Ask Matt Asay About Ubuntu and Canonical
A week after the announcement that open source advocate and blogger Matt Asay is leaving Alfresco for Canonical, in the role of COO, Matt has agreed to answer your questions about his role at Canonical, his vision for the future of Ubuntu, or the prospects for open source as we begin to emerge from recession. Usual Slashdot interview rules apply. (Disclaimer: Matt is on the board of advisors for Slashdot's parent company, Geeknet.)
In the 21st century, why is it that we still don't have a simple, user-friendly tool to help both home and enterprise users to migrate their existing documents and settings while performing a Linux install?
http://blindscribblings.com - Tasty pop-culture in conceptual fashion.
>(Disclaimer: Matt is on the board of advisors for Slashdot's parent company, Geeknet.)
Revealing the interests of parties involved is good journalism. But unless the author feels this means they consequently have no obligation to objectivity or accuracy, it isn't a disclaimer - it's a disclosure.
However, given that even many diehard GTK developers seem to have serious issues with GTK, and there is some dissent over how to proceed with GTK 3 in the future, why not at least consider a future Gnome built upon Qt?
The problem is C vs C++. It pretty effectively rules out any real sharing of code bases and means that to write Gnome/Qt, you are pretty much starting from scratch. I think KDE just tried that and it was a long and nasty road. I don't think that many enough would embrace Qt/C++ to see it through and it'd never work quite the same, the danger is that you'd only get a bleak shadow of what Gnome should be and get all kinds of flamewars going.
Unfortunately,
Live today, because you never know what tomorrow brings
I'm guessing we should remove SAMBA and FAT support while we're at it. Hope you don't like to access those USB drives. Oh yeah, you shouldn't be using h.264, mpeg (of any kind) or a number of other container formats other than Ogg + Vorbis/Theora.
You're not honestly comparing apples to apples here. There are protocols and formats and codecs that aren't native to Linux true, but they are either relatively simple and feature complete or the standard is open though non-free. For example let's take H.264, both ffmpeg and x264 should be able to decode any valid H.264 stream. WINE and Mono on the other hand are trying to implement some of the core features in a vast platform, they're like constant hackjobs to bugfix and update in a neverending stream of things that don't work and each new thing comes as a "surprise".
Don't get me wrong, I've fiddled with WINE quite a lot and done git bisects to find regressions and it's extremely useful in doing things on Linux that otherwise plain wouldn't work, but I also see how much of it is stubs and hacks and unknowns that they duct tape together to make it mostly work. It's never going to get done, it'll always be a crutch to lean on never a real leg to stand on. I don't in any way think they should be removed, but I see them as very much less ideal than most of the things you mention.
Live today, because you never know what tomorrow brings
There are at least three libraries that mask the audio problems in GNU/Linux with reasonable dependencies (for a game): OpenAL, SDL and libao (and at least the first two are also available on Windows/OSX)