GNOME 2.14.0 Release Candidate

It's very nice, but I'd prefer a different hover effect on the top links. Something more subtile like just a brighter background. Also the hover effect on the gray link-list feels a little counter-intuitive since it darkens instead of lightens.

And maybe a small amount of padding inside the text boxes would be better.

Implemented! ;) What do you think?

Nice. :) I still don't really like the prelight on the main linkbar though (the one with About GNOME), mainly because of the flipped gradient. The border is also shifted one pixel up on prelight, that's probably not intentional. That might be fontsize related, because the shifting varies when I zoom in or out.

Yeah I am having fun with that. You get different padding to the border between different web browsers. Some pad to the inside of the border, others pad to the outside. What is even weirder is that it my broswer it pads to the outside of the top border and the inside of the bottom border. I may just need to go back to a solid color to avoid these issues. Have any suggestions?

I actually think it's a bit to shiny for my taste. Reminds me a bit of KDE :-(
But to be honest, I don't really care that much

agree... it's really nifty. feels refreshing :D

Gnome 2.14 is very cool in most ways. The performance improvements in particular are amazing.

But gnome-screensaver in its present form is a mistake. It will, sadly, make most people see 2.14 as the "now-they-even-want-to-take-away-my-screensaver-release".

gnome-screensaver offers almost no advantage to the user (compared with xscreensaver), but introduces new bugs and incompatibilities with existing software.

From reading the g-s faq, at

http://live.gnome.org/GnomeScreensaver/FrequentlyAskedQuestions

it seems that the main feature of g-s is to prevent people from having a screensaver that says "The CEO is a bastard".

Is this the new spirit of Gnome?

They did leave it alone. gnome-screensaver is for those of us who couldn't care less how fast their Matrix simulator is scrolling. :P

I guess people need something to bitch about with every release, at least nobody can argue that this is hurting their productivity this time. :)

gnome-screensaver is actually helping productivity. Since it is not backwards compatible with xscreensaver at all (by design, it seems), it happily activates in the middle of game-playing and videowatching (with every mediaplayer I've tried, except totem).

And even better: it locks up X in the process, thus increasing productivity even further.

A stellar piece of software.

Hope you filed bug reports. That is pretty annoying behavior.

Right, if they don't fix one bug, they will obviously fix no bug. Sometimes...

If the screensaver kicks in during a game while the user isn't idle (didn't happen to me yet), that would be a serious flaw and much different to an application not properly disabling the screensaver timeout. Besides, the fadeout effect is very slow and interruptable, so in that sense it is much less annoying than when the same happened with xscreensaver while watching a movie.

you mean "he gave reasons in 3 of those bug reports and even created a FAQ", then yes, you're correct. Just because you remain unpersuaded does not mean that he didn't offer a defense of his actions (or lack thereof). And please remember that just because some people are screaming against gnome-screensaver, does not mean that there isn't a quiet majority in favor of it. You can't please everyone...

You're demanding that developers to do work for you, gratis. You need to convince the developers that you're correct, not the other way around. The presumption of correctness lies with he who does the work. If you don't like a product and can't convince its developers to change it in your favor, simply don't use it.

This word "reasons", I don't think it means what you think it means. The stated reasons seem quite clear and rational to me. For instance, if something has no obvious advantage and doesn't help any user achieve any of their goals, then that should be reason enough to reject something.

If your goal is to fiddle with stuff aimlessly, or if your goal is just to always have everything (useful or not) exactly like it always was, then you are probably capable of installing the relevant old stuff.

Screensavers just aren't very useful.

No one said that there was a quiet majority in favor of it. Just that because some people were vocally against it *does not necessarily preclude* that many, non-vocal others are in favor of it.

It is of course logically possible that everyone, except for the ones that speak up, like the new screensaver better than the old one. However, it is also logically possible that dogs fly, except for the ones that you and I have seen.

The requested functionality exists, and the developer was kind enough to point you to the relevant docs. All that remains is for you to tell the mplayer guys, which shouldn't be too hard since you took the time to tell the gnome-screensaver guys. It's not wholly unreasonable to ask mplayer to change their upstream app. The following dbus command line should work, for instance:

dbus-send --session --dest=org.gnome.ScreenSaver --type=method_call /org/gnome/ScreenSaver org.gnome.ScreenSaver.InhibitActivation

According to the gnome-screensaver FAQ using dbus is not a viable solution: "Please be warned that the gnome-screensaver DBus API is currently unstable and may change in the future."

It doesn't give any hints about what to use instead, though.

gnome-screensaver breaks pretty much every app that needs to prevent the screensaver from activating (video players, openoffice.org presentations, full screen games, etc.). And I consider that a bug in gnome-screensaver, not in mplayer, vlc, xine, openoffice.org, etc.

For example: the code that disables the screensaver in xine is 4 years old, and has worked perfectly ever since. And now it doesn't. Why is this a good thing?

Talk about constructive! A pity nautilus-actions doesn't seem to be available for Ubuntu, perhaps there's a deb package somewhere.

Thanks for the tip, anyway. Since nautilus-actions is going to be part GNOME (IIRC) how about including this action as a default?

rehdon

perhaps there's a deb package somewhere

... like on the project's web site :)

rehdon

> The famous nautilus's recursive permission bug is still here. And yes, it's been there in every release of gnome for six years!

I was one of the people who tried to do something instead of just talking. Instead of bitching around here, I ask you to come up with use cases on usability list, so that we can decide how much of chmod's power we should expose to the user.

My perception was that the main use case is a cdrom->hd copy, which could/should be resolved as described in a bug report. The kernel should expose the "correct" permissions and on the read-only media.

Can you come up with other use-cases? :)

Yes, you do ask use cases for something that's been around for six years. You always always ALWAYS elicit requirements and express them as use cases from your stakeholders, ie, the public, developers, and community.

If you just go ahead and implement, what you perceive as fixing the problem might not be what others perceive. Ie, you might expose too much/too little of chmod's functionality, depending on the application it's being used for (perhaps a kiosk and a user's desktop would have very different ideas for requirements).

Also, many nuances have to be considered. Can your design be disabled if the administrators don't want to give this ability to users? From what contexts will you be able to use this feature? What indicates success/failure of an operation? There are many many MANY details.

This is how software engineering is done. Any school worth its salt will teach you that. The grandparent poster was starting the first step in the process - elicit requirements so that a design can be conceived to satisfy those requirements.

It would seem that Gnome devs do get it.

I'm the author of the "bla bla" posting.

My use-case:
An admin uses a GUI where he shouldn't, i.e. he just wants to chmod. Because he precisely knows what he's doing with chmod, we should provide a little helper entry allowing him to poke around with all the chmod parameters, parsing them and applying them as desired. Thus, we just wrap chmod in a completion entry.

Other use-case:
A user copies data from CD, and wants to get write permissions.

Or:
A user wants to grant all people, or all people in his group read access to a particular directory.

Were these all use-cases? :)

> Can you come up with other use-cases? :)

Sure. Imagine family computer with several users. Each of them unloads their pictures from phones and digital cameras to the comp. They aslo want to share them, so that others can modify and see those pictures. As you already know nautilus don't play nicely with recursive permissions. So the question is why should your 12 years old daughter learn how to use terminal? Oh, right - she doesn't have to - just show her konqueror.

You see, this is a general problem (which started some 6 years ago I guess) with gnome development today. Somebody (novell/ximian) puts some disconected moron in charge of very important piece of software (like nautilus or screensaver). The maintainer then claims that he knows how users use or might use this particular software (in most cases he doesn't have such a knowledge). He ignores feature requests, bug reports, lots of feedback in forums, mailing lists and other places. Then he takes some functionality out of program and calls this regression a "feature". But the worst bit is that this new crippleware is pushed upon users.

Do such actions encourage participation in gnome? No!
Do they improve gnome's image amongst users? No!
And the most important question: can we still consider such a development model to be open? Unfortunately no!

Sorry for the harsh words, but asking for usability examples after 6 years of ignorance is just like spitting in users's faces.

Read "man umask" and set it up system wide the way you want it.

This particular bit of your animated reply

Most users don't care about the features you talk about.

seems to fall in this part of the original post:

The maintainer then claims that he knows how users use or might use this particular software

And how did you reply to that? Let's see, wasn't it perhaps with a

That's a straw man.

Funny, huh?

rehdon

No, the maintainer doesn't ignore people. But people just claim features without contributing anything. "Recursive permissions" were not rejected, but nobody except me came up with a patch as of writing, and my patch was still not ready for production use.

Others should come up with UI ideas, test patches, and so on, but they prefer to rant in this forum instead of doing something :Z.

Apologies if it wasn't immediately clear from my previous posts, my frustration is not directed against you but against the Nautilus maintainers. As I said I've given up on Nautilus for non-trivial file managing, that's the sad truth of the Gnome desktop at the moment (for me at least).

rehdon