4 ms·

It may be a cause or effect of this, but the open source culture has fractured in the 2000s to many smaller projects or libraries instead of these huge mega-pro

by bergie 14y ago

It may be a cause or effect of this, but the open source culture has fractured in the 2000s to many smaller projects or libraries instead of these huge mega-projects (say Linux, GNOME, KDE, Mozilla, OpenOffice) that mostly come from the 90s.

coldtea 14y ago
I'd say it's an effect. Without corporate funding large and co-ordinated projects (Gnome, KDE, OpenOffice et al) have fallen by the wayside. That's why progress has got so slow the last 6-8 years in so many parts of the OSS world (and on lots of proprietary software, that lazily relies on OSS). Things like Gnome or OpenOffice are essentially the same as they were 10 years ago, whereas the proprietary alternatives have gotten tons of changes. GTK+, the main toolkit behind Gnome has one (1!) major contributor (which complained about the issue). In a proper world, something like this, used by so many millions, should have at least a full time team of 10 persons. Contrast the additions to GTK+ in the last 10 years with the development of the corresponding Cocoa libs, and it's like night and day. A tree-view widget here, a spinner there, ... Small libraries and stuff can still flourish under such an environment, but not big, co-ordinated projects.
- bergie 14y ago
  I'd agree otherwise, but GNOME is a bad example. In the last couple of years they've rewritten a lot of their stack. All APIs are now on GObject Introspection, the GNOME Shell is a completely new thing, written in JavaScript, new set of default applications, etc. But then again, they seem to still be receiving some amount of corporate sponsorship from Red Hat. A few years back (when Nokia stopped supporting the project) things looked a lot more gloomy: http://wingolog.org/archives/2008/06/07/gnome-in-the-age-of-decadence http://wingolog.org/archives/2008/06/07/gnome-in-the-age-of-... Then again, any desktop environment or toolkit is bound to be somewhat unsexy these days when the focus is on web and mobile. I also wrote about this back in 2008: http://bergie.iki.fi/blog/gnome_in_decay/ http://bergie.iki.fi/blog/gnome_in_decay/

by bergie 14y ago

coldtea 14y ago

I'd say it's an effect. Without corporate funding large and co-ordinated projects (Gnome, KDE, OpenOffice et al) have fallen by the wayside. That's why progress has got so slow the last 6-8 years in so many parts of the OSS world (and on lots of proprietary software, that lazily relies on OSS). Things like Gnome or OpenOffice are essentially the same as they were 10 years ago, whereas the proprietary alternatives have gotten tons of changes. GTK+, the main toolkit behind Gnome has one (1!) major contributor (which complained about the issue). In a proper world, something like this, used by so many millions, should have at least a full time team of 10 persons. Contrast the additions to GTK+ in the last 10 years with the development of the corresponding Cocoa libs, and it's like night and day. A tree-view widget here, a spinner there, ... Small libraries and stuff can still flourish under such an environment, but not big, co-ordinated projects.

bergie 14y ago
I'd agree otherwise, but GNOME is a bad example. In the last couple of years they've rewritten a lot of their stack. All APIs are now on GObject Introspection, the GNOME Shell is a completely new thing, written in JavaScript, new set of default applications, etc. But then again, they seem to still be receiving some amount of corporate sponsorship from Red Hat. A few years back (when Nokia stopped supporting the project) things looked a lot more gloomy: http://wingolog.org/archives/2008/06/07/gnome-in-the-age-of-decadence http://wingolog.org/archives/2008/06/07/gnome-in-the-age-of-... Then again, any desktop environment or toolkit is bound to be somewhat unsexy these days when the focus is on web and mobile. I also wrote about this back in 2008: http://bergie.iki.fi/blog/gnome_in_decay/ http://bergie.iki.fi/blog/gnome_in_decay/