Wavbreaker-thp merges

Some people don't think too much of forks. I think the absolute world of them. I think they help everything in the day-to-day in keeping software moving and becoming more Free. Well, the news is that Wavbreaker's author moved wavbreaker to sourceforge, in a huge opening and it opens it up to a community effort, which before was much more a mono-developer effort. The results are beautiful; Wavbreaker has had a release with the newly merged code and it's much faster, better looking, and it is much more intuitive. Wavbreaker, as all software, has it's warts, but being in the community, at least now I can blame it on myself that it's not fixed.

Gentoo's Problems

OK, I've seen everyone comment on Gentoo's problems. I'm not going to link to them all, there's plenty of opinion to be seen out there. The way I see it is that a large open source project will have 10s or 100s of conflicting personalities. The main thing to do is to keep them seperated as much as possible. For instance, look at how the Linux Kernel guys do it. Take a look at how many different "Linux Kernel" mailing lists there are. One does not depend on the other. Sure, they have to have one lieutenant to push things to Andrew and Linus, respectively, but let's keep it nice and separated. Let's not have to answer to any authority, but only do things because we're scratching our itch. And when you do have to speak to someone, try to do it with some tact. If we feel the need to disagree, let's not get personal. Try to keep things civil.

Look, I don't pretend to have all the answers, but follow the above and I can't imagine how problems persist. The main problem I see with Gentoo right now, are rude people, personal attacks, etc. I can't say I didn't see it coming.

As a Gentoo user, not a developer or anything else that's crazy, I've been personally attacked, called a liar. I refuse to post bugs on Gentoo's bugzilla. Not to say I won't in the future, but I can fix all my problems that I have on my computers, and will not deal with their attacks. Ever since then I've wanted to see Gentoo fork. Something with a better focus, quite possibly more receptive to other package managers. A distribution where nonsense isn't tolerated. I'm actually looking for something that is not such a democracy, that has a sensible leader and does not need a council, etc..

Gentoo needs competition, I don't consider anything that isn't Gentoo compatible as competition at the moment. Something like Sabayon with a much broader goal. Gentoo's goal was broad, but it's like a spoiled bread at the moment. But as I said, it sucks less then everything else out there at the moment so I'll stick with it until it's no longer viable.

Wavbreaker's fork

There's one program that I use on a daily basis, that I would say, goes rarely used. That's wavbreaker. It simply does just as the name says. Breaks wavs up into smaller pieces. I record daily to a 24 hour raw file, which I then convert into baby wavs and edit them from there with wavbreaker. It is notoriously slow for read and write, and development has stalled since a few months ago. Well, it looks like someone has forked development, fixed performance, added usability stuffs and will hopefully weave it's way back into the main branch. Of course, just like any other software I could be critical about it, like it doesn't support 4GB files as per the wav spec, but I actually give it all blessing because if it wern't around I would have to dual boot to take care of my wavs as there's no other program in the linux world that does what it does, and not nearly as efficient.

So, I hacked on it a bit today, the fork that is. I'm going to try to get some of the patches that I've been carrying from wavbreaker into the fork.

Another day, another less than optimal OSS problem

Well, today I have yet another problem in Linux, it's a problem that has happened, more or less, since I've been using Linux. Video Players.

Generally, I believe OSS has the strict advantage here when it comes to video players. We have so much technology behind our backends; they play about anything you throw at them, they consume so little memory, and they run on vintage hardware. Where's the problem? The GUI.

gmplayer - This one sucks less in the whole category, which really is a farce, because you would look at the preferences option and run to the toilet. The skins, which in general I deeply hate, actually improve it. I can make it seem to be livable in this case. The main problem I've found with it is it breaks with Pulseaudio, or Pulseaudio/ALSA plugin, unless I've compiled in support with mplayer, with an out of source patch. Even mplayer works, just not gmplayer. Also, probably the most hacky feeling out of the bunch.

xine. Xine-libs are wonderful, and they'll help fuel some of the great things that are supposed to be happening in KDE4, but xine-ui is horrible, not only that but it is documented to be broken beyond repair.

Smplayer. Obviously this developer never needed to compile it, because I never got past that with some error. Not to mention it has a custom make system that sucks horribly.

Kaffeine - Oh the suckage that is kaffine. Horrible UI, general instability. I have a hard time believing that it's an semi-official KDE project.

VLC - Compiled the latest stable, a snapshot, then I even tried the trunk. None worked for me. At all.

totem - Depending on Gnome is definitely a strike against it, but then I tried it and it _seemed_ not to suck. Then it hung on a video.

It's really horrible to say, but all of this has made me pine for the days of running Window Media Player (with Windows 98/2000).

In some of the above examples, granted, that it may not strictly be the UI's fault, or more correctly the UI may not be the cause, but the UI should definitely not lock up or crash, no matter what happens in the backend. Some of it maybe even a distribution problem, granted that Gentoo has it's warts, but I've seen some of the frontends in other distributions and they fare no better there.

Gentoo's Failure [overlays.gentoo.org]

Welcome to my blog. This is where I'm going to post about tech stuff, such as MPD and Gentoo. If you're looking for someone who puts a positive on Linux technology, this is not the place to look. I know what sucks and what doesn't; Linux sucks less than everything else at the moment, at least for me.

Because of the nature of my blogging, I'm not going to attach myself to any 'planets'. Enough said.

Now, I've been running Linux for about 7 years, it's been a really good 7 years. Some things still suck, but I've tried the alternative and they suck more.

I've been using Gentoo for about 4-5 years, everywhere. It sucks less than the other distributions at the moment. I would like to see it grow past what it is, but I am not crazy. I do realize it has a target audience, I know it will never get past. Now, onto my topic.

Gentoo's overlay feature is genius. It allows anyone to update their own packages, with minimal effort. It allows anyone to easily share their changes, by posting it wherever on the internet. And best yet, it encourages ebuild improvements.

My problem? Not with overlays, but with the site http://overlays.gentoo.org. Somebody at Gentoo had the idea to start a centralized place for Gentoo developers to host their overlays. It has so many problems, but I'm going to try to list them here, as I haven't seen it posted elsewhere.

1. Exclusivity - It would have been brilliant to find hosting (which wouldn't be that hard imho) and make a community overlay site. Ebuilds take up little to no room, and hosting would have not been hard to find. Of course, in this case you'd have to trust the source, but we all do that on a daily basis nonetheless. Gentoo developers already have an overlay, the main Gentoo repository, I don't know why it's such a bad thing about giving the rest of the people some free hosting for theirs.
   
2. Navigation? - Just take a look at the site. It is impossible to navigate. Impossible to find what you're searching for.
   
3. The RSS feed does not give good links.
4. Trac? - Now, I love trac. I have tried to talk some people into getting it to be the MPD default, it's a great software, but it is either not up for this monumental task they're trying to make happen with it, or they didn't correctly set it up. Seems to me there are better solutions they didn't try to find. It would have been better having a little more thought put into this.
   
5. Searching - Yes, it's a subsection of navigation, but it must be mentioned, no search engine?

Don't get me wrong, I found free hosting for my overlay long ago, and I wouldn't like the idea of changing it, but bringing other Gentoo users into the fold would have been good for everyone, the project especially.

I have never seen something so mismanaged, something that was such a great idea, that was implemented so horribly. It would be better having nothing than having overlays.gentoo.org.