wingolog

to guadec!

3 August 2011 9:19 PM (gnome | guadec | gnu | guile | javascript)

I'm very happy to say that I'm going to Berlin on Friday, for GUADEC, the GNOME conference.

This will be my seventh GUADEC, and my first as an Igalian. I'm very much looking forward to seeing old friends, as well as to meeting new folks.

I'm especially looking forward to soaking in the energy of GNOME 3. GNOME 3 is the best user-facing project that Free Software has going right now -- it's a dynamic group of creative, talented people, making users' experiences with computers beautiful and natural.

The only other Free Software project that is focused on users in the same way is Android, and it is its own corporate kettle of fish. I hope that it's GNOME that ultimately wins, against the hoarders of the software world, from the workstation down to the phone, because GNOME brings with it our participatory culture of software as a means for human emancipation: that we can work together to make a better world. We've been working on this Free Software thing for some 30 years now; we're not a fad, or a hype: we are people working together, making our world. We can win, and humanity deserves it.

. . .

This year GNOME is again sharing a conference with KDE, another "desktop environment", to use the outdated term. I have pleasant memories of hanging out with KDE folk at the last joint conference we had, but I don't think it's healthy. Last time we were together because it was convenient for Nokia. Now that Nokia sold us both out, it doesn't make sense.

I do wish we could collaborate in meaningful ways -- and we do, in lower levels in the stack, and partly at the window manager level -- but we're working in the same space with two different cultures, and colocation ultimately dilutes the impact these conferences can have among the members of these two projects. So my KDE friends, let's have one last fling this weekend in Berlin, and thereafter see each other in Brussels, at FOSDEM :)

My seven GUADECs marks me as some kind of old fart now, I suppose, and it's true that my own work with GNOME is mostly in the past. I used to work on GStreamer, but don't anymore; Guile is mostly where I'm at, free-software-wise now. I wrote some bindings for the GNOME libraries back in the 2.x days but have not had the time to update them to the modern 3.x world. I fear that I'll just have to pass that project on, though there does not appear to be much hacker interest in maintaining API compatibility.

Anyway, as I said, I mostly work with Guile now. If you're interested at all in Guile, please pull me aside, and I'll be happy to talk your ear off ;-) I've also been doing work with the V8 JavaScript implementation recently, for work, and I'd also be happy to talk about that: compilers, optimization, and what-not. But kids, remember: JavaScript is a fad. Guile is the official scripting language of GNOME.1

See you in Berlin!

1 ;-)

7 responses

  1. Denis Washington says:

    Great to hear that, I'll hope we'll meet - I admire Guile (which brought me to Scheme) and your work on it! A nice chit chat about Guile, the state of Scheme and the R7RS standard (yes, I'm the one who recently sent that exhaustive commentary messages about the third draft to the scheme-reports list) would sound like a great deal of fun. :)

  2. jospoortvliet says:

    Heavily disagree on the "doesn't make sense" of the shared conference. KDE and GNOME didn't do the first DS just because Nokia wanted it. It was done because it makes sense. There is 95% overlap between goals, technology and culture of KDE and GNOME - no matter how hard people claim there isn't.

    As far as I am concerned, projects like Scribus lead the way, embedding Clutter in a Qt4 application. Intel's doing good work with MeeGo too and I applaud the efforts by KDE's Oxygen developers on oxygen-gtk. SUSE works on a cross-desktop MIME stuff and there's more going on. Far more useful than further polarizing.

    You compare GNOME to Android. Ha. Saying it's a David vs Goliath is still too much of a compliment, GNOME doesn't stand a chance. What's GNOME's marketshare? Too small to count, huh? Oh, and the same goes for KDE of course. If you ever want to get anywhere, MAYBE you MIGHT have a chance together. Otherwise, if you're fine staying in the margins and enjoying the lack of attention there, keep fighting.

    And if you want to blame the sponsors I can only speak for my company (SUSE) but I can tell you we're sick and tired of having two different teams doing as much duplication as they can, arguing and fighting all the time. I'm quite certain the KDE-vs-GNOME is a major reason why Linux never got anywhere on the desktop and it's silly. And a waste of money, something I'm not prepared to pay for or support. And I'm quite sure I'm not the only GNOME/KDE sponsor who thinks this way. Canonical has already set some steps in this direction, Nokia and Intel don't care about the GNOME vs KDE sillyness etc etc.

  3. Andy Wingo says:

    A pleasant surprise, Denis. Let's do meet up!

    Thanks for the note, jospoortvliet. You make some good points, though I disagree about the arguing and fighting: there isn't much of it, because there simply isn't much to say. GNOME and KDE collaborate very well on the lower levels, but ultimately success is about making an impact on user's lives, and that is mediated by the user experience. Free Software on the whole is enriched by having two good communities here, but it does not follow that GNOME and KDE should share a conference.

    Like I said, I am looking forward to this one, but I hope that in the future, GNOME will choose to use its main conference as a means to create software that reaches users, not as a means to network with other underdog projects.

  4. jospoortvliet says:

    It's all fine to focus on user experience, and in that regard it's fine that GNOME and KDE keep their separate way. Obviously, if I had my way, the two would be integrated but I realize that won't happen until it gets really really cold in a very very hot place.

    Until then, the amount of collaboration is still far less than it should be and that's a main reason why both projects can't focus as much as they should on what they (should) do best: the user interface.

    Nepomuk, tracker and zeitgeist work together but their respective developers would have a VERY hard time convincing me GNOME and KDE are better off with the separate technologies than one well thought-out, integrated whole. Esp here, the different focus and history (cross-university research project vs practical, commercial-led quick search) makes that each has it's own, specific downsides. To me, the collaboration feels too little and too late, to be honest.

    Just reading about libfolks, evolution-data-server and Akonadi. Seriously, Akonadi was developed to be a cross-desktop, cross-platform solution for all kinds of PIM data caching, scalable, easy to use, bindings for any language and platform. Still I doubt somehow that those developing copy-cats in the years after Akonadi was written ever had a good chat with the Akonadi devs. And that happens again and again in all kinds of area's.

    I get that someone wants to develop a webcam app like Cheese for KDE although I still think it's silly and needless duplication. But the underlying PIM stuff, which is extremely hard to get right (proven by the fact that nobody got it even remotely right yet) is just plain NIH, if you ask me. Sad.

    To add insult to injury, there are still people in both camps who have no problem to loudly claim they 'hate' the other camp. They should get a medal for their honesty, then be kicked from both communities and fired by their boss for being stubborn idiots.

    Unfortunately many of them are not only incredibly stubborn and short sighted but also brilliant developers and I don't expect things to change in a positive direction. However, I hope that as Gold sponsor for the DS and with my own personal involvement in the organization SUSE can help close the gap a little. And I hope and call upon other distributions and companies as well as all individuals involved in either or both communities to do the same.

  5. Kefas says:

    Wait a minute, jospoortvliet! Isn't that also valid for the different distributions? I mean I think the number one complaint you hear first about using GNU/Linux is the number of flavors of it out there. This is overwhelming for newbies who already on their first day must investigate and choose one out of the myriad of unfamiliar distributions. This is why your comment sounded to me like the typical newbie complaint "Why don't they just make one Linux?" I really can't imagine your company (SUSE), RedHat, Debian and PCLinuxOS all merging into a single entity and producing only one distribution. Because these are the colors in the free software community which make me grateful. We can pick and choose. And ever since I've gone all-out in using free software, I realized that I became more picky about everything in my computer (the only side effect of free software). So I like the fact that I can choose between LibreOffice and KOffice, Opera and Firefox, Amarok and Rythmbox, etc. And IMO, SUSE is doing a great job satisfying both GNOME and KDE lovers. Sure, everything would've been much simpler if there's only one option for everything, but I don't want to go back to the no-option world (Windows).

  6. saulgoode says:

    "I fear that I'll just have to pass that project on, though there does not appear to be much hacker interest in maintaining API compatibility."

    I am not particularly interest in the Gnome stack, but I am interested in creating Guile bindings to other libraries. However, I am at a bit of a loss as to how to go about it. I think I have the current state of snarfing figured from your work on the Gnome GTK module (and it doesn't look like it's changed much since the SCM days), but I don't know where things currently stand with regard to 'smobs'.

    The documentation available appears to be in need of updating and while I've had some success with creating a few language bindings for a particular task, I don't know how to approach a large project (especially in an "approved" manner).

    If you find the time after your return from Guadec, I would greatly appreciate any information you could share on this.

    Regards.

  7. jospoortvliet says:

    Kefas: Of course it's true about the distro's, and that's why openSUSE is working so hard on cross-distro stuff with others like Fedora, Mandriva, debian etc.

    I wouldn't claim there should be ONE desktop, because there ARE differences. But they're mostly about the final UI, design stuff and the like. We could eliminate lots of duplication without loosing identity, I'm sure. Why would GNOME not have a dependency on certain Qt things? Why would KDE not incorporate certain GNOME things (ok, it does already, but more).

    Choice is good but not for the sake of it.

Leave a Reply