JulioBiason.Net

Old-school coder living in a 2.0 development world.

When open-source fails

with one comment

When I started using Linux, around the year 2000, you could use a very simple window manager or you could use GNOME. GNOME, at the time, had this very very cute window manager called Enlightenment, which was also a royal pain in the ass to use. It required certain settings on your X to be enabled or you couldn’t use some features (like key-bindings.)

The thing about Enlightenment is that it was really nice on the eyes. Its themes were extremely good looking and there were lots to choose from. I could say that Enlightenment was the Vista of that time.

One of the things that actually did happen a few years later was that GNOME changed its default window manager from Enlightenment to Sawfish. Sawfish, although not that good looking, was way more configurable and, apparently, the author was willing to make it more integrated to GNOME than Enlightenment. No biggy, GNOME changed window manager, but Enlightenment had its own fan base, so they took separated ways.

More years later, the Enlightenment team announced the start of release 17, also called E17. They plans were big: they would use a lot of new libraries and it would be fast and you would get even more eye candy, with shadows and real transparency and real time updates on icons and such (almost what you get today using Compiz.)

The biggest problem with E17 is that it didn’t survived its own promises. Every step forward in development was followed by two steps back. Features added were moved to yet another library and everything had, apparently again, to be refactored again. All libraries were being constantly hacked and never had any releases.

It was almost a year after the E17 announcement when the XFree team announced some new features they were planing, which would allow any window manager to use features like real transparency and real drop shadows and almost everything that the E team promised. At this point, anyone would think “So the E team joined the big dogs and help them to develop something for the community.” Well, wrong. They decided to keep their plans and don’t look around.

Even more time later, XFree announced XDamage and XRender, two features that paved the way to the current compositor-enabled window managers. Even weirder, there were two projects that managed to do that, one lead by RedHat and another by SUSE. Problems? No, they decided to talk and found a way to merge their projects into a single entity, not fragmenting the community and giving a fair change to everyone.

So, since the E17 announcement, we had a major release of X, now forked into X.org, opening it to the great community (XFree was not that open with other people suggestions), several libraries and almost every single desktop is using these new features and getting some very nice eye candy every day.

And what did happen with E17?

Last week I downloaded gOS, a light-weight distribution which uses E17 as window manager and desktop environment, using the sources from the repository, as there are no formal release yet.

What I saw was a alpha release of… something. The keybindings still don’t work properly, the themes are more proof-of-concept than usable (too much animation and very few helpful things), their widget set sometimes decides to ignore the current theme and falls back to the default one and, still, it feels like the window manager is always trying to get in your way and annoy you in the worst possible way and never ever help you.

More than 7 years and still no stable release. In that time, every single desktop environment managed to slim down and get more eye candy and be more user-friendly.

I don’t know about you, but I think E failed.

Written by Julio Biason

April 10th, 2008 at 10:14 pm

Posted in Tech, Thoughts

One Response to 'When open-source fails'

Subscribe to comments with RSS

  1. Elive it’s what you’re looking for.

    plan9ner

    11 Apr 08 at 02:41

Leave a Reply