“GNOME 2.4 brings to the Linux desktop considerable polish, accessibility and consistency. This release is a culmination of the work done by commercial vendors and the GNOME community, as evidenced by the fact that three vendors–Sun, Red Hat and Ximian–have already shipped desktops focused on the GNOME 2 platform. The end result is a pleasant desktop that is nimble, attractive and unobtrusive. While it’s not perfect, the foundation is now there and the overall product has matured.” Read the in-depth review of GNOME 2.4 at ArsTechnica.
ArsTechnica: Inside the GNOME 2.4 Desktop & Developer Platform
About The Author
Eugenia Loli
Ex-programmer, ex-editor in chief at OSNews.com, now a visual artist/filmmaker.
Follow me on Twitter @EugeniaLoli
It is sorta not out yet, isn’t it?
It is expected Wednesday. Arstechnica probably tested the latest RC, which is pretty much the same as the final, plus a few bugs maybe. But overall, nothing that would keep them from writing a review for 2.4 in general.
I am almost looking forward to trying this one out, either on FreeBSD 4.9 or OpenBSD 3.4 when they come out. The accessabilty features look as though they’ve matured alot over the last little while, and the UI has been passable since 2.2.
I doubt that it’ll make me switch, but I do think that this release will give KDE a run for it’s money…
Have they ever fixed it so that you can search by filename inside the dialog boxes. Sheesh that has bugged me forever.
Also, does this release fix that extremely annoying delay you get whenever you ask a launcher menu to pop up that you haven’t displayed yet? That is a tremendous annoyance on slower machines.
And why does everything have to be so damn gray? Hopefully Red Hat will be smart and not leave it as the default color.
I am almost looking forward to trying this one out, either on FreeBSD 4.9 or OpenBSD 3.4 when they come out.
If you are talking about packages ship with FreeBSD 4.9, then the Gnome 2.4 won’t make it in because the ports freeze starts tomorrow. FreeBSD 4.9 will shipping with Gnome 2.2.
Ummm … type a path/filename into the box, and press tab. it autocompletes like bash.
I seems GNOME is trying to catch up to KDE forever now, and this release show it.
Their new panel, movement of special menu applets, mini calendar for clock, SVG, post it notes desktop applet, have been in KDE for a while now.
Probably the most significant features KDE does not have is HIG apps, snap to grid, text to right of icons, too bad nautilus still sucks.
I understand this is a point release, but seriously very few serious improvements have been made and it all seems to be just polish or copying like snap to grid and text to right of icons from Windows.
“I understand this is a point release, but seriously very few serious improvements have been made and it all seems to be just polish or copying like snap to grid and text to right of icons from Windows.”
Nothing wrong with polish. That’s one of the complaints people have had against Linux. Besides GNOME has been more about building a good infrastructure, than “Look how pretty I am.”
Probably the most significant features KDE does not have is HIG apps, snap to grid, text to right of icons, too bad nautilus still sucks.
Having a HIG is nice. But having a HIG when apps even created by the founders of GNOME don’t follow word by word – I don’t see the point. Besides usability, the main purpose of a HIG is consistency, something not lacking with KDE apps.
And for the rest of the comment, the main features of GNOME 2.4 wouldn’t be seen by users, rather developers. And as for copying, everyone in the industry copies from each other – from Apple to Amiga, from Windows to OS/2. That is nothing bad unless the immitation is done badly.
kindly:
>Their new panel, movement of special menu applets, mini calendar
>for clock, SVG, post it notes desktop applet, have been in KDE for a
>while now.
libRSVG was “first” in gnome, kde recently got the proper capability to render svg images in the DE.
other than that, when you design a desktop enviroment, a “desktop” there are some common denominators you have to use, clock, post-it notes etc.
nautilus otoh, is getting there, konqueror is still not a filemanager it is more a … kparts technology preview.
>I understand this is a point release, but seriously very few serious
>improvements have been made and it all seems to be just polish or
> copying like snap to grid and text to right of icons from Windows.
so i gather you have never actually used gnome, if you had, you wouldnt have written that flame..
reported.
> Probably the most significant features KDE does not have is
> HIG apps, snap to grid, text to right of icons
Actually, this is not true. KDE had a HIG long years before GNOME. They even have snap to grid longer than GNOME.
I would be thankful if you could investigate properly before replying. Such misbeliefs and wrong assumptions often leads to misunderstanding which then results in harrassment and unnecessary flames.
kparts are nice, but they’re like female vampires; they’re nice too.
> nautilus otoh, is getting there, konqueror is still not a
> filemanager it is more a … kparts technology preview.
You think ?
I mean I claim to know both desktops pretty well and I must admit that from personal choice I prefer the way KDE found their solution with Konqueror over the solution adopted by GNOME with Nautilus. I spent some years into GNOME and of course some months into KDE as well (simply by using it and testing it..) and I must admit that Nautilus never felt like a serious filemanager to me. It lacks so many things and has a lot of problems applied to it even now with the release of 2.4.0 it still contains many things. Only to show up a few examples:
– Counting Files in directories is broken and will be shipped broken with 2.4.0. Bug is filed some months ago.
– Emblems. Ever noticed that when you open the Emblems menu entry that you get more emblems shown that e.g. when you open the Emblems sidepane or rightclick on the icons ?
– The grid for snap to grid seems to be not adjustable and thus some icon themes adjust ugly while others adjust nicely.
Well these are only some examples and for sure Konqueror has it’s issues as well but what I want to say is that the overall care of applying patches or hacking on Nautilus is not as good as I would like to see it. And I have to admit that Nautilus is a big project for what it returns to the user. Compare it to DirectoryOpus Magellan II for example. Even the thingy with Nautilus Views has been chewed many times even on GnomeSupport.org’s forums about the pros and cons. I am a bit sceptical how the future of Nautilus goes.
We have the slow development, the problems with proper testing of patches before they get applied (broken Directory count). Then we have a bunch of people who like Nautilus to become a OO based Filemanager and others who disagree, then we have the corner of people who like Nautilus to get rid of the Views while others do like them. Before these questions haven’t been answered what do you think may happen or how do you think we can help to get it on the road ?
Konqueror doesn’t seem to suffer from these questions and problems because of the kparts concept and I personally pretty much like the concept (and only the concept).
nautilus otoh, is getting there, konqueror is still not a filemanager it is more a … kparts technology preview.
Nautilus developers like to call their work as a “shell” as oppose a file manager.
Long time no see, Renling.
counting files in directories, works here, elaborate.
emblems, a non-issue.
the grid, a minor cosmetic issue, the grid could have smaller cubes.
as for the future of nautilus, it goes wherever the maintainer chooses it to go, his guidelines and those guidelines that can come out of discussions between the developers.
what i think may happen is that the development will continue along its evolutionary path and add new features that make sense alongside optimising and refactoring the codebase.
what i think will help to get it on the road is for people to stop talking and start doing, the doing part seems to be something a few people cant get around to, so instead some people choose to criticise the “faults” and leave it at that.
as we wrote in the article, nautilus is not quite there yet, but it will be, eventually thatis.
as for konq, konqueror is as discussed vividly on the konq list.
the concepts are interesting, on one hand you have konqueror that uses kparts to make the functionality, on the other hand we have nautilus which uses bonobo to some extent and then goes a step further with a script facility. i’m currently wading through the source of nautilus so i’ll get back to that later.
yes, we got bought up along the project by a minor enterprise so that is taking up most of my time.
its good to be back rajan ;D
“We have the slow development, the problems with proper testing of patches before they get applied (broken Directory count). Then we have a bunch of people who like Nautilus to become a OO based Filemanager and others who disagree, then we have the corner of people who like Nautilus to get rid of the Views while others do like them. Before these questions haven’t been answered what do you think may happen or how do you think we can help to get it on the road ? ”
Sounds more like the Open Source process in action, than anything specific to GNOME. I’m certain something similiar happens on the KDE side.
> counting files in directories, works here, elaborate.
This is pretty much strange because it has been a confirmed issue. I even went into the #nautilus channel and got supporting feedback for this issue. By the way I’m refering to a more recent 2.4.x issue here.
Well you are right that usually the maintainers decide which way they go since they provide the code and somehow target it into a direction. But I also think that they are not sure either which way they go and how they should satisfy all types of people.
The usage of the scripting facitly and bonobo are for sure the big benefits of Nautilus. I think we’ll see how 2.6 will be and also watch out what users have to say once 2.4.x is officially out and announced everywhere. At the final end they are the audience who need to use it
im rather interested in the process towards 2.6 since the focus is going to go into the N part of “GNOME” .
the fruits of that has already made it into gnome-network, ergo the mdns fileserver/personal webserver/ very interesting.
Nuff said.
Oh .. and gnome 2.4 rocks .. been using gnome-2-3 from CVS
Well, Nautilus was the first Bonobo app to reach 1.0, or so boasts Eazel before they went under.
How long to you think it will be before it is ready on freshrpms.net???
…and then we will yet again see that gnome is still way behind KDE in usability and configuration options.
…and then we will yet again see that gnome is still way behind KDE in usability and configuration options.
This is typical comment from Troll, no matter if GNOME, KDE, Windomaker,… user.
So once again, if you arn’t able to comment in a constructive way, close your mouth and keep using what you decided to use and be happy with it!
just my opinion
H.a.n.d.
Golum at work dot ork
No.. this is a typical response to a comment absolutely devoid of *constructive* criticism.
Form proof and solutions for your criticisms. Then you will be taken seriously.
Can we get to a normal level of conversation again please ? This Thread has been started so nicely.
> This is typical comment from Troll…
I wouldn’t call this a Troll because he has a lot of valid points and I have the tendency to agree with him in several ways even while supporting GNOME. I was seeing GNOME as the ultimate Desktop myself for quite a long time and thought of it to be the most enchanced open source Desktop there is until I tried KDE myself and I must admit that I was happily impressed by the progress it made over the years. Even going from KDE 3.1 to 3.2 (CVS) it made a huge noticable jump. This is of technical and usability point of view and not simply 3 changed icons and ‘wow Nautilus is faster’. Due to their choosen framework, development language and strict development rules they made a lot of progress. Not to mention their clear roadmap where they decide before what should go in and what goals should be reached. Everyone underestimating it is an ignorant.
I sometimes wish that GNOME is as strict and disciplined in following similar goals but their road is ‘develop as far we get and release it’. A lot of people replying here should seriously spent the one or other day to test both Desktop throughly but I doubt to expect this because of bias.
“All over, GNOME hypes for having this and that feature, but always when it comes out, Mr. Pennington have decided remove even bit more configuration options… and people who want to configure their desktop and know what they want and dont want anyone else to decide for them how their desktop should work or look, get disappointed. Like me. ”
If they get dissappointed with the direction GNOME follows they should switch to the other desktop system.
KDE and GNOME follow different directions. What GNOME does in regard to configuration options makes perfect sense if they aim for the corporate desktop market and not the typical geek’s desktop. IMO it’s good that KDE and GNOME differ so much, it would be disappointing to have two fine desktop environments which are the same. So we have two desktops for different people, different needs. Actually I think there is demand for more (specialized) desktops, because the flexibility of both KDE and GNOME is of course limited, whereas needs are not.
I do and did test both of them and will continu to do so, but then again I come to different conclusions then you do so I’m probably biased as you call it.
What I see in KDE is the sheer hype of it all, Incompleteness, bugs and crashes may only be mentioned and discussed if it is about a version before the current version, the current version number is always rock solid and the future is always heaven.
Gnome on the other hand is less about hacks that mature and grow without guidance and goal, but more about doing it good and make it solid, like gstreamer and its applications (totem ed). Sure Totem works great with the xine library, but gstreamer is the chosen (with reason). So totem isn’t in the current release (2.4), KDE would go for xine and slap things on it when needed, gnome has a goal with its media api and implements it and calls it stable when ready.
Quite different then the view you hav he oGalaxyo?
I didn’t want to call him a troll, it was just the way he was commenting! Of course he has valid points… but It is always the same thing. When there is a new GNOME version, or only a new handy feature, there is always someone, who is bashing GNOME by telling KDE/FLUXBOX/Wathever has that feature for a long time, and GNOME is still way behind… in trolling manner! I’m just a little tiered of that. Isn’t it ok, that people are looking forward to a new version of software? constructive critisme is ok, but not in a trolling manner!
I don’t like flamewars as little as you do. And I also think that we should keep the quality as the thread started. For my part, sorry for the harsh comment!
H.a.n.d.
Golum at work dot ork
> If they get dissappointed with the direction GNOME follows
> they should switch to the other desktop system.
This would also include a bunch of people developing and contributing for GNOME then. What do you think would be better arranging with these people or tell them to go away knowing that you loose a bunch of developers and contributors because someone decided so and give people from outside the feeling that it’s a closed community ?
GNOME is communitywork and people contributing for it should be respected. When someone show up on a mailinglist telling that they do not like the direction GNOME leads then they should be taken seriously. If two GNOME heads decide that GNOME must go a corporate way and flout the opinion of other developers then this could be really dangerous.
I would like to quote an interesting reading which I found on SlashDot it’s written in good english and sounds mature in the way it was replied so it’s up to you wether you read it or not but I think it clearly has some points. One reply is from Havoc Pennington and the other from someone else.
http://slashdot.org/comments.pl?sid=52921&threshold=-1&commentsort=…
And finally I would welcome if we could get back to a normal conversation. A bit more respect so we can all get along.
Has anyone here read teh Arstechnica review at all. Please, read the whole damn thing and then post. And stop calling each other trolls here. All you have to do is report them. And no one sees their posts.
Freedesktop.org is the reason a GNOME notification icon knows how to run in a KDE system tray and why rich object copy and paste works consistently across desktops.
Rich copy and paste works across desktops? This is news to me. Unless you count “text only” as rich.
There was a discussion on d-d-l about this a few days ago. Eventually we’ll have to catch up with Windows in the rich object transfer field…..
is only done by ppl that still havent found the menu-entry called “GConf”. I am not calling you stupid for this, but don’t call the gnome-developers names for moving a lot of congiruation away from the preferences panels into GConf.
I mean, it’s known for a fact that the gnome filosophy will encourage this, because it doesn’t distract less knowledgable users and knowledgable users will still know how to find the further configuration (through gConf)
It’s not because some ppl can find them (they are hidden on purpose btw) that they are not there
Thank you for saying that. There is a policy and if you have anything against the policy then say so.
It surprises me that some people do not seem to want midde ground. On one side there are those who complain that Linux makes you have to know the famous/infamous dotfiles (.gtkrc and so on), and on the other a group of ‘power users’ who want to configure everything from GUI. When GNOME provides a GUI to aggregate these options in one place, whilst leaving the text editable files, they find favour with no one. Strange.
Why is that KDE and GNOME users have to jump down each others throats? They’re different desktop environements. Deal.
GNOME has less options than KDE. This is a design decision, clearly targetting different end-users.
It’s not like you can’t install KDE, and GNOME on the same machine. Heck, on one of the X11 servers at work, I’ve spotted simultanious use of KDE, GNOME, E, fluxbox, and XFCE4, and often all using both GTK and QT apps.
If you don’t like it, don’t run it – some one else deciding to run something else doesn’t limit your options in any way.
I was already writing down a detailed and calm response to oGALAXYo, when I realized that Havoc wrote exactly the same in his response to the posting Galaxy linked to, but people weren’t getting it either, so there is no point for me to try it again.
However, what I’m really hoping for is that all those who bitch about the direction GNOME has taken after 2.x, will go on and keep working on 1.x under a different name or fork 2.x to be more like 1.x (under a different name). This could be done without breaking compatibility with anything and you wouldn’t even have to duplicate work if you do it smartly.
Then every user could use what he wants to and every contributor could choose which project he wants to contribute to. Survival of the fittest. Then those of us who like the new way GNOME is going (including most if not all of the main contributors and many geeks, mind you) could continue enjoying it without this kind of tiresome discussions full of negativity and down-talking of great efforts (which made me really happy with my Linux system again, after a long time).
Just ranting will not change _anything_, because in the end, it’s those who do the work who get to decide and if those disagree with you, you can rant on as much as you want.
I beg you, please fork.
I find it odd that people come to the conclusion that GNOME is geared towards the corporate, or normal user, and not the geek, power user types. In fact, I find that GNOME is good for both sets of users, myself being in the second category.
GNOME isn’t designed for the first set; rather, it’s designed to be usable. The focus on HIG compliance, and polish, rather than simply adding features, and the focus on developing a proper library to developer from (GStreamer over Xine, for example) all means that you have a desktop that is intelligent.
Let me explain in more detail my thoughts. I appreciate the fact that when the GNOME developers make a release, they promote the fact that they have put a lot of effort into polish it. Not everything is a new *feature*. I look at the polish, and the enhanced usability has a new feature. For example, some of the new Epiphany things they have planned aren’t adding new things, but making what’s already there smarter.
As a programmer, I like this. My desktop suddenly becomes more intuitive. I used KDE for a long time, and I did appreciate the level of options they offered. However, there is something to be said about spending time adjusting options, and simply having the options work by default.
There is an argument floating around here implying that the greater the number of configuration options, the greater the usability of the resulting software is. Nothing could be further from the truth. Poorly designed software from a UI standpoint is poorly design, no matter how many options you throw at it.
In fact, an UI that promotes the fact that you can change it around is usually a sign that less thought was actually put into the design of the UI; the sentiment “if the user doesn’t like it, they can change it” prevails.
The other usability aspect is the underlying development library. GNOME has a habit of developing exactly what it needs to get the job done, while I find KDE doesn’t want to reinvent the wheel. Both ways have benefits. KDE is able to leverage an existing code base while GNOME has to develop a new one, rather then spend time developing applications for it. GStreamer is a prime example of this. GNOME could have gone with Xine, or Mplayer, or something else entirely.
However, they wanted something a certain way. And rather than settle on a existing solution, they developed their own. This may not seem like a benefit to the user, it may not feel like a feature to be proud of (“Great, but such and such was able to do that for 5 years now.”), but in fact, it is.
The old mantra about reinventing the wheel is flawed. You see, the idea behind the saying is that if it’s already been written, don’t write it again. Use what is already written. This is something that really grew out of early programming practice with regards to writing functions, and creating modules.
However, people try to apply this mantra everywhere. For example, people scream why we need another browser, everyone should just develop and improve Mozilla. Or why another desktop, just focus on making one super one.
And in each case, these people are running with the idealism that we should reinvent the wheel.
But people really do “reinvent” the wheel. Goodyear isn’t selling the same wheel every year. They have new models. Pirelli isn’t the same as FireStone. Winter tires, summber tires, all wheels. Big wheels! Monster trucks, horse drawn carriages, trains, planes, and automobiles.
All have different tires, all have different wheels. The concept of a wheel, it’s round and it is used to transport things is the same. The same with web browsers: It’s an application that lets you view web pages. The same with desktop environments: it allows you to run applications in a uniform environment.
Anyways, I digress. The point of creating these carefully crafted (for their specific needs) is that when all is said and done, developers will be able to utilize this code more easily. Their applications can all speak a language that allows better communcation. And this has a direct benefit for the user. Rather than have to worry about how to do something for an application, it’s already been done for the application developers. So suddenly, that new Gnome program is suddenly “media” aware with little work from the developer.
A powerful desktop isn’t the one with the most options (even still, with GNOME, most of the configurations have been moved to GConf), but that is the most usable. And more options doesn’t mean more usable.
Gnome is way ahead of KDE in terms of accessibility, if you’re talking about accessibility for the disabled. The ATK toolkit which works extremely well with GTK in Gnome 2.4 allows a level of accessbility to those that are disabled that isn’t found in KDE currently.
Gnome 2.4 I believe allows US Section 508 compliance.
I’ve used both KDE and Gnome for periods of time, and ended up using Gnome, simply because all of my applications are either Motif or GTK based anyway and the some of the window manager hints they use KDE doesn’t honor. I think the other reason is that KDE hasn’t been very stable on my systems for whatver reason. I’m used to constantly get the little Dragon error box complete with backtrace 😐 Admittedly, the latest version of KDE barely crashes at all so that’s better. But, I still don’t have a compelling reason to switch Application wise…
Thanks for everything. GNOME is a wonderful environment to use a computer with. I appreciate all the hard work and dedication of all people involved. Thanks again.
I like gnome. So what? Somebody else likes KDE. So what?
It is just a frickin’ desktop. Choose the one you like the most. It is called choice. Most distros besides Redhat have already chosen KDE for you. If you don’t like it or Gnome try Xfce or even non-linux, bsd and use Windows or even Mac OS X.
It is time to focus on the real shortcomings of the entire Graphical Interface for the free *Nixes from X on up. Everyone wants to blame X but there is enough blame from the widget folks working XUL, and QT and GTK on to the Gnome and KDE people right down to the folks that code OpenOffice. I am not interesting in another KDE vs. Gnome flamewar. Personally, it would be nice to lock all the above mentioned folks in a room and have a real discussion on making the interface nicer but more importantly faster for the end user.
Gnome 2.4 is a nice progression. Is it as mature as KDE? No. Why? KDE is an older project. Gnome 2.0 was a huge change from the backend to front and back again. They had a lot of ground to regain.
I’ve really grown to like ArsTechnica. Their reviews are more technology reviews than “good or bad” reviews. They have good record for avoiding partisanship with their reviews. Look at their architectural comparisons of the P4, Athlon, Opteron, G4, and G5 as good examples. As a rule they tend to look at the positive and negative aspects of each design, often leaving the question of which is “better” up to the user.
I actually like that they didn’t mention KDE in the article (Qt only mentioned as background info). Some will complain that they are avoiding an important issue, which is certainly true. However, by avoiding the KDE vs. Gnome discussion, they are able to focus on clearly describing the technologies of Gnome 2.4. I found the article technically informative, but useless for choosing between KDE and Gnome. Since I’m a geek, this is fine for me. I can make that decision for myself.
I also agree with the ending point of the article that many of the Gnome technologies are still immature, but that they are progressing nicely and the future roadmap looks promising. Obviously, it would be nice if things happened faster, but sometimes doing things right takes time.
Ok this may sound inflamatory but I have to ask:
> clearly describing the technologies of Gnome 2.4
What benefits will new technologies give if they are not throughly used ?
> the future roadmap looks promising.
Show me the roadmap you are refering to.
> but sometimes doing things right takes time.
Define ‘doing things right’ in your context. By the way this is a question only directed to Roy.
What benefits will new technologies give if they are not throughly used ?
I don’t see Roy discussing the benefits of gnome technologies, just that Arstechnica reviewed them well enough.
Show me the roadmap you are refering to.
In the article they refer to things like more multimedia apps integrated (rhythmbox, totem), a better file selection dialog and the like. It’s in the article, and guess that’s what Roy talks about.
Define ‘doing things right’ in your context.
Why in his context? In the article they talk about how only technologies and apps that work will be included in future gnome releases. If you read the article, then Roy’s post, it makes sense.
By the way this is a question only directed to Roy.
He left an email for private contact, and this is a public forum, so don’t try to discard anyone’s else opinions just because you want to bash Roy for thinking gnome is good.
Geez, galaxy. Will you ever stop? I agree with Spark that you should stop bitching whenever a gnome news appear, and just do a fork if you don’t agree with gnome’s direction and its main developers.
As you see in Ars Technica, there are people that like gnome’s progress. But you still go on those rants like if the only valid gnome is the one you dream of.
I wasn’t talking with you, the questions were targeted to Roy. The Ars Technica ‘Roadmap’ is nothing NEW. I recall where Totem was already part of GNOME then taken out again, same for Galeon which was part of GNOME and then taken out and even Rhythmbox was part of it and then taken out. What clear Roadmap is this when it’s changed over and over again ?
A Roadmap is something you decide once and trying to achieve during the development process. Once the process has been achieved you can go in an itterative process and set new goals for a new Roadmap. Same Roadmap for the nice Fileselector. Yes it shows up on 2.0 then sorry we can’t do that we shift it to 2.2 then to 2.4 and now it will be shifted to 2.6 and I bet my pants that it will be shifted to 3.0 because they need to integrate GNOME-VFS and GConf into GTK+ and I wonder how and when this will become possible. Even GStreamer we hit 2.4 now and it’s still ‘strangely’ integrated into GNOME. There is nothing in GNOME that uses GStreamer besides the GNOME-Media module and it’s not sure which of the apps supplied with GNOME-Media are GStreamer’ed.
Don’t be naive. The GNOME Roadmaps are Roadmaps which (known from the past) are being changed more often than people change pants in a year.
Doing things the right way ? Which way is the right one ? Yours ? Maybe mine is right who knows, Maybe the way of Eugenia is right or maybe the one from /. It’s not like 5 people work on GNOME it’s a lot of people and you already know how often I repeated that not every developer is excited about it but that are facts you usually ignore in the favour for a quick flame.
Same Roadmap for the nice Fileselector. Yes it shows up on 2.0 then sorry we can’t do that we shift it to 2.2 then to 2.4 and now it will be shifted to 2.6 and I bet my pants that it will be shifted to 3.0
The fileselector is not part of GNOME, it’s part of GTK. And now your pants are in trouble. Man, i’m gonna laugh so hard at you when they release the new GTK, you bet i won’t forget about your pants.
It’s so funny how these KDE boys get so jealous everytime there is some kind of Gnome/Gnome-related release
They just can’t help trying to find problems in the Gnome project… you know, jealous.
Victor.
> The fileselector is not part of GNOME, it’s part of GTK. And
> now your pants are in trouble. Man, i’m gonna laugh so hard
> at you when they release the new GTK, you bet i won’t forget
> about your pants.
And that’s how misunderstandings often start which often end in un-needed flamages.
The Fileselector is indeed part of GTK+. But the Fileselector which should be included into GTK+ is a GNOME one at least it depends on GNOME-VFS and GConf refering here to the LibEGG Fileselector. To integrate exactly this Fileselector you need to get GConf AND GNOME-VFS to the GTK+ layer.
By the way I’m developing my own GNOME applications like Atlantis and CVSGnome and you can simply belive me that I spend a lot of time into these things or do you think I show up here unprepared and talk without proper background knowledge ?
Jason – your post was excellent. You pretty much summed up the reasons why I have chosen GNOME over KDE.
-G
You obviously disagree with the direction of Gnome2. But disagreement is going to happen whenever ANY decision is made. Indeed, it has alienated some Gnome developers, but hard decisions are necessary in any large project. The people currently in charge of Gnome aren’t trying to piss off Gnome developers. They think that these decisions will produce a better result in the end. Maybe it will. Maybe it won’t. That isn’t my point.
Anyway, if enough community developers feel they are being ignored, perhaps they should fork Gnome development. Also, they could try to change Gnome leadership. I’m not sure how the Gnome leadership is chosen, but there is some sort of voting process, right?
You seem to be the one not understanding. The GTK+ fileselector will not depend on gnome-vfs. It will be pluggable which means that _IF_ gnome-vfs is available, it can be used. If not, the fileselector will be an interface to the regular filesystem. It will work either way; using it in an environment which can extend it gives more advanced possibilities like SMB and FTP, using it in an environment which doesn’t add anything leaves you with a fileselector that only is able to access the regular filesystem.
You wrote in a previous posting that the developers should listen to anyone who shows up on a mailing list. I strongly disagree with that. What if I show up on a KDE mailing list and demand they remove Konqueror? Should they listen to that? I think they would be rather stupid to do that. I respect very much that the developers stick to doing what they believe is right instead of succumbing to pressure from what I believe is mostly uninformed people. I generally get the impression that they know more about the things they develop than I do. But hey, maybe I am just superdumb?
If you don’t like Gnome, don’t use it. I am not sure why you have such a beef with the Gnome developers, but it is rather tiresome to read. You keep repeating the same rants over and over without adding anything new. If you want Gnome to improve, you might want to contribute without that know-it-all attitude you keep displaying.
It is my impression that Gnome is heading in the right direction. I waste a lot less time fiddling with settings nowadays than I did when I used fvwm or Gnome1. And I can’t think of a single configuration item that has been removed which I miss. None. It works beautifully for me, so at least they haven’t failed to cater for every user out there.
> What benefits will new technologies give if they are not throughly used?
I’m not sure I understand your statement here, but I’ll give it a shot. If I’m missing the point, let me know. Are you referring to the fact that a lot of Gnome applications don’t fully utilize technologies like GStreamer (though GStreamer is not technically a Gnome project)? Some technologies like Pango are being used. I already use Sound Juicer and Rhythmbox, so I am using GStreamer, though I’m obviously not using it to its full potential.
> Show me the roadmap you are refering to.
Maybe the term “roadmap” was somewhat misleading (though 2.6 is set to contain some nice stuff). Maybe just “future” would have been better. I was just referring to what I’ve read on many of the individual project pages where they discuss features they are planning to implement in future releases.
> Define ‘doing things right’ in your context.
Mainly, I’m referring to a well thought out design. This wasn’t a grand statement that Gnome was doing things the one right way, just that I’d prefer a well thought out design that takes a little longer to a hack that can be done quickly. Specifically, I was referring to the file selector. It is taking a long time because they are reworking the programming interface. I’d prefer that they take the necessary time to come up with a good solution rather than rush out a hacked up solution.
Ok from the Fosdem 2003 writing they plan to achieve the goal of the plugable Fileselector but from further conversation February this year on the GTK+ Mailinglist (reference):
http://mail.gnome.org/archives/gtk-devel-list/2003-February/thread….
You see by follwing the thread that getting the mimesupport of gnome-vfs may be required and thus needs to be taken to the GTK+ level. Say it differently. To get gnome-vfs support you need to make gnome-vfs available to GTK+ before GTK+ get’s configured. GTK+ will see ‘hey gnome-vfs is installed I like to suppor the mime type’. This was my point from what I know the libegg Filemanager is chosen for this thing otherwise the people wouldn’t spent hours into committing stuff to CVS and when going through the code you also see that it requires GConf for storing whatever preferences. This means when they (the GTK+ and GNOME devs) are going to implement this (the plugin system for either normal Fileselection or the enchanced one with gnome-vfs mime support then you can expect that this implementation is done in one go. My point.
No I have nothing against GNOME but I am realist and see things as they are offered today and not how people try to make it belive in the public by writing baggy articles that show GNOME in ways which do not reflect what we have today.
I don’t want to make an harrassment from my writing here and appologize if this is what you people understand from my writing but I’m pretty sure that not everybody understands what I like to point out.
Thanks for explaining the questions. Yes now I understand you a lot better and I must admit that I fully agree with you. I was a bit curious when I first read your comment because many people simply refer and baggy things which do not reflect what is now. But yes the way you explained it here is wise and make a lot of sense. Let’s see what will happen. People are already developing on HEAD for 2.6
Ok, the gtk+ fileselector is going to suck right up to the gtk+ 2.6 release.
What to do? Use Ximian. It has some decent fileselector hacks for shortcut buttons at least.
For everyone else use the fileselector here:
http://members1.chello.nl/~h.lai/gtkenhancements/
Download script and run script follow the prompts and you are done.
Its in Debian unstable too. It isn’t great, but its good enough that I don’t think about how much it sucks when I use it.
I think the fileselector has just become an issue to beat on Gnome with. People, it’s getting taken care of. Eugenia and other so-called ui experts can wail on it all they want (it’s their function in our community in a way) but a few folks have turned it into a buzzword for the “Gnome lags behind other desktops” argument.
KDE has a million issues. XFCE has a million issues. Gnome has a million issues. Windows has a million issues.
I prefer Gnome over KDE. I’ve tried a bunch of desktops, and from some feel polished and solid, and others do not. In my opinion, Gnome has this feel, and KDE does not. It’s not because of issues like file selectors and snapping to grid. When a Gnome developer implements something, more effort seems to be placed in making it feel smooth and done properly. When you open up nautilus, and click and highlight and move files and directories around, you get smooth colored icons, that colored box, and a general feeling of “solidness.” With KDE (last time I checked was 3.1; if it’s changed then I apologize in advance), you get the Windows 95-ish dotted line for the selection box, poorly colored icons (especially in the text area), and flickering.
I am not a novice, but I take less features and better execution over a bunch of features and poorer execution. Gnome seems to pay more attention to quality, and kde seems to prefer adding functionality.
I notice you didn’t mention OS X, nice.
I think everyone is making a mountain out of a mole hill over this damned file selector. Granted, I don’t use GNOME/Linux that much these days, but when I do it doesn’t irritate me at all, I don’t even notice it. Would half you people be whining if you hadn’t heard someone else whine first? No. Hell, Finder in OS X is where the grief is at. It’s not a big deal, get over it.
Jason – your post was excellent. You pretty much summed up the reasons why I have chosen KDE over GNOME.
What is wrong with FooBarWidget’s file selector?
What is it missing?
There is a thread on gnomesupport.org for it where you could post. He has already included a number of ideas from other users into the interface.
Put together a response and let your voice be heard sir!
oGALAXYo wrote:
> [snip] same for Galeon which was part of GNOME and then
> taken out and even Rhythmbox was part of it and then taken
> out.
This is not true. These modules were never part of any official GNOME Desktop modules list. Not. Never. Not true.