Slashdot Mirror
Gnome 2.18 Released
xdancergirlx writes "Gnome 2.18 was released today (on time as usual). Detailed release notes are available. Nothing revolutionary in this release but definitely some nice new features, bug fixes, and improvements."
← Back to Stories (view on slashdot.org)
Personal security is now fully integrated into the desktop, allowing digitally signed communications, encryption of emails and local files, and user-friendly management of personal keys. Internationalization records progress in all directions, with support for vertical text layout and a full Arabic localization matching the quality standards. The official release incorporates essential tools for developers, which hopefully will contribute to get more and better software for the GNOME users.
What's more important, for the first time we ship online games, chess with a 3D look, and endless Sudoku entertainment.
Good thing we've got our priorities straight.
Riiiiight.. cause that summary really screamed hype to me. I see you got modded up too, moderators can't even be bothered reading the summary now?
Fuckin' Slashdot.
How we know is more important than what we know.
Just as some examples:
- As an end-user why can't I extend applications by simply dragging and dropping features from one application to another? i.e. Dragging a search box from one app to another.
- I have 1000s of photographs. How can these images be automatically categorized and displayed most effectively without having to manually add meta-data. It should be sorting images by looking at similarities between pictures, date taken and other automatically generated information
- I have 1000s of mp3s. How can these songs be automatically categorized by mood, tempo, etc without manually entering in meta-data? Think of it as Pandora with your own music collection.
These are some of the type of things that would make using a computer easier to use.Are open source desktop developers so focused on trying to make it "easy" for Windows user to convert they get Microsoft tunnel vision and can't innovate?
It's the year 2007 and we have desktops with the same intelligence as those back in the early 80's.
I think the thing they were going for in gnome is to start integrating, not just password management, but identity management. Thus, Gnome's new feature manages both gpg and ssh keys.
The biggest problem I have with GNOME as a user is Evolution. Simply put, Evolution needs to be scrapped in favor of something else. Its Exchange functionality is non-functional, and its calendar could be easily replaced by something else. Why not just do what they did with the default browser and fork from Mozilla? Surely, it'd suck less.
Nautilus is in dire need of a code audit, just to ensure that everything in there is up to par. Hells, if I were in charge at GNOME, I'd probably stop developing new features in Nautilus and work on the audit for the next cycle.
Honestly, though, the one thing that hurts GNOME the most is the six month release cycle. If they'd even just use a single one-year release cycle, just to clean things up, they'd be in much better shape.
All that said, though, GNOME is my desktop. It's what I learned first, and honestly, KDE's configurability just scares me. Also, I remember too well a time when KDE looked like shit out of the box. Thankfully, that's no longer a problem.
Haec merda tauri est. Ceterum censeo Carthaginem esse delendam.
Not fully automated, but we live in the internet world where an encyclopedia written by Wiki is among the most used references in the world...
Namely, I'm talking about MusicBrainz. Programs will analyze and produce a fingerprint, and MusicBrainz will do a fairly good job of matching that fingerprint to the track. From there, tempo, mood, etc could all be community stored info. More esoteric tracks suffer, but as Wikipedia shows, things that don't work well in theory can sometimes work surprisingly well in practice... Esoteric tracks generally have a more fanatical/enthusiastic fanbase to offset their lack of popularity. Hell, such a system could one up the GP's requested behavior and be able to make recommendations of tracks based on community opinion, both implicit (tracks that tend to be submitted by the same people and rated highly) and explicit (users specifying related tracks).
The photograph conundrum he poses is harder, since generally photographs are personal things. The low-hanging fruit of Date taken and some other things is handled by EXIF data most cameras record, and most photo managers deal with, but looking at similarities in photographs without context is more along the lines of the difficulty you bring up. Some heuristics would probably do interesting things, but a lot of environments will look too similar and sometimes related images couldn't be picked out by a person without any context. For example, a pictures taken of a landscape with some buddies on a road trip would group with some other buddies on the same roadtrip in a bar, no one could ever tell they belonged together without knowing the group and/or the circumstances. Simple fact is, if you have time to take your pictures, you have to be ready to organize them if you care, because no one or nothing could ever do a sufficiently accurate job on such individualized data.
On the drag and drop a widget (in his example 'search'), that seems goofy and impractical. Drag and drop a text-entry widget that happens to be a search into an app with multiple child panes, wtf do you search? What if the child widgets don't have any text to export, or else format it differently? Anyone adding a search widget to most structures knows the complexities and pitfalls, occasionally it is a simple 'add toolkit search and do what makes most sence', but if your program is doing things that people care about, the situation is almost always too complex for that.
However, specifically to his search inquiry, things are being tackled in a more structured way. I.e. beagle is intelligent about the filesystem and a number of popular programs and how they manage data, and how it makes sense to organize it. A popular app emerges and developers who know how to index it right and present it have to manually add the intelligence to do the right thing, and it's effective at keeping up because of a sufficiently healthy development community.
However, in a more general sense of applications sharing features more intelligently, the good old pipes of the command line set the precedent here. NeXT brought that into the GUI world and extended it to know more about the context of the data and whether the operation was applicable before a user selected it. They were/are called services. I.e. you have a text editing application. It had a menu item called 'dictionary'. Well that menu item was actually a third party app that registered itself under the name 'Dictionary'. That same menu item and app would also appear in your Terminal application, letting you spellcheck your *nix commands, since that would be so effective... Probably also in the file management that dictionary item would appear. If you had text in the active context, it would spellcheck that. If it were a file, it would know and spellcheck the file. It's similar on a very basic level to the right-click context menu in windows explorer, but much more flexible and pervasive. Don't know how well it would scale in a highly competitive software market place (many companies wanting a 'Search for related info' menu item would undoubtedly happen and then it gets interesting), but it seems like the best approach to get close to what he describes.
XML is like violence. If it doesn't solve the problem, use more.
Linus' patches don't "fix" anything.
They remove an unnecessary and artificial restriction -- and also apparently simplify the code, which is always a good thing.
they add one feature.. in particular, the ability to configure left, right and middle click to do what you like. Which, ya know, is useful to like 3 people.
It sounds pretty useful to me... Obviously the MS-raised proles will never use it, but many more clueful people use Gnome too ("like, ya know").
We live, as we dream -- alone....
No. But some variant of the patches are in trunk for the next release. It really just adds a config option though. Not as big a deal as the brouhaha would suggest. :)
Also, with the focus on Mono applications, Gnome seems to be getting slower and even more bloated with every release.
Right. Check that to see if you can change your scroll wheel sensitivity. Google it too. Good luck.
Green reminds people of SuSe/Novell, I suppose
You have to love it when things get so slow around here that we post story blurbs that explicitly say they aren't news.
The big thing with changing what the mouse buttons do for many people is simply which buttons your interface thinks are the left, right, and middle, assuming you have more than 3 buttons on your mouse/trackball.
I have an MS 4-button trackball that consistantly comes up with the right button in a position where I would never think to right-click, and the MS (intellipoint) software fixes it with no trouble on Windows, but other interfaces may or may not give me the proper choices to fix it (even games can give some trouble in this respect). It's also nice to have the configuration options for the other buttons, though I generally leave them as default (back and forward in the browser/explorer windows under Windows).
Look, kid, don't put words in my mouth. I didn't say GNOME rules. What I said is that the functionality you want isn't in metacity and you're going to need Beryl to get there. It's all Free and free, so I'm not sure what you're complaining about.
In fact, I don't think you really understand the Unix mindset, which is that everything starts small and simple, and you build more complex tools from the smaller tools. Come with me over to Sun-land for a moment for a prime example that has nothing to do with Linux. A lot of people think that Solaris is an operating system. It is not! It is a distribution of SunOS. It happens to be just about the only distribution thereof, but regardless, I shall continue; Solaris is the combination of SunOS, the operating system, and today GNOME (but formerly CDE, and before that OpenWindows), the windowing system. Both of these are made up of smaller packages. For example, Solaris today includes GNU tools. And the windowing system consists of the X server, the windowing environment, and the assorted X clients.
In other words, you're complaining about a non-issue. You don't run GNOME on its own. Arguably, you don't run GNOME at all - it just runs for you when you log in with a GNOME session. What you're running (when you turn on the computer) is the distribution. If it doesn't include Beryl, you don't get that functionality. Big whoop. Find a distribution that includes the things you want (although I don't know of anyone actually including beryl at this point) and run it. You'll get your Expose-like functionality.
You can complain all you want about how GNOME doesn't do everything you want it to, but what you're missing is that if you use Free Software instead of Apple or Microsoft proprietary stuff, you have the ability to swap in any window manager you like instead of being stuck with what you're provided.
"You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"