The Notable Improvements of GNOME 2.22

Michael Larabel writes "Phoronix has up a list compiling eight of the most interesting improvements on track for GNOME 2.22. These improvements include the Epiphany browser switching to the WebKit back-end, transition effects inside the Evince document viewer, a new GNOME application for taking photos and recording videos from web cameras followed by applying special effects, a mouse tweaking module for improved accessibility, and a new GNOME VNC client. On the multimedia end, GNOME 2.22 has a few new features appended to the Totem movie player and the Rhythmbox player. Totem can now search and play YouTube videos and connect to a MythTV server and watch past recordings or view live TV. Rhythmbox now can utilize FM radio tuners, integration with new lyric sites, improved Podcast feed support, and even has support for communicating with newer Sony PSPs. There will also be a standalone Flash player and flash previewing support from the file browser in this release."

Comments

am I missing something here?

[rucs_hack]

The main reason I like gnome is that its a fast window manager with a low cruft index. This looks to me like Gnome trying too hard, and adding too many capabilities to what is, so far as I understand it, just a window manager. Why, for example include vnc? It's not like seperate client/servers for this task aren't available, and most are pretty good.

Is all this new stuff going to slow it down, that's the thing that interests me. If the team have too many things to maintain, just how good a job can they do?

Re:am I missing something here?

[Ckwop]

This looks to me like Gnome trying too hard, and adding too many capabilities to what is, so far as I understand it, just a window manager. Why, for example include vnc? It's not like seperate client/servers for this task aren't available, and most are pretty good.

This is a very good point. Linux is so flexible because each project has a different agenda and a different set of design criteria it is trying to satisfy.

I think that Gnome should not try to be a direct competitor to KDE. KDE is huge, has tonnes of software included with it and tries to be everything to everyone.

We need a desktop environment that does that.

However, this doesn't mean that Gnome should try to be this too. If it tries to, it will lose. KDE's software base is absolutely huge, and it's all controlled from a series of close-nit projects. Gnome would struggle to match that style of development.

Gnome's advantage is that is simpler and less complex. It is my view, Gnome should be a like a good text-book; it is complete not when there is nothing left to add, but nothing left to take away.

Free software is about choice. You don't have a real choice when both options put before you are the same. The differences between open-source projects are not weakness but strengths. Being different allows you to choose your software according to your needs; it allows you to adapt.

Simon.

Re:am I missing something here?

[Excelsior]

First, can you not wait to see the result before jumping to the bloat criticism.

Secondly, the world does want eye candy (see OSX or the IPhone). Gnome is competing with OSX, Vista, KDE 4, and others. In comparison, Gnome is behind in the eye candy department.

I know I've converted more people to Linux by showing them Compiz/Beryl/Compiz Fusion than anything else, by far. When I show them Avant Window Navigator and Compiz in the same desktop, they are snatching the live CD from my fingers. Like it or not shell huggers, eye candy sells.

When you say "do we need more eye candy", I guess you are referring to the "we" that is 0.8% of the browsing public using Linux. In that case, I guess "we" don't need eye candy. But "I" would like to see more people interested in open source and free(dom) software, and eye candy in Linux is one of the best ways to make that happen.

Removed .NET yet?

[metamatic]

Any chance that they've removed the dependency on Microsoft's patented .NET technologies via Mono?

(Yes, I know you can manually remove bits of the Gnome environment to get rid of Mono; but the Gnome environment by default includes Mono.)

Re:Different designs

[Anonymous Coward]

There are, in fact, Qt bindings for C, Objective C, Ruby, Java and many other languages. (QtPython is probably the most widely used.) I'm not sure why you think it's so much more difficult to write bindings for a C++-based API.

Maybe because he's used them. As somebody who was written (and maintained) large programs in both PyGTK and PyQt, I must say that these projects are a great argument in favor of using plain C.

The PyGTK project, for the past several years, has released an update to its bindings within a month of the corresponding GTK+ library. PyQt's last version lagged by about a year. (Not a good sign.) Long after the Qt people had gone to Qt 4.0, we were stuck with PyQt for Qt 3.3, because that's the latest stable release. (It was even hard to get PyQt fixes, because the few PyQt developers spent almost all of their time trying to get PyQt 4 out the door.)

PyGTK has always done just what I expected it to. In PyQt, we always seemed to have trouble with object lifetimes. Qt tried to be just a bit too clever with C++ destructors and object lifetimes and such, and it just didn't translate into HLLs. Between this, and needing to pass C++ method signatures (as a string!) to signal handlers, you really had to write C++-in-Python. It's not really possible to write PyQt code without constantly thinking about the C++ layer. In GTK, I never thought about C: I just wrote Python with a Python GUI library.

I also find the phrase "quirky C required for Gtk+" funny. The C used by GTK+ has always seemed perfectly normal to me. If you want quirky, take a look at the "meta-object compiler" (yet another preprocessor!) that Qt's C++ uses.