I seem to be one of the few program managers on the Media Center team who reads our beta newsgroups on a daily basis and I must say that I find the experience to be just a little bit frustrating. While I’m tempted to respond to every post I’d like to clear up as much as I can in one blog post and see if I can get this weight off my chest by posting some thoughts here on the feedback we receive.

Build Quality

One thing I find frustrating is seeing you all struggle with the builds we deploy to you. I certainly know there’s problems with them, I install a new build at least every week at home; but I’m paid to install it. You’re not. You’re installing it because you love Media Center. So when I see you struggle with our bugs, bugs that prevent you from really testing the software I get frustrated that we can’t be dropping you better builds. Fortunately, every build gets better and better and a surprising number of you are very patient. Thanks for your patience and thanks for your continued participation. Every bug filed helps (even if we resolve it won’t fix/by design).

We did add new features!

I read over and over about how we haven’t actually added any new features to Media Center this time around and that there is no compelling reason to upgrade from Media Center 2005. Puzzled by this I decided to pull together a list of what’s changed (not an exhaustive list by any means, we still have a few key features hidden up our sleaves):

• Support for 64bit machines
• You can upgrade to Windows Vista Media Center from XP
• Media Center is included as part of two Windows Vista SKUs
• No need to buy a Media Center. You can install yourself.
• Domain join
• Available worldwide in every locale that we ship Windows to (160 new locales! 15 new languages!)
• More content on screen in our photos/music/videos/TV libraries
  Faster perf for the music library

• Digital Cable Tuner (formerly called OCUR) support
• It was very hard to use a mouse in MCE 2005. We’ve made some big improvements to mouse handling.
• Likewise, hard to use with a touch screen before, should be better now
• Run on your Tablet PC
• New start menu to get you to where you want to be faster
• Start photo slideshow from Music Now Playing
• Now playing item on the start menu, should be more discoverable
• More ways to slice and dice your music collection
• New music Now Playing
• Way better queue management
• View photos and videos by folder or date
• Mini TV guide
• TV favorites/most viewed
• TV categories is now discoverable
• TV guide is an overlay
• Easy to get to TV categories
• Thumbnails in recorded TV library
• PAL exhaustive channel scanning
• Microsoft DVD codec
• Native burning solution
• Extender platform. Now any hardware manufacturer can integrate a MCX into their TV, DVD player, etc.
• Tighter integration on start menu for third parties. You’re no longer buired in More Programs
• Windows Media Center Presentation Layer, now you can build apps that have the same fidelity as Media Center
• Windows Presentation Foundation, re-use your Avalon code to build Media Center applications
• Hotstart

If you look at that list I see lots of features that you’ve asked for over the years: 64 bit, native DVD codec, native burning, upgradability, retail availability, better development platform, etc.

Something not on that list that we’ve likely spent more time on than anything else is getting Media Center to integrate with the Vista codebase. It was a lot of work getting Media Center to work with Vista and all the changes to the things we’re dependent on like drivers, graphics infrastructure, sound infrastructure, networking, etc. We paid a very high development cost to become a part of the Windows Vista SKUs but you’re seeing the result in terms of worldwide distribution, retail availability, etc. And yes, sadly this meant we couldn’t do other things we wanted to do like a movies library, or more TV format support, or DVD playback on Extender. All this takes time and we only have some much time before we need to release and hard decisions had to be made. Understand that we want some of the features we cut just as much, and possibly more than you do (not surprisingly many of the developers here are very hard core Media Center fans).

Who is our target market?

You’re probably wondering why there are features that you want and that we want but aren’t included. For instance the biggest request I get as the DVD program manager is about why we don’t support streaming of DVDs in DVD changers to Extenders. What we as program managers do when we get a request for a new feature or a decision change request is step back and ask ourselves how many users this impacts. Media Center is used by millions of users and with Windows Vista will likely be used by tens of millions of users. Of those tens of millions of users how many are buying DVD changers? 10,000? 20,000? Is it worth investing weeks of dev, test, and usability time on a feature that will be used by 0.01% of our user base? Or should we spend more time working on building an Extender platform enabling any hardware manufacturer to build an Extender? Or should we spend more time striving for the highest DVD playback quality? I know many of you are unhappy with some of the decisions we’ve made but hopefully we’ve made a decision that is going to have the biggest impact for our large user base. Also rest assured that there will be another release of Media Center, this is not the last release! We’ll hopefully get to your favorite feature eventually. And if we don’t, brush up on those coding skills, after all, Media Center is a platform.

What can we do differently?

While I like to think we’re already pretty progressive in dropping builds on a pretty regular basis (how often does Tivo have large pubilc betas?) and open about what’s going on (for instance our participation in newsgroups, chats, blogging, etc.) we can still do a better job. The biggest area of improvement I see is growing the window of feedback. Currently once we drop a build there is very little time available for us to respond to your feedback. In the future I would like to see us drop builds early and often so that we have a bigger window of opportunity to incorporate your feedback.

I’d also like to see increased participation of my team members posting, blogging and providing more transparency about the decisions we make so that you can understand the options we had and the tradeoffs we made. Contrary to popular belief we don’t purposely try and make your lives difficult by designing complicated start menus :).

Conclusion

Hopefully you now have a sense of where we’re coming from and I’ll keep an open mind about where you’re coming from with your feedback.

Comments(68)

Making the decision to remove features (or regress functionality as we call) are never taken lightly. For this release I’m actually the only program manager who removed whole features areas (caller ID and messenger).

Both these decisions were painful but made together as a feature team and then were communicated very broadly within the group before actually happening to ensure we were making the right decision for the product. For both cases we had very good reasons to remove these features. We removed Messenger because Windows Messenger which have a very large dependency on was removed from Windows Vista and it would have been a very large investment for our team to build our own Messenger backend functionality from scratch. And we removed Caller ID because of the high on going dev and test support cost for a feature used by a very small number of people (none of our major OEMs ship caller ID modems and those who do disable the feature in MCE because TAPI conflicts with most fax software) while third party addins do a better job.

While we really had no choice about Messenger I know that by removing the caller ID functionality our dev and test teams have been able to focus their attention on areas of the product used by several order of magnitude more users. So unfortunately I can’t point to a new feature and say, we cut caller ID but implemented XYZ, because that’s not the case. I can say that the dev who would have spent their time on caller ID is now better able to focus on delievering a solid Media Center platform for our developers. A platform which has already delievered better caller ID programs then what we built.

I’m sorry we cut these features but it’s the right decision.

Update: Check out my co-worker Charlie’s blog for his spin on this, Our Loss Is Your Gain:

So, what company will be first to take advantage and deliver experiences which put our originals to shame (using the new Windows media Center Presentation Layer Application model, of course).
Well said!

Comments(21)

Ian Dixon has a post with video about using the Windows Vista version of MCE through his XBox 360, Running Vista Media Center on my XBOX 360 (with Video):

Last night I played with the extender function of Windows Vista. This is the first time I have got my XBOX 360 working as an extender with Vista.
Using the latest build of Vista (5472) I setup up the XBOX 360 as an extender. The setup was very simple, it seems slicker that the XP version and maybe a little quicker.
Once up and running it looks exactly like Media Center running on Vista, with the new UI and sounds.

Comments(0)