It's hard to believe it's 9 months since I last blogged, but in truth very little has changed with regard to the whole Silverlight situation, and I didn't want to get pulled into the whole 'Silverlight isn't dead' shill nonsense that dominates the blogs.
On a personal note, I've just finished my second contract (this time for a Swiss investment bank) where Silverlight was mandated from on high, despite the fact the real requirements for the application's UI were for simple, fast data entry and the delivered app cost way more, and is far less maintainable than if a much simpler design and more widely-used technology had been used. It's great that they mandated Silverlight - it gave me some work that helped pay the bills - but it was the wrong technology for the job, which required lots of 'reach' and very little 'rich': an inappropriate usage of Silverlight, something which I've seen far too often over the last three years.
The icing on the cake of this particular experience was hearing the same manager who'd mandated 'Silverlight across the board' refer in literally the same breath as he extolled the virtues of Silverlight, about 'the new world of iPads and iPhones and other devices'!
So here's a message to managers considering Silverlight 'across the board': if you're on some sort of back-hander from Microsoft, or have some hidden agenda for choosing a basic technology, at least do the basic research before telling the people paying your bills that you will run across all devices!
Not that Silverlight doesn't have its place. I'm still a fan of the technology WHERE IT'S APPROPRIATE, it's just that too many organisations I've worked at seem to be using it where it's inappropriate, with the result that their customers are paying far more than they need to for an application they don't actually want or need.
By the way, if you want to read some compelling arguments around where Silverlight IS appropriate I recommend you check out Jeremy Likness' excellent book on Silverlight Enterprise Application Development. Finally another book I can recommend on top of Pete Brown's introductory volume.
Update: There is also an excellent blog post comparing HTML5 and Silverlight that I forgot when I first posted this blog entry
For me, and a few other developers that I respect, Build will be 'make or break' on whether we stick with the Microsoft stack - Yup, they've pissed us off that much with their complete inability to undersand the marketplace, the competition, or how to deal with developers who help them sell their platform.
The problems with Silverlight and WPF are well-known, if not widely publicised, and exist mainly around huge memory leak problems and poor performance.
Hopefully the 'lessons learnt' from Silverlight and WPF, will fix all that in the new 'Jupiter' XAML-based technology. Let's just hope it isn't the rushed, botched job that resulted in the Windows Phone 7 fiasco or in the usual Microsoft 'smoke and mirrors' take on things they pretend that Jupiter is Silverlight 6!
I'll be kick-starting this blog over the next week as more news comes out of Build. This is Microsoft's last chance, after all the mistakes of the last few years, to redeem themselves and show they can compete with the likes of Apple, Google and Adobe. Let's hope they don't screw it up, although I have to say the signs are not great. Here's just one example (of many over the last 12 months) that indicate they're losing the fight even with their own best advocates.
Couple the different changes of the last 12 months with Steve Sinofsky's unbelievable arrogance when discussing developers or DevDiv rather than WinDiv, the effective removal of Scott Guthrie from public blogging, the departure of Ray Ozzie and Bob Muglia, and a whole multitude of other events, and it's very hard to keep faith or have any kind of optimism that Microsoft have finally woken up to all the mistakes they've made over the last couple of years.
Silverlight is not dead, for reasons Scott Barnes, former Silverlight product manager, pointed out in a tweet this morning: "If Bill Gates got a tattoo 'Silverlight is dead' on his forehead and announced it the same day..it'd still take 5yrs+ to kill it marketwise" (Actually I think he's too optimistic and it will take less than that for it to be killed because it has so few mainstream users in the first place).
But it's on life support (there's a reason why the usual high profile bloggers have all been quiet for the last six months), and a rushed Silverlight 5 RTM release to 'clear the decks' for next year's Jupiter is not the 'proof' that Silverlight has a strong future that many claim it to be, just as calling 'Jupiter' Silverlight 6 is disingenious to say the least.
If you're following Build remotely, as I am, make sure you follow folks like @MossyBlog (hilariously cynical about what Microsoft say vs what they do, but nearly always right), @TimAnderson (ruthlessly honest journalist who won't be bullied by Microsoft marketeers) and @mtaulty (Microsoft employee who is loyal to his employer without resorting to being a shill and has a knack for unemotionally getting to the heart of what's important). Amongst all the noise of the 'wanna-be MVP shills' (or just the plain naive/stupid), their analysis of the next 5 days announcements will be essential reading.
I'll also be making my own quick comments on my new Twitter account of @IanSmithUK (Yes, @irascian is gone for good :-)).
To quote a line from one of my favourite movies: 'Fasten your seat belts. It's going to be a bumpy week.' (last word changed to make more appropriate! ;-))
Footnote: Can't help quoting my friend Dave Evans who describes Jupiter as 'a gas-filled planet completely inhospitable to human life' ;-)