad. The President of Server & Tools in PDC just came out and pretty much implied that the race between HTML5 vs. Silverlight internally is over. The winner by way of Presidential nominee is HTML5.

It's easy to assume that maybe Mary Jo got it wrong, that maybe she took some journalistic spin to the overall story and tricked BobMu into saying things he didn't want to (it's what Journalists tend to do sometimes). Think again, Mary Jo doesn't play that game and its exactly why she gets these types of interviews in Microsoft, so why start now?

It's also easy to assume that maybe BobMu is just some empty jar head executive who says a few buzz words here and there? someone who typically isn't informed of the inner workings of one out of many products that fall within his portfolio? sorry, that's not true either. Each quarter when I was in the team, we'd have what we call "RTB's" - Reviews of the Business. It's that point in time where the team would put together a PowerPoint deck that covered everything from roadmaps through to metrics associated with the said product. BobMu was not only informed, he'd make decisions that we'd all react to post such meeting. He was informed and unless he was heavily medicated, he meant what he said.

What's the story then?

It's not like I didn't warn all about this turn of events a few weeks ago, (read part one of this post). The story isn't whether Silverlight is or isn't dead. I don't think Microsoft could even kill off Silverlight to make way for HTML5 just yet (HTML5 is simply still a science project in the market). I think what we are really seeing is a company as large as Microsoft in chaos.

You've got a President doing PR 101 mistakes, You've got a marketing team that double down on a single product instead of their entire UX Platform portfolio, you've got the Internet Explorer team writing their own messaging that confuses the masses against existing messaging. You've got an IE9 demo at PDC that smells, tastes and looks like a previous one in MIX07 only without the word Silverlight in it? You've got Silverlight not making an appearance at PDC which isn't a bad thing given MIX is really the party for Silverlight, but given market conditions - YOU SHOW UP.

Bottom line is this, the entire Server & Tools business within Microsoft is in dire need of marketing reform. The strategy coming out of Redmond is chaotic at best, the design and develop discussion has obviously changed within the belly of the beast. The problem is, they've kind of forgotten to inform the masses of this and we're only just starting to see glimpses of the inner truth now - and its frightening the kids especially when its Halloween time!

I did want to dedicate this post to how Microsoft has shut down the entire "designer engagement" strategy across all divisions, but clearly this is simply a symptom of what we're now seeing unfold online.

Microsoft is by all outward appearance shutting down its vision of the circa 2007 UX Platform, it's now winding it back to secondary citizen in favor of the new shiny object, HTML5?

I for one reject our new HTML5 overlords.

The problem with moving Silverlight & WPF back to the end of the visibility line, is that nobody really has sat down and asked existing rich client developers what they think of this new vision? it's a forced entry into the market mixed with a whole bunch of messaging from the Internet Explorer team that's labeled "Trust us, we have this covered" seal of quality assurance.

The one team in which has breed so much distrust in Microsoft. It's probably the biggest cancer within Microsoft and is the main reason why the Consent Decree exists.

The cold hard reality is that most developers actually probably don't want to go back to Circa 2005 development with extras (i.e. JavaScript and HTML suck). The entire HTML5 strategy is basically a mess on its own as you've got Browser catch-up's that still need to be done.

You've got issues around browser owners looking into ways of forking the HTML5 concept to suite their own agendas? You've got tooling coming out slowly and half baked? You've got a mixed reaction of what HTML5 actually means? You've got anxiety over whether or not JavaScript and HTML can scale? you've got millions of devices today that just can't load HTML5? You've got at least a 2-5 year latency effect world wide of enterprise even considering HTML5 in its current form ... the list just goes on.

It's one thing to get onto a soap box and declare a true x-platform strategy like HTML5 the future? it's one thing to say "open standards or bust" as being the mission statement of the world's future software ecosystem.

Someone just point out where the strategy exists for moving people both willingly and unwillingly across the desktop/plug-in divide over to this new world, because if Microsoft is running this show, we're all basically f#$%ked - I say this as right now, they couldn't organize a virgin in a brothel to get laid (as they would be too busy fighting over which girl was the prettiest).

Silverlight vs WPF vs HTML5?

Pete Brown last week released a blog post around the future of WPF which talked about successes and hints at the future of the platform. Pete did something extremely hard in making this post come together, he went internally and asked a simple question "Where is this bus heading?" and that's just before PDC2010 as well - big hat goes off to Pete for pulling this together, as many have tried and failed that little mission.

It's still not enough though! - now before you grab your pitchforks and declare me a jaded hateful ex-WPF/Silverlight team member, hear me out.

The reason I say it's not enough is that we just heard 200+ engineers are working on Silverlight/WPF and looking at the new additions to WPF, i can't but help wonder how thinly the team are spread. There is a lot of surface area to cover inside WPF, the biggest of which is around performance and getting line of business grade features onto the table. The WPF team are reacting to the data they have and unless there is radical changes since October last year in the way they get this data, it's still a ways off (the product usage data etc inside Microsoft is simply disconnected and a mess, features are skewed between what looks fun vs politics etc).

It's not enough, there needs to be a consolidated marketing strategy around the product(s), there needs to be an Evangelism rhythm that maps out how this drum beat gets played out worldwide, i.e. its one thing to announce how you intend to build something - its entirely another to actually get that message to every developer you possibly can.

It also needs to connect back to Silverlight. It needs to fit in with how developers can navigate the ye olde "It Depends" response from Microsoft. The guidance Pete used was old, I know as it was something we crafted back in July last year - "Use Silverlight until you hit a wall, then go WPF" was pretty much the summary we came up with (even then I remember thinking, that's just bullshit but what else can we say? WPF is dead? :D)

WPF also needs to connect back into HTML5? so how does the new IE9 overlords and WPF play in the same sandpit together? at what point do you separate the two? Windows 8 team have ideas on this, but I'm pretty much betting that the HTML5 story will get more air play in that pool of brilliance.



Lots has been said in the past month, bottom line is this. The technology is currently a big software buffet, we have loads of options and different portions on the table to pick from. We need informative views more so than ever now, given the emerging mobility vs. desktop discussion and more importantly how all these pieces fit together.

Microsoft lacks the marketing muscle right now to answer these questions, they simply just don't have the maturity needed to lead this vision forward. You've got pretty much majority of the executive branch abandoning ship and the competitors they used to just sweep the legs out from under are basically starting to get their act together.

Adobe for one has its act together finally, I've watched this company for years fumble around in the dark around this entire discussion. At MAX 2010, they not only connected it together but they did so in a way that is slow, simple and has the appearance of saturation + ubiquity.

Microsoft's shows up and starts waving its hands in the air about Internet Explorer 9, Azure and how Silverlight is now winding down - not to mention zero WPF discussion (except for Rob Relyea - owner of WPF Team - picking up the Developer Platform & Evangelism divisions dropped ball and doing a PDC session on WPF).

Bob Muglia needs to really take a hard look at his organization tree and start putting together a plan of reform. This isn't a technology problem anymore, it's a marketing one pure and simple.

As for Silverlight Marketing Team getting ahead of the PDC2010 fall out? – “Out of Office” summarizes it all.

Related Posts:

  • woodjoe

    @MossyBlog good post. But it’s not all marketing. There is a technical connection missing linking their dev strategy to HTML5.

  • DenisBasaric

    @MossyBlog View from the ISV side:

  • I certainly can’t speak for all New(ish) developers out there. But I can certainly speak for myself; This is a massive black-eye for Redmond.

    I’m still learning the basics of development and programming. That said, I spend a lot of time trying out tons of stuff on tons of platforms. I had super high hopes for Silverlight and WPF.

    If WP7 is a flop, what happens? They are so far behind the pack, that I have to wonder if WP7 has even a fighting chance, and I hate to say that. The Consumer needs more choices not less.

    If they really want a fight chance, why didn’t they make Silverlight functional on Android (somehow; Browser or native) as a primary goal. Cross platform in the mobile space is the holy grail and everyone seems to ignore it, except Adobe.

    The technology along with marketing is the problem. As a student, I need to buckle down and make hard choices about my primary toolset. The more I think about this, the more I need to stay away from Microsoft. I really, really like VS2010 and Expression Studio in comparison to the plethora of tools I’ve used with Java, Python, JS and HTML.

  • MindscapeHQ

    @DenisBasaric We just posted our views, similar to your organisations views:… (cc @MossyBlog )

  • hybridweb

    So, should I go back to my pre-2000 roots and pick up MFC again, where I left it? Or maybe download MinGW, Qt and Eclipse C++?

    Need guidance, as there is a 1-2 year shifting time.

    (I’m saying this publically in the hopes that someone at MS is reading, and beginning to see how perilously close they are to losing developers. I can’t believe that I’m the only one with these kind of thoughts…)

  • hybridweb

    @ Michael Martinez:
    You know, I can almost see in my mind’s eye all those grey beard hard-core Win32 types, just sitting there, saying “I told you so”. But for a few years, I might be among them.

    I don’t know how this is all going to play out, but I know where I am going to begin investigating: the old C++ and TrollTech Qt path. Qt is still evolving nicely; they are adding a declarative UI package to the system; they have support for scripting through JavaScript (if memory serves me).

    If enough of us can convince MS that we’re serious about switching if things don’t improve, maybe they will listen (Or is it too late for that, Scott?)

  • sheva

    I’ve got 3 years of sweat and pain skilling up on WPF / Silverlight. 6 years of .NET sweat, and 11 years developing on the MS platform.

    Personally I am going to see how this pans out in the market. If I notice WPF/SL falling away in terms of job ads then I am abandoning the Microsoft stack. I’ve got a few gripes I could live with, but all this FUD on top of it all means I am starting to get pissed off with it all.

    However, I must say in the recent short term, I have been overwhelmed by the amount of work available in SL/WPF, and it is paying very well. I have been approached by 3 foreign companies to join them (WPF), and my contracting rate has increased by 50% in the last 12 months due to the demand.

    I haven’t really decided what I will do, but I am thinking of skilling up in Flash/Flex. Oracle make a fabulous RDBMS which I have used and enjoyed, or perhaps I will go all NoSQL and CQRS for data persiste

  • @MossyBlog:
    Great article! I completely agree!

    I was completely appalled by the lack of WPF/Silverlight at PDC. I’ve been a follower of Mary Jo’s blog for a long time, and she’s one of the most respected journalists out there. I don’t believe there was any trickery involved here at all.

    Silverlight is the new .Net Compact framework. Or more specifically, the development platform for Windows Phone 7 (which oddly enough doesn’t support HTML5 either). Does this somehow admit failure of that device if you read between the lines? If HTML5 is the future, and WP7 is Microsoft’s mobile platform going forward, you’d think they would have at least had a demo of IE on the phone capable of playing HTML5 content at the PDC keynote.

    I personally see this as Microsoft basically “giving up”. Not only on the front of content delivery technology on mobile platforms, but on their own operating system. It seems they are too busy playing catch up with Google and Apple and wanting to be a “me too” company to care about taking radical measures to be a leader. Their answer? Make a really fast browser. Google and Apple aren’t going to be beaten at their own game. Just how long before Webkit performs as well if not better than IE9? Clearly the real strategy here was to win back browser market share on the desktop thats been declining over the past 5 years. And in the end, it will be futile.

    @ DenisBasaric & @ MindscapeHQ:
    I’ve noticed a lot of companies re-investing in Winforms (the company I work for uses Telerik’s library).

    However, I have to wonder if the reason why there isn’t so much of a market for WPF components is possibly because WPF is far more customizable than Winform. WPF component companies seem to thrive on teams that are familiar with Winform but are new to WPF. I know that was the case for us, and as we’ve become knowledgeable of xaml/WPF/Silverlight we have found ourselves slowly taking the Telerik controls out of our applications because all they seem to do is add another point of failure for the application without any real added value.

    @ hybridweb: But then you would have to use C++ 😛
    Qt’s x-platform strategy isn’t quite as portable as HTML5 (or Silverlight for that matter). AND it’s owned by Nokia (which isn’t a bad thing unless they decide to change strategy and/or get acquired by some other company less friendly to open source.).

  • albsure

    Hi Scott, I heard your interview last month on that Silverlight podcast and immediately knew it was serious! The thing I’ve learnt in life is that you ALWAYS listen to the person who has been or is on the “inside”. The inside is the reality, the outside is just spin.

    I’ve been saying to a colleague that MS has an inconsistent strategy for years. Its like they have the classic “too big” syndrome where they have so many autonomous divisions with differing visions of the future, yet to the outside world they look schizophrenic.

    WPF/SL/LinqToSQL etc.. are great technologies that will die because of politics and not because of their usefulness. Thats whats so sad about it. If they were just bad products (like Java Swing etc..) I’d understand, but they’re not bad at all. Just victims of a company that has too much money and no real direction going forward. Their in the worst position, a company that doesnt have to do anything to survive is always the worst company to be in. They can relax on Windows and Office licenses for years and not have a any problems. They are not hungry at all. And as you have rightly pointed out, they are just a massive “me too” company at the moment.

    The sad thing with software development is that you have to devote your life (social life etc..) to learning something that you probably wont be using in 5 yrs time based on company politics, not on the usefulness of the tool.

    I dont know of any industry in the world that changes as much as programming. I can honestly say that in the 13yrs of me being in the software industry, nothing I learnt in my first few years is relevant to my job now! Not one thing! What other industry is like that?

    The thing I’m learning is to stay away from Microsoft tbh. Yeah, its lucrative if you managed to hedge your bets right but maybe the HTML5 crew have a point. Slowing down the pace of change and building on your existing skill sets allow you to have careers for 10-20 yrs, not 3!! If you knew HTML, Javascript, Photoshop 10yrs ago you could still be in work now.. If you were hot at VB6, RDO, COM your probably not working now… Says it all really..

    Finally, the reason why WPF and Silverlight have problems is that there are no killer apps that show off the technology. There is no “gmail” which shows of AJax. There is no “BaseCamp” that shows off Rails/MVC. MS only just got round to developing Visual Studio in WPF and that is half and half. If there is anything hot its in an intranet somewhere never to be seen by the outside world.

    No one has shown the world what is so special about SL/WPF from and end users point of view. So why should anyone adopt it? Bet their careers and companies on it? If the company that makes it isn’t even using it develop all their software, then why should we?
    Its all simple marketing and common sense. Something that due to the political infighting, MS have failed to get right. And the reality is, they don’t actually WANT it to be right. People in the company don’t want other people’s projects to succeed. That is a dysfunctional situation.

  • D

    After 10+ years on the MS stack I’m tired. I’m tired of MS pushing the next latest and greatest only to kill it a couple of years later. I’m tired of half baked controls and approaches. It took ten years, but I’ve learned now not to trust the MS stack. MS, you’re on top of very unstable ground. You’re going to lose us, the developers, because of your inability to come together as a company and provide one solid vision for the future. The internal politics is killing the company and what are good technologies. I’m tired of wasting years getting a technology under my belt only to have it put out to pasture. I’m tired of having to defend unsound business decisions because all they want to do is sell sell sell. Time to seriously look at other technologies to get the job done.