Alastair, It was interesting to read your thoughts as these kinds of letters were ones i use to crave back when I was in my ivory tower in Redmond :) that being said, I've read this kind of thing before and I guess you can take some sense of positivity that you weren't the first and definately not the last to write this kind of thing. Sad part is the fact you guys have to write it in the first place.
There are and is a number of challenges facing the WPF on-ramping: Migration. ~~~~~~~~~~~~~~~~ Just like with any technology getting 6million+ .NET dev's to embrace anything new faces its initial challenges - mostly around the acceptance of the "new" but that aside, when a company like Microsoft expects people to move forward then there is a tax to pay, the first being is your developer satisfcation levels (scores) will drop (as confidence gets lowered given people have to learn) and lastly you need to invest heavily into providing these folks bread crumbs to follow into the new world. I mention this as inside Microsoft you're goaled on these types of metrics namely around Product and Developer satisifcation levels and once these scores drop, people freak out which in turn creates this chaotic effect where in many ways each subsidary around the world says "I have the solution" and then floods the interweb with the said solution. This in turn creates churn for developers like you as not only do you have these initial on-ramping issues around change and new but now you have to seperate signal from noise. Furthermore given the beta life cycles of Silverlight/WPF over the years you also have to seperate old/depreciated vs new and so on... to be blunt, its a freaking mindfield of stupidity at the moment. However, thanks to places like Stackoverflow and in parts social.microsoft.com forums you can find some salvation given the life span of the products (ie it's fairly well seeded in to developer market but by no means is there one definitive site that answers the most basic problem). The stark reality is, I know WPF/Silverlight better than most given i've had initimate knowledge of it since its birth but there isn't a day i go without cursing the technology in a cubicle somewhere - most recently is my experience in trying to make LINQ to Entities work the way I expected it to in a WPF solution. Maturity. ~~~~~~~~~~~~~~~~ The overall maturity of WPF developers is low whilst the ubiquity of WPF is high. Silverlight has high amount of developer base but low about of ubiquity. I raise this point simply because Microsoft asked everyone to really focus on Silverlight so well everyone kind of did, but in doing so it shifted the developer base away from WPF into Silverlight which in turn creates this supply/demand issue world-wide around WPF. Having this deficit of WPF mindshare creates an issue for future on boarding, as in turn you want a healthy amount of developer base to talk about their problems / success online in order to help pave the way for new blood. The reality is though you're more likely to yield answers in Silverlight than WPF if you're not looking in the right places that is (right places meaning you've figured out a pattern around seperating noise/signal in WPF google searches hehe). I'd say there needs to be more mature/senior WPF experts unloading into more centralized locations than the disparate ones at place, as in order to retain the Allaistair's of this world, it just needs to occur for future seeding. Shing Object Syndrome. ~~~~~~~~~~~~~~~~ Assuming you have mastered the Maturity/Migration issues above, then comes what I call the wave approach to shiny object syndrome. As again, inside Microsoft you are given commitments and scorings against weird metrics that roll up to a health report index of how investment in products are going etc (reviews of the business etc aren't fun meetings to have). The downside with this is well most staff are told to double down on the new wave, so in our case you're going to see a lot of momentum around Windows Phone 7 and Silverlight in the next fiscal than before. Now, there isn't an issue with this per say - hey embrace change - it does however create more distance between technologies like WPF and the new world order, as in reality there needs to be a backfill of staff who's job is to sustain the adoption of the old wave's (Silverlight 4 / WPF 4) aswell as seeding the new. This doesn't happen given old wins aren't as good as new wins, so less of a vicotry email bang occurs if you walk into a meeting and said "I grew last year's or year before's metrics by 20% this year!" vs "I grew this years metrics by 20%". In order to defeat this, it requires street evangelism to occur which is people on this list talking about the value propositions of WPF and helping people like the Alastair's out when they hit a wall (and hard at that). This requires a better sense of community and it also really needs in parts the help of Microsoft on an subsidary based level to help nurture and foster these relationships more (ie code camps, msdn user groups, online QA's etc). This is because you want to raise confidence levels of a new adoptee to a stage where sure it's hard to learn, but if the said person(s) feel like there are more others around them they are more likely to gravitate and polynate there learnings more in a deeper fashion. This isn't occuring :( Should you stay? ~~~~~~~~~~~~~~~~ So should you give up? It depends :D - but there is the reasons why the state of play is what is, its not perfect it's not entirely Microsoft's fault and you aren't alone. Winforms productivity will always trump WPF as well it did all of the above well - too well! :0 so you have habitual + confidence levels at an all time high. There's just no easy "How to migrate from WinForms to WPF" story being told and thus why you're confidence levels are dropping despite your maturity around .NET being probably high. Regards, Scott Barnes http://www.riagenic.com _______________________________________________ ozwpf mailing list [email protected] http://prdlxvm0001.codify.net/mailman/listinfo/ozwpf
