Why Silverlight was rigged to fail and my time leading it
riagenic.com
riagenic.com
It boils down to an egotistical hypocritical jerk uses fights over the Silverlight marketing website and strategic plan to try to illustrate problems with the Silverlight project, but fails to make any useful points other than that there was some corporate infighting.
This simply isn't a good article, please don't vote it up, and I recommend you don't waste your time.
It also seemed very poorly written and ill-considered. If I were interviewing this person for a job I'd be very put off by the tone, the solecisms, and the lack of clarity.
Given the recent barrage of positive tech articles about the upcoming Windows 8, I enjoyed a look from the inside at what looked from outside like a remarkably successful product (at least on the marketing side).
The conclusion the author was trying to sell (I think) was that Silverlight didn't have a five year plan so that is why it failed. But it sounded like no matter how well it was doing it could have been killed simply because HTML is the new hot thing.
Honestly he comes away as a guy who didn't have much influence and when he didn't get his way, got aggravated and started fights. And when the product was killed off he is trying to spin it as though he knew it was going to fail all along and wasn't listened to, but to what point I am not sure, it isn't helping him.
Edit: and him openly admitting to being the leak for a major tech news lost him all sympathy from me. People who leak news that they know they shouldn't rub me the wrong way. For someone who does consulting this seems like a kiss of death for future jobs.
Something this long shouldn't be written in a single draft. If it's worth nearly 10,000 words, it's worth doing properly.
TL,DR: Guy realizes he was a jerk at the start of Silverlight. Continues to be a jerk, but doesn't notice. Oh, and Silverlight is dead.
Otherwise you leave yourself open to criticisms like: "If he can't execute on the small things in a blog post, then no wonder Silverlight is dead" etc.
> A guy I barely knew but was in charge of the video side of Silverlight?
> Fine, I thought maybe I can get a fresh perspective and start something new in this team?
> I’m riding the Junior bench on all these ideas that he stole from not only Brad (my old boss) but also me?
Seriously?
And on content, it's stuff like this that shows where he's at:
"A guy I worked with Joe Marini, tweeted about Nokia phone he was excited to hold in his hands. I like Joe, he was a nice guy and I have a lot of time for him and to see him get fired over a bullshit tweet about Windows Phone 7 and Nokia?
They should be firing Brandon Watson down, for failing to market the stupid phone not firing the guy who spent most of his days highlighting the positives associated with the product?"
Joe Marini did an ill-advised tweet about Microsoft's most important partner. I'm sure Joe is a great guy, but if you ever want to stop leaks, this is what you have to do. I get the feeling that he likes Joe and would keep him. When you're running the show you have to fire people you like.
And fire Brandon Watson? Last I saw he was not head of consumer marketing, he was head of developer marketing. And while Windows Phone hasn't caught on at retail, it's doing a really good job with developers. Again, it sounds like he has a personal beef with Brandon and would fire him based on that. The marketing problem with WP isn't developer facing, it's consumer facing.
I found the bit on Brandon particularly jarring - Brandon runs developer marketing and is doing a really good job of getting developers onto a phone that people aren't buying much. Just look at his activity here on HN or him getting Scott Adams to try out WP or how responsive he is on Twitter/email to any developer out there.
I empathize with a lot of what Scott said (though I'm not sure terrible marketing sites were the core problem Silverlight had) but he's off the mark with this one.
"Now I just had to figure out how I was going to tell me wife that we are about to move to Seattle as I should point out, at this point I hadn’t told her I was interviewing for the gig. Pack up your life and move to Seattle.
I eventually wore my wife down who just gave birth to our second child Emily, so there I was organizing my entire family to move from Australia to Seattle. My wife had never set foot in the US in her entire life, and here I was dragging her away from a promising career at Ernst & Young to work for $20k USD less than we were earning in Australia (Not including her wage either)."
However, Silverlight's underlying technologies are still alive and well. Not much consolation to MSNBC programmers or whoever, who develop interactive Silverlight websites, but hey at least their skills directly translate into Windows/Windows Phone.
Will browsers be able to keep the innovation up without plugins as a playground for testing new ideas?
His actual complaints about the Microsoft group he was in can be succinctly summarized:
. Politics
. Lack of Strategic planning
. Difficulty executing
All reasonable points for nearly any Microsoft group, but I have to wonder why it is he feels such a determination to burn any possible bridges he may have left behind.
My own read on Silverlight is actually all strategic problems -- unfortunately a group that could have been working on something like V8 with awesome UI enhancements for metro three years ago spent their time competing with, essentially a dead platform.
The major break in architecture with Metro is a gamble, but not a bad one. It's an aggressive, fairly fast gamble from a company that's gotten way too slow over the past five years. I think we'll see better product coming out of MS over the next three years if they can keep this up.
Frankly I blame Steve Ballmer. The buck stops at the top. He's not a technology guy. He's just a middle-manager who happened to be in the right place at the right time.
Compare this to Google, where I work. I'm just a software engineer with no reportees but there are precisely four people between me and Larry.
Take any software engineer at MSFT and find out how many people are between them and Steve. Frankly, I think that'll be a telling (damning) statistic.
I refer you to the (excellent) My first BillG review [1]:
> In those days, Microsoft was a lot less bureaucratic. Instead of the 11 or 12 layers of management they have today, I reported to Mike Conte who reported to Chris Graham who reported to Pete Higgins, who reported to Mike Maples, who reported to Bill. About 6 layers from top to bottom. We made fun of companies like General Motors with their eight layers of management or whatever it was.
I was always impressed with the quality of work that they were capable of, but it was never clear to me how to involve them in the actual business of building products. To wit, I can't remember a single member of product marketing present at any product review up through the senior-VP level (which, after Bill checked out, was effectively the top of the management chain, unless you wanted to do something cross-company such as include Office for free in MSDNAA).
I'm a software engineer with no reports at MSFT; there are 5 people between me and Steve. YMMV.
The problem with Microsoft's org chart is its width, not its height. The best metaphor is "warring city-states", with a large helping of "the left hand does not know what the right hand is doing". Each Microsoft product team or department has the ambition and manpower to believe that they can build a world-changing product - after all, this is Microsoft! We can make a new, better standard for web content if we want to do that! Go big! At the same time, no individual product team has the official authority to quash other projects that might duplicate, negate, interfere with, render obsolete, or drain resources from its own project.
The key word is "official". When a decision has to be made about a troubled or behind-schedule project, the more-important divisions will typically get their way. Software development being what it is, there will always be several troubled or behind-schedule projects in a company the size of Microsoft. Then you have to sit down, compare future plans and bug lists to schedules/budgets/available developers, and decide what to do about it. It's in those meetings that Microsofties discover some product teams are more equal than others.
This is what the original article sounds like. There's an ambitious product, yet it suffers for lack of attention and resources. Some developers are overworked (their project is too large) while others are bored by inaction (their project is too small). It suddenly comes to the attention of Team Foo that Team Bar has built a system which does not use Foo, even though Foo is supposed to be Microsoft's all-purpose framework for your Bar needs (Team Bar did not know about Team Foo and vice-versa). The problem is recognized, yet nobody actually seems to be in charge of deciding what to do. Finally there is a come-to-Jesus moment in a "corporate realignment". Foo: Yes or no? Microsoft retroactively discovers that actually, it's OK with Foo not taking over the world. The project ends, often with the notable developer/evangelist/whatever in charge saying that they were very excited about the Foo technology but it seems Microsoft didn't agree. Some level of bitterness is expressed at the fact that Microsoft never said so up front.
Google are next, I thought you would see that when you link to an old post where Microsoft talk about being lean from your own post that talks about Google being lean
Start with one person in charge of the company. Suppose he has ten assistants, and each of those assistants has ten managers they coordinate. Suppose the managers have ten employees each that they interact with, guide, and coordinate. So for a thousand-person workforce, the company has three layers of management in this model.
1. It’s entertaining if you are curious about turf wars in Microsoft
2. He mentions a boss he detested launched http://thetirefire.com/ – timing out for anyone else?
† "Leaking IE9 secrets?" Posted just now, sept. 2011, with IE 10 in public beta.