Older blog entries for louie (starting at number 94)

deekayen says:

I submit to you that people hacking on Gnome off in their own corners is what brings the coolest, new things to Gnome. The way I see it, the atmosphere behind thinking to create a "functional, stable release" is more focused on current features. Gnome needs both free spirits and "stabilizer" people...

Just to clarify, I don't think that people hacking off in their own corner on cool new features and such is a bad thing- this is often how innovation happens, and we need more of that, not less.

What I was worrying about (sorry I wasn't clear) was when an entire company goes off and stabilizes/bugfixes/backports on their own fork of a stable/deadish branch. [Something that Ximian has been quite guilty of in the past.] Most of those fixes end up being lost to the trunk, and the companies end up wasting a lot of time backporting that could have been more productively spent stabilizing the trunk. And the companies throw away the QA benefits of sticking closer to the trunk as well. The community could benefit a lot from that effort going into trunk, I think, and so could the companies, if it was structured right- they'd get more functionality, faster, with better/broader QA and less waste.

9 Apr 2004 (updated 9 Apr 2004 at 14:27 UTC) »

the blog tool needs a 'do what I mean, stupid' option.

Just read Havoc's post, and I do think he has it mostly right- GNOME does do it better than most. That said, some comments on his, Benjamin's, and Jeff's posts:

  • I don't see why having the Foundation employ some hackers would be a bad thing, as long as it remains a fairly small number when compared to the size of the community. It would be great for admin to be done by a Foundation employee, for example, and I think having most bugzilla maintenance done by someone with more neutral affiliations and more permanent committment probably wouldn't be a bad thing.
  • Being more transparent about where influence is coming from is not enough- companis must be transparent about where the influence is going.
  • Benjamin, remember that we are all doing this for the 'right' reasons. We're not doing it perfectly, but I haven't met anyone at Red Hat, Sun, or Ximian who isn't doing this with the best of motivations. I hope we can all be actively aware of our tendencies to look inward (as I mentioned with the UI stuff) and avoid it. There is no perfect answer, though.
  • At least for me personally, there is no difference between what I wanted as an unemployed volunteer and as a paid contributor- either way, I wanted the free software desktop to survive and improve and eventually win. Having a higher-level language to write in is (to me) perfectly in line with that- if I'm still dealing with app crashes because someone did their pointers wrong in C in 2-3 years, we'll have lost.
  • It is ridiculous to claim that Novell is writing apps in Mono as a 'carrot' for the community. We're writing apps in Mono because we think C# is the best language out there for writing large apps in, or in the case of f-spot, because the original author had heard a lot about C# and wanted to try it out. If that has a carrot effect, fine, but that's never been mentioned in either of those cases, to the best of my knowledge. (FWIW, it has been mentioned in the evo case- and you'll note that is mostly on hold until the community has more clarity on this issue.)
  • Other than that quibble, I mostly think Jeff's analysis of the language thing is pretty damn good- it certainly jives with my sense of where things should be going.
  • GNOME has never been healthier, from a code perspective, than when Sun, RH, and Ximian were all actively hacking on GNOME 2.0 and 2.2. The next cycle threatens to have all the corporate types hacking on 'enterprise' versions of 2.6 instead of 2.7, which I think would suck a lot. I'd like to see us focus instead cooperatively on making 2.8.0 be the most stable, functional release ever, and shipping that, instead of hacking off in our own private corners. But I don't get to make these decisions, particularly not for Sun and Red Hat :)

Anyway, that's one morning's ravings. NB: I dreamed about a Novell employee who snapped and shot all her co-workers in the non-existent Minneapolis office last night. It was scary as hell, and may have affected my mindset this morning :)

[Later] One other thing about Havoc's post- I don't think the thrust of my usability post was about communication, but rather about investment- it's not sufficient in my mind for the usability community (regardless of who employs them) to say 'here is what we're doing'; it must also educate about how it is being done so that others in the community can learn and add on. I've tried to do this with bugsquad; the evo team failed to do this for a long time but is now more actively trying to promote this kind of thinking, which is awesome.

9 Apr 2004 (updated 9 Apr 2004 at 13:47 UTC) »

I'd like this entry to be blank, but advo's rss generator appears to bork if you have a blank diary entry. And no one actually reads this via advo anyway. ;)

9 Apr 2004 (updated 9 Apr 2004 at 03:55 UTC) »

Had a decently productive day today- triumphed over bug 139195 (which I frankly didn't expect to do- thanks to the people who helped reproduce it) and otherwise did some minor cleanups.

Misirlou cooked up a nice, but scary, query for bugs with patches. There are over 1300 of them. 1300. Yeek. I'm hoping to do some cleanup on this list, and produce some useful queries maintainers can use to help clean up their own mess, later tonight.

It's nearly spring- had a very nice sandwich for lunch, while sitting on a bench outside. Look forward to doing more of that at some point soon.

[Later] Whipped this up for maintainers to check on patch statuses, but it still sucks- if you mark a bug needs-work, it'll still show up in that query. Bugzilla code still scares me sometimes :) Still, though, it can give you a rough idea of the number of patches in the system for you- probably some scary large numbers in there for some maintainers.

You know you're too obsessed in IP when you go to a perfectly wonderful Vienna Teng concert and spend half of it thinking about street performer protocol and the RIAA. Oh, and The Animators rocked too.

Slept in after last night's depressing Duke loss. 8 point lead in 3 minutes. Arrgh. On the plus side, a year after the trip, got my pictures from last year's trip with krissa posted, finally. And I got my taxes done. Not a bad day, at least after yesterday. Oh well- go Tech!

3 Apr 2004 (updated 4 Apr 2004 at 22:26 UTC) »

When Dave and I were in Nuernberg, we saw a robot taking pictures of some of the remaining stained glass in St. Sebaldus church. Decided to look up the robot this morning; they have a very neat site, and it is reassuring to see that somene is collecting and storing such things. The French are doing some similar things with monum.fr but Vitra (the glass scanning people) seems to be a pan-european project, which is cool.

Just read Havoc's post on usability. I don't think there is any challenge in building a closed organization that churns out usable code that happens to be GPL, which it sounds like is what RH is planning. I think the challenge (and I hope Ximian and RH's design teams can work with the community on this problem) is creating an open community where design is similarly top down- or at least where user-centric design is the dominant paradigm. We got a long way- lots and lots of people come to #gnome and say 'I want to make my program HIG-compliant.' But we sort of stalled there. The HIG covers only spacing and look and feel- it isn't a GNOME Usability Design HOWTO, and it seems like the people who could write such a doc are getting hoovered up and into the relevant companies. That's not bad- I'm happy for them, and I hope both ximian and RH can write great software based on their skills. But we won't be successful as a community if all our best people are working purely on internal projects- we need some resources working on something better than Havoc on Preferences to educate newcomers about the GNOME design philosophy and the GNOME design process. Obviously documentation can only do so much; it isn't a replacement for real training. But having something to point people at is really important, IMHO, and it doesn't sound like the direction anyone in the current usability community is pointed at. And that's too bad.

2 Apr 2004 (updated 2 Apr 2004 at 15:17 UTC) »

I was pleased to discover last night that my alma mater has purchased the public domain. Take that, Lessig.

Been a long week, but had a lot of fun last night at the release party. Good to see the RH guys, and good to hang out some with Joe and Robert, who I don't hang out with enough. Dave I see too much of. :) Nothing quite like 20ish slightly drunk gnome hackers crowding into Tosci's in harvard square. Very surreal, lots of fun. Hopefully Joe and Robert will post pics sometime tonight. We need to organize boston@gnome.org or something like that so we can do these at least monthly.

85 older entries...

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!