Edit: I've locked the thread from further comments. The conversation got out of control with such negativity that it ceased to be helpful to those involved or the community.

Niall Ginsbourg posted this the other day…

“…unless you’re after some pretty specific (and less than useful) changes offered in the Win7 incarnation of this SDK – my best advice to developers would be to completely give this SDK a miss – and instead revert back to Vista Media Center SDK /along with Vista Dev platform (if you do plan on persisting with Media Center development).”

Bad advice. I mean really, really bad.

Windows 7 is generally accepted by the industry as a whole to be much better than Windows Vista. By the time all is said and done it will sell loads more copies and be much more prevalent than Windows Vista. The Windows Media Center platform has quite a few improvements for Media Center Markup Language (MCML) and the Managed Code Object Model which gives you the most seamless and elegant chance to have a great experience.



Categories: Software Development Kit | Windows 7 | Windows Media Center | Comments [10] | # | Posted on Saturday, August 22, 2009 1:48:39 AM (GMT Daylight Time, UTC+01:00)   
Saturday, August 22, 2009 4:21:27 AM (GMT Daylight Time, UTC+01:00)
I think you have taken me well out of context there (a bit of a cheap shot really).

The entire statement I make on my post pretty much explains this reasoning - which I'll repeat here :

1. Write an app with Vista MC SDK and it will work on BOTH Vista AND Windows 7 Media Center. (My Advice)
vs
2. Write an app with Windows 7 MC SDK and it will ONLY work on Windows 7. (Your Advice)

Which one would give you greater reach/audience - and hence what makes commercial sense to you today? What about in 6 months time... and then another 12 months time..? The answer is still the same - you'd be crazy to dump the Vista audience for no additional features of value.

The question is not about which 'platform' is better - quite obviously Windows 7 is a better 'O/S' - it's just that the new MC SDK does nothing to take advantage of it.

The 'improvements' to the 'managed' code object model in the Windows 7 MC SDK are so minor and hence essentially 'useless' in context of those writing managed applications for Media Center. As you know - personally I have 4 x Managed MCPL applications on the market. Apart from the 'bugfix' regarding the screensaver on extenders - there is really nothing at all that the new SDK offers me over the older one. That is - I cannot find a single compelling reason why I would want to recompile my apps under the Windows 7 MC SDK/API (and alienate the Vista audience) - and I can't find a single compelling feature being offered to me in Windows 7 MC SDK that I'd want to take advantage of (and release a new version for).

If I'm missing something there please tell me - but having 'helper' classes for problems you can already solve in Vista MC doesn't provide me a compelling case at all. Quite obviously - noone there wanted to go near any of the MCPL 'rendering' side of things (there isn't a single change offered here in the Win7 MC SDK).

If the 'new' SDK actually had of taken advantage of the 'new' features that had evolved since Vista was released - then it would have been a very different story - for example :

- Multi (and Single) Touch UI support
- Proper support for v2 Extenders (ability to query Capabilities, Codecs, Profile etc)
- Fixes to all the v2 Extender / MCPL related bugs
- Support for the completely changed underlying EPG database / PVR scheduling system
- Hooks into the new UI functionality
- First class support for Silverlight (the only UX getting full attention at MS)

... and then there's all the things that *should* have just been in the Vista SDK to begin with - like support for capturing/using the 'more info' button on the remote (and right mouse button)...

I do hope that *if* there is a Media Center in Windows 8 and a SDK/API created for it - it will expose and utilize new platform features introduced in Windows 8 - and not just be there to 'catch up' with features introduced by a Vista or Win7. (even though right now the SDK is now a full O/S behind where it should be - so that will be needed too)

Also - I'm very curious to see what actual features of this new SDK are going to actually be used by the 'online spotlight club' members out there (obviously the only recipients of any love with the Windows 7 MC SDK were 'MCML Web Apps') - and if they will even actually start making MCML Web apps again. (the last few new '3rd party' experiences we have seen in Media Center seemed to be all flash, Silverlight and WPF).

Anyhow - I invite your readers to read my full blog post in it's full context (thanks for the link) - and make their own educated decisions.


Saturday, August 22, 2009 5:33:51 AM (GMT Daylight Time, UTC+01:00)
I think the disconnect for you *personally* with regards to the Windows Media Center platform in Windows 7 has to do with the fact you don't write web applications. You write local applications which might leverage the internet -- but that's a different thing altogether from writing applications which run on a web server to deliver an experience to the customer.

I tried to set expectations accurately in my original blog post about the platform for Windows 7 -- you can read it at http://blog.retrosight.com/WindowsMediaCenterPlatformInThePDCBuildOfWindows7.aspx. The summary: Small and targeted for web applications. Your personal expectations continue to be higher for some reason, as though we promised something larger and then backed out (we didn't -- it's been plainly stated since October 2008).

It's as though you can't move past the fact you are disappointed.

Maybe my first paragraph in this comment is the issue: You are primarily a local application developer (at least for Windows Media Center) so I can see where you might be disappointed with the paradigm we chose to focus on for Windows 7: web applications. As a long time desktop developer I can sympathize -- the transition the software industry is in towards the web can be painful.

You've also seemed to believe the resources for platform were MUCH bigger or more important as a whole for the Windows Media Center team than they were in reality. The scope was largely not in our control -- we had to make tough choices given the resources. At the end of the day, I still believe we made the right choices even though it alienates some developers.

Am I just way off base here...?
Saturday, August 22, 2009 8:42:07 AM (GMT Daylight Time, UTC+01:00)
I have to agree with Niall here. If I take advantage of the new Windows 7 features in my applications, I will be either dual maintaining my applications for Vista and Windows 7 (which I tried and found too onerous), or I will just support Windows 7 and loose a lot of my user base. Not everyone is going to upgrade to Windows 7 in October.

The features in the Windows 7 SDK that are mentioned in your blog post are really in the category of bug fixes for issues that the community have found ways to work around. If they were in Vista, I would use them. Since they aren't, I'll continue with my workarounds so that I can support both the Vista and the Windows 7 user base with a single code line. It's a no brainer as far I am concerned.
Martin
Saturday, August 22, 2009 8:46:48 AM (GMT Daylight Time, UTC+01:00)
I actually do write a lot of web applications (have been doing so commercially for 13 years now - since the days of ASP 1.0) - just not for Media Center - and not Media Center Web Apps for 'content aggregators/providers' (as unfortunately all the ones I do deal with don't want to be in the Media Center space).

Since I'm not a content provider (nor are the 90%+ of other MC developers who use the SDK) - the space which I can provide value for Media Center users in is writing locally running 'managed MCPL' applications - which aren't constrained by a sandbox. I presume most of my readers who develop for Media Center are also in similar place (aren't content providers and are not developing online spotlight content) - so my advice is targetted at them. However - it still actually applies to MCPL Web apps - because as mentioned - there's very little offered that you couldn't already do using another technique. (ie : Web Service calls could already be done without the new Data classes in Vista if you used CGI parameters and dynamic MCML / Host element).

I agree you 'set some expectations' in October 2008 (2+ years after we were asked for feedback on MCPL) - and I similarly 'expressed my disappointment' at that same time via my blog - although pretty much 'held back on any final/difinative advice' as I still had some hope for change (and Windows 7 had only just been revealed to me). Since then - quite a few more things came to light - for example such the massive disconnect between the new EPG/PVR database and functionality and the SDK. (this was not fully realized until after the MXF related specs were released in late February).

I still believe it was somewhat presumptuous (and almost insulting) to all the developers who provided their feedback on MCPL when it came out - to not get MS first real response (or intentions) - until AFTER the SDK was 'feature complete' (PDC build). There was no 'feedback loop' in that - and hence no real opportunity for any developers to have any say in the Windows 7 MC SDK (as no information was revealed about what would actually be in Windows 7 to take advantage of before then either).

So after seeing (and fully understanding) the massive advances with Windows 7, Silverlight v3, 7MC and other areas - and then NOT seeing a similar set of advances with the 7MC SDK (and getting a complete picture on what 'could of been') - I've blogged my 'updated' thoughts (and increased level of disappointment). It's not about me 'not being able to move past disappointment' - it's actually me being able to clarify and quantify why I am (and sharing those thoughts on my blog for those who care to read it).

I appreciate you were under resourced - but it's where that 'reduced' effort was focused that was the disappointment for me. I am still yet to understand why these 'tough choices' were made (to only satisfy MCPL Web apps with bells and whistles - and ignore the majority of what MC developers wanted) - as I am obviously yet to see any actual beneficiaries from these decisions (perhaps there's a whole lot of mcpl web apps about to come out I just haven't seen?).

ps - thanks Martin - just saw your comment. Glad to know I'm not going crazy... 8)

Saturday, August 22, 2009 10:17:25 AM (GMT Daylight Time, UTC+01:00)
Niall,

You are not crazy. I completely agree with you. You would have to be mad to write an MCML local app that targets 7 exclusively (at least for the next year). In fact you have to be a little mad to be writing any MCML based app in the first place.

To think that Windows 7 is the first time us developers can fully leverage the "azure initiative" is both naive and mad.

So here I am reading this blog post, and it reeks of the sentiment that: Windows Vista ... that is sooo last year, forget about it. And there is plenty proof of that, you go and release KB967632 which breaks the stop button on Vista and go ahead and completely ignore fixing your regressions.

Sure, if the story was:

* You have a complete Managed API that is not XML based that allows you to do everything (hook in to remote keys, set repeat rates, load images from memory, build UIs programmatically because your crazy markup language has an exposed underlying object model).

* You have complete interoperability with WPF and Silverlight (Deep Zoom) (parts of the screen can be rendered with WPF).

* You have a GUI that can be uses to design UIs instead of pushing angle brackets

I would agree completely with your sentiment, but as Niall said, the changes are not ground breaking, its just a bunch of overdue bug fixes.

The argument that the team is tiny is pretty weak, look at what Phil Haack managed to achieve with his tiny team in the last couple of years.

Sam
Saturday, August 22, 2009 5:08:07 PM (GMT Daylight Time, UTC+01:00)
Somewhere along the way I failed to set expectations for the three of you (and perhaps others) that Windows 7 was going to be much more constrained than prior releases.

There are some rather nice improvements for application developers in general if you will take a closer look and figure out how to make use of them. Case in point: Niall is correct that web apps could exist on a richer scale in WIndows Vista using 'CGI parameters and dynamic MCML / Host element'. While possible, it certainly wasn't apparent to web application developers -- and the number of contortions you have to go through is hacky at best because the approach was never intended. It works but thats literally just because of the luck of the draw for how MCML works in general and not some intentional design. The new Data Access Model Items provides a much better, well thought out approach to the same problem.

You guys really need to take a *positive* look at the platform in Windows 7 rather than armchair quarterbacking and complaining ad nauseum about your personal disappointment. You aren't doing the community of developers any good continuing down that path. At some point in the future there will be more users of Windows Media Center on Windows 7 than Windows Vista. The early adopters of the platform for Windows 7 will have an advantage over those who choose to stay only on the Windows Vista platform alone.
Sunday, August 23, 2009 1:01:35 AM (GMT Daylight Time, UTC+01:00)
Charlie,

What community of developers? There are 268014 questions on stackoverflow and barely 4 media center related questions. The community is tiny, there are probably less that 20 devs doing this kind of development.

http://stackoverflow.com/search?q=mcml
http://stackoverflow.com/search?q=mcpl
http://stackoverflow.com/questions/tagged/mediacenter

There are barely 5 posts a month on sandbox at the moment.

You didn't even build us an app store, so getting penetration into the market is just too hard. There is no question that Big Screen products / MyTV / Media Browser and OML are the most popular plugins for Media Center and none of us are particularly pleased with the direction this is taking and the speed of change.

You don't go and complain about your customers because they are unhappy about your product, you go and fix your product.

Sunday, August 23, 2009 1:50:19 AM (GMT Daylight Time, UTC+01:00)
It's not about 'positive' or 'negative' - it's about 'good' and 'bad' advice. Telling developers to start coding and releasing products which are 'Windows 7 only' without there being a tangible benefit - and alieninating the large % of the target users still on Vista is just 'bad advice' (and would be commercial suicide for them).

The advice I gave follows the exact same reasoning that dictates that it would be really bad commercial sense if Microsoft released 'Windows 7 only' (or to be fairer Win7+Win2008 only) versions of Internet Explorer 8, Silverlight v3 SDK or upcoming MS Office 2010/Visual Studio 2010 products (just to name a few).

Maybe - just maybe in a few years time when/if Windows 7 MC has large penetration (ie. 85% or more of users are on Windows 7 instead of Vista) - only then would it make sense to consider developing against the new SDK.

Large Windows 7 penetration/dominance is clearly not the case 'today' - and developers wanting to release solutions 'today' (or even in a year) - who don't specifically need any of the minor improvements to the SDK - should stick with Vista SDK. (and get the best of both worlds). Those who want to target a 'broader' market and get some potential IP re-use and portability beyond the walls of Media Center and want a much more sophisticated UX developer platform - should also consider other technologies like Silverlight and Flash for better ROI and Time to Market. (they may not give you the optimal deployment/integration solution for Media Center - but unlike MCPL - portability to and from other 10 foot , 2 foot and mobile platforms will at least be a possibility). I think this is likely the exact same commercial reasoning behind decisions to not use MCPL for Netflix and the Flash Based Sport's channel in Media Center. (two of the most recent non MS produced products in Media Center).

In this blog post (and comments) 100% of a small amount people disagree with your statement (and until I see people actually supporting/agreeing with what you have said and providing valid reasoning - I can only conclude that would I've suggested is right).

Whether or not certain usages of the Vista MC SDK are 'intended' or not 'intended' and considered 'hacks' etc are also not valid points. The MC SDK provides a set of raw primatives for developers to piece together to produce solutions (and little 'real world' examples are given on how to use it). Being inventive and innovative with those building blocks - and realizing implementations that aren't immediately 'apparent' is the very essence of succesful development with MCPL - and those that can't do that will ultimately struggle with the platform.

It's not about 'setting peoples expectations' - it's what is and what isn't. We aren't 'armchair critics' either - we are part of the core user base on the MC Dev platform - and I know firsthand that both myself and Sam not only have well known products out in the marketplace (that would probably have a much larger headcount of 'serious'/'regular' users than the MC SDK) - but we have spent huge amounts of time developing (and struggling with) the SDK. If people like us aren't considered 'qualified enough' to make educated comments about MC development - and our input is labeled 'armchair quarterbacking' and 'ad nauseum' - then I'm not sure who is.
Sunday, August 23, 2009 1:58:27 AM (GMT Daylight Time, UTC+01:00)
>> You didn't even build us an app store, so getting penetration into the market is just too hard. There is no question that Big Screen products / MyTV / Media Browser and OML are the most popular plugins for Media Center and none of us are particularly pleased with the direction this is taking and the speed of change.

Yes exactly. Not only that - but last year when MS bought out TGB they removed the google adsense (our only real way of targetting the MC userbase) - but didn't provide any mechanisms or places for commercial developers to tell people about their products. Thank god for the other places like TheDigitalLifestyle and Australian Media Center community site and the efforts of those and other bloggers (although unfortunately none have the same audience or global reach of TGB)..

>> You don't go and complain about your customers because they are unhappy about your product, you go and fix your product.

I couldnt have put it better myself - you nailed it with that comment.
Sunday, August 23, 2009 3:38:52 AM (GMT Daylight Time, UTC+01:00)
Amazing.

At no time did I ever say develop exclusively for Windows 7 and not pay attention to Windows Vista.

I think this conversation has jumped the shark. It's so much now about what we (I) didn't do for you as to not be helpful to the community.
Comments are closed.
RSS
RSS 2.0
Search
Sign In | All Content © 2014 Charlie Owen

This is a personal weblog. The opinions expressed here represent my own and not those of my employer.


Powered by newtelligence dasBlog 2.3.9074.18820