Slashdot Mirror

From memory, the MS system has a mechanism for key recovery and/or an admin back-door. This makes sense in an office situation (e.g someone leaves or is fired) but it still leaves me wondering about other backdoors. It doesn't sound like paranoid-tinfoil-hat-crowd level encryption to me.

Well, sort of. IIRC, EFS can store a second copy of an encryption key encrypted with an administrator's (key? password? Been too long, but the interface exposed to the administrator should be the same.) I'm not an EFS guru, but I believe that the internals of the storage structure are public ... I don't see how or why Microsoft would start backdooring the thing, especially when they do things like sell Windows to military and other clients (granted, they may have non-EFS policy, but still...).

dm-crypt was introduced in kernel 2.6.4. This uses the new device-mapper API to encrypt a file or device. From what I've read, it's a much cleaner implementation than cryptoloop.

Could be -- thanks for the link, as this is new to me -- but when the author is talking about cleanliness, this is a kernel hacker or at best a developer interface. The problem is that there is still no setup done by the distros to make this particularly accessable to the users -- I don't want to have to manually play with PAM on each system to convince it to hand off a password to an encrypted loopback layer, and really, even the most do-it-yourselfer would probably prefer to have certain basic things (particularly related to filesystems) automatically set up -- maybe kept in the open so that the workings are accessable and configurable, but this is not a process that benefits greatly from a lack of a standard mechanism or an easy setup.

As you point out -- these are block-level-device interfaces, and the code needs to be filesystem level. Dunno about whether Linux can do it, but it might be possible to produce a "pass-through" filesystem that does encryption on files (though directories wouldn't be feasible...dunno how EFS treats dirents). Dunno if Linux filesystem code can call back through the VFS, either, and if not that'd shoot this down. This approach would be nice and generic, at least, so you get encrypted filesystem-level encryption under just about any filesystem, though the interface'd be a titch less nice for the fstab user than having encryption features just available in each filesystem.

About virtual terminals: I gather that most distros now use gdm, kdm, or even good old xdm to provide an all-graphical login. None of this clunky startx stuff. Much more prettier looking as well :)

This is true; I mention it because it has been a significant problem in the past (oh, and folks that use xhost + are still around, too).

About user switching: I think KDE recently (last year?) added a "switch user" feature. I'm a Gnome user, so I'm not really sure. I know I've seen it recently, but I can't find it in the Gnome foot menu here. So it's probably in KDE.

Hmm. Interesting. Looks like it provides a front end in the screensaver for firing up a new kdm instance in a new xorg instance.

Apparently the GNOME project has its own efforts underway in the same direction.

Ever heard of development versions?

i tried suse the other week. modem and wireless card don't work with linux. not buying new ones. back to windows for me.

It's available in Epiphany.

You can now see a user's online status inside KMail and inside KAddressbook, if you have Kopete running. You can also click on an address to IM them, from inside either of these applications.

To me this is a killer feature. Gnome still has a bounty out on it ( http://www.gnome.org/bounties/IM.html

*sigh* I'm getting a headache from you "I don't understand spatial" guys.

The spatial metaphor works better when it's sitting on top of a meta-rich environment. That's were the GNOME storage project comes into play. Maybe one of these days, you'll all realize that developers aren't stupid (doesn't stop you from using their software I noticed). You all chewed and spit out the Mozilla team for XUL and other decisions (Surprisingly no one left because of it). Now it's the GNOME teams turn.

Well, it certainly never made it into redhat9 - which was a very gnome-centric distribution, and is not that old.

I'm pretty sure it DID make it into RH9, but certainly not going to install that one to check...

lists, default and mandatory settings couldn't be changed until recently (in CVS, about 2 to 5 weeks ago).

This response illustrates my point far better than anything I've said. How many years has gconf been around? I would have thought that configuration of a configuration system would be a higher proirity than crafting a windows GUI registry editor lookalike - but that's looking from a perspective where everything can be a file and a web browser is an application and not part of the OS.

Really? Isn't it just the opposite: the GUI registry editor lookalike is indeed a lower priority than the underlying configuration system and that's exactly it took a long time to implement those in that gui thingy? Anyway, looks like I was wrong about lists, gconf-editor in FC2 does support them and is older than 2 to 5 weeks.

Defaults and mandatory settings are for sysadmins, usually one-time thing after installation and are probably scripted anyway so nobody missed gui for them, easily doable with command line tools.

Well, it is very much an old school unix style of things, and if the gnome developers were like that they would have started with flat files before progressing to a database, instead of starting with the then trendy XML thing

You can't get much farther from old school style of things than having one big database file that stores preferences of every application! XML tree is somewhere in middle ground between of that and flat files.

which defeated the whole purpose of XML by only being changeable by gconftool/gconftool-1/gconftool-2.

I really can't see what's supposed to be so unchangeable, but just in case, gconf developers agree, simplification of the XML format is on TODO.

gconf is, indeed, by it's very nature for storing only user preferences, instead of every last setting on system, if that's the same thing as "toy for keeping track of icon positions", I do not know, system-wide daemon (instead of each having own gconfd process, I assume) however seems to also be on that list...

Neither the registry editor lookalike GUI nor xml backend are the main focus, they can be changed, or removed, but the fact that apps have a consistent and easy to use API that implements necessary things for reading and storing configuration.

But it's obviously at least partly a matter of taste, and you can't convince someone who doesn't want to be convinced, I didn't intend this to be a semi-religious flamewar, just needed to point out that "gconf != need to reboot"

linux has usability standards? problem is there are probably 5 of them and 90% of linux apps don't follow any of them

Why don't you check out the GNOME Human Interface Guidelines. KDE has their own, but both share a common base-set of standards outlined here.

Stick to gnome applications under a gnome desktop, everything should be fine. The most disturbing thing for me is Balsa, written with GTK+ 2, has an address book manager written in GTK+ 1.2, which sticks out like a sore thumb.

Linux apps have some way to go, but the situation is a thousand times better than the mess we had a few years ago. It's getting there.

They have not purchased licenses they are obtained currently royalty free, see here. Will it stay royalty free? Seeing as it's MS probably not.

Which version implemented this? Is this version only available in CVS or has it had a wider release?

Well, there's a CVS commit with comment " Added support for editing keys, adding new keys, and unsetting keys." dated 2.5 years ago, no idea which version that corresponds to, but it's been quite a while since that anyway. So I'm guessing you must be referring to something else entirely... lists, default and mandatory settings couldn't be changed until recently (in CVS, about 2 to 5 weeks ago).

Good, so something sane like berkeley db

You mean something like this? It's long since been removed though, simply because nobody used it.

can be used apart from the format which until recently (if you are correct) was only changeable with undocumented hacks. There are/were a lot of keys you just couldn't get to any other way.

I wouldn't exactly call 2½ years "recently" but whatever, and those command line tools are not anywhere neal the level of "undocumented hacks" even if they do (did? seemed fine for the very simple thing they do last time I checked) not have the best documentation. And if you're going to use bdb-API (I sure hope you weren't planning to db_dump, text edit and db_load) you are doing "hacks" anyway, why not go trought the gconf itself?

Tasklist can group icons together when multiple instances of a program are running. A number in parentheses appears to next to the application. Clicking on the icon brings up a menu listing all of the running instances.

Java becomes more like a usable programming language every day!

The functionality of mailcap, later cloned in win32, GNOME and KDE, is now available to Java as JDIC!

And JDNC provides powerful GUI functionality, including an XML format for describing GUIs, a la Glade, wxWidgets XML resource files, XUL , XAML. Of course, they are all completely incompatible. Also worth mentioning here is GNU Enterprise, which has seemingly similar aims to JDNC.

I don't know what all this has to do with ActiveX. For all I know, ActiveX was MicroSoft's intended replacement (or umrella) for Java and JavaScript. It has met with mixed success, having met with popelarity mostly as a malware target. JDIC and JDNC look like decent proposals to me. It's good to see Java maturing both in speed and functionality. The more choices, the merrier, after all.

Havoc Pennington recently stated that he also sees legal issues by using MONO technology in GNOME read here and here.

There are another few comments made by Havoc Pennington this year but I wasn't able to find the links. So you people should better be sceptical by using MONO technology because MONO is moving in the gray area here and is probably becomming a big legal issue.

Nat Friedman's dashboard is absolutely nothing like Konfabulator. Doesn't even remotely do the same thing. The OS X Dashboard is basically a rip-off of Konfabulator, yes. But Nat's Dashboard and Beagle address the problem of searching/indexing metadata on the entire system and not throwing away data that can be useful later for searching.

GNOME a dissapointment.

Well, the GNOME Human Interface gudelines and the KDE User Interface Guidelines are there exactly for this. GNUStep probably uses the OpenStep ones, since it is an OpenStep replica.

Hint: in absence of a single authority having the power to dictate how thing should be done "OR ELSE", this is the only sane way to do it: define reasonable guidelines and tell developers about them: many will follow.

Or Open Office

They are both free alternatives and I'm sure I'm the first to inform everyone of these.

Try out Gnumeric - it's a native Gnome application but is rather lost because of OO's "mindshare".

One of the best spreadsheets for linux, gnumeric has support for 100% of Excel's functions as well as most of its other features. Its one of the highest quality and most stable pieces of software I've ever seen for linux. Its amazing they overlooked this as competition.

50 bux for a spreadsheet app? I'll stick with the free Gnumeric instead.

I really love KDE. I just wish that ARts was more easily capable of using network transaprent sound. I know that it should have the capability, and I have seen posts like this but, somehow, all the guys who'd love to see it happen for thin clients can't seem to make it work.

Esound is not very good, but at least it works easily over a network.

If I'm wrong (plus all the other guys working on this), and there's and easy answer, I'd love to hear it, but until then, I'm lashed to GNOME and IceWM, and my favorite KDE sits out there in the"want it, but can't have it" pile.

a.g.o also has some really nice stuff.

It's possible, but slow, to trawl through the sources at ftp.gnome.org

My preferred method of getting the latest GNOME stuff is garnome.

You still have to wait a day or two while it is compiling, but at least you don't have to download everything youself. And you get all the goodies of compiling from source, like switching on optimizations and stuff.

This way I can have the latest GNOME desktop even on my Debian Woody machine.

The GNOME devs are actively working on this issue.

See for example

http://mail.gnome.org/archives/desktop-devel-list/ 2004-May/msg00028.html

http://mail.gnome.org/archives/gtk-devel-list/2004 -April/msg00065.html

You should start seeing the improvements in 2.8.

The GNOME devs are actively working on this issue.

See for example

http://mail.gnome.org/archives/desktop-devel-list/ 2004-May/msg00028.html

http://mail.gnome.org/archives/gtk-devel-list/2004 -April/msg00065.html

You should start seeing the improvements in 2.8.

If you run something like irssi in screen, and then open a second tab in the window, you will have the first window turn entirely blank-on-blue background.

These two bugs refer to this problem. Apparently setting TERM=vt102 helps, but this problem keeps g-t from being the 'perfect' terminal emulator for the GNOME desktop (it means at least one person I know uses konsole on gnome.)

If you run something like irssi in screen, and then open a second tab in the window, you will have the first window turn entirely blank-on-blue background.

These two bugs refer to this problem. Apparently setting TERM=vt102 helps, but this problem keeps g-t from being the 'perfect' terminal emulator for the GNOME desktop (it means at least one person I know uses konsole on gnome.)

Have you used the Bitstream Vera fonts? They are truly brilliant. I use the monospace for all my terminal windows.

This is great! I certianly wish more companies would adopt this kind of a driver development model.

I know the GNOME Foundation has also done a similar bounty system, recently.

This is great! I certianly wish more companies would adopt this kind of a driver development model.

I know the GNOME Foundation has also done a similar bounty system, recently.

Stolen today from Calum Benson's blog:

Q. Why do women love England footballers?
A. Because they're on top for 90 minutes and still come second.

... I'd recommend you spend some time using it to actually _manage_ files.

As one of the gnome devs points out, when people test a file manager, they often go and browse around their files. If they do this using spatial, they'll come to the conclusion that it sucks. But that's because spatial _does_ suck for browsing files - if you want to look for something, use the file browser (it's right there on the main menu).

But spatial is incredibly good for day-to-day file management. I finally got round to reorganising my home directory yesterday, and it's incredible how easy spatial made it (after all, file reorganisation is a task which you _want_ loads of windows open for).

So, before you attack spatial nautilus, try reorganising a few directories with it, because that's the sort of task it really shines for.

Every one of your concerns has been addresesd in the design. It may not be your cup of tea, but do give them a little credit. (Though it wouldn't hurt if they made some of these features more obvious)

• When I want to open a folder, I don't want to open every folder before it in the tree -- too many extra clicks.
  If you actually use the folder a lot, add a shortcut on your desktop. Otherwise, you can always enter a location with the familiar C-L (and tab completion works).
• Similarly, I don't want to have to close all those folders which are open for no reason, again too many clicks.
  Control-Shift-W will immediately close all ancestors of your folder
• How do I quickly go back to the parent folder? Oh, it's in the menu. Three clicks.
  Alt-Up. Or, if you want to close your current folder as -you do it, Alt-Shift-Up
• What if I want to go three folders up? Three menu clicks!
  Alt-Shift-Up three times
• And then there are the problems with the "Filing Cabinet" analogy -- if my filing cabinet at home had well over a thousand folders in it, and happened to also have folders inside folders (and honestly, what kind of a real filing cabinet has nested folders?) then I would take a real long hard look at my life. Computer files transcend real filing cabinets.
  Granted. But the Gnome HIG acknowledges that when using metaphors (like the desktop), "it is important to neither take the metaphor too literally, nor to extend the metaphor beyond its reasonable use." The guy who wrote this article's just stupid -- the official HIG are much smarter.
• Windows get in the way of other windows. Too many windows. Have to move windows around (and find windows that get behind other windows... silly windows) just to copy/move/open/reach a file.
  Using the Ctl-Shift-Up and Ctl-Shift-Down to close windows as you open them will help here, as does Ctl-Shift-W, but that's beside the point. One good thing about spatial is that it remembers where you put windows. If you position them in a convenient, non-cluttery way (i.e. camera USB card opens on the right, "Pictures" opens on the left), nautilus will remember forever.
• Most of the great little shortcuts I'm used to (yes, from windows) have been removed. Example: I'm in a directory of, say, 10,000 files. I want to get to the one called "testnumber5384.c". I start typing the filename, expecting the file manager to know what I want and automatically jump to the files fitting my typing. Nothing happens. I sort by name, then proceed to scroll through 5,383 other files (an imperfect science at best) before I finally find the file I'm looking for.
  Sounds like your Gnome is broken. Typing should work as expected, and does for me.
• And of course, where the heck are the hidden files in the file chooser??? Forget opening any config files in gedit (not that you would).
  Hidden files are hidden for a reason -- your average user doesn't want to have their Home directory filled with as many files as they have programs installed. But if you want to see them, there's an obvious option on the very first page of the preferences dialog

Sometimes we underestimate our users, anyone who's using office is able to use openoffice. It'll be an adaptation time and maybe some complains that things are not were it used to be, but people are not likely to changes. But they did it before, I remember Wordperfect, Amipro, those were killer apps, but died and people got used to MS Office.

And if it's necessary people get used to openoffice and gnome or kde more easily than you may have expected.

So I guess the term for Linux is "feature-rich" but the equivalent term for Windows is "bloated".

Research before you insult someone.

What are you talking about, its not like the're sticking bloging or P2P into the GUI or anything dumb like that....

If there is anything Epiphany is *not*, it is not clunky. It is downright unclunky, to coin a new term. It's the unclunkiest browser I've ever seen. Perhaps you are confused... you sure you don't mean Konqueror? If so, I understand what you mean.

Epiphany is certainly not as full-featured as other web browsers, but it's damn good at *browsing*. That is the point, read the manifesto. The lack of features is not a mistake, nor is it a result of laziness -- it's due to a different set of design preferences than you are used to. That doesn't mean you should use it... what the fuck do I care what you use? I do think that your particular criticism of it is unwarranted though.

1. The Menus should be much more customizable; treated like folders that you can click and drag into (I hate to say this, but "Like Windows").

2. Better Video control properties; take advantage of XFree's extended features and have options like TV switching and such.

3. Better preferences; the control panels are quite lacking.

4. Other aesthetic enhancements that will make gnome pretty enough to compete with other window environments (like win XP's or OSX's). Smooth scrolling, the zoom-on-hover icons in OSX are sweet, and _drop shadows on windows_ would be real nice.

5. Some kind of Linux-version-of-Active-Desktop would be real nice, so I could have an IRC session running as part of my wallpaper,anchor the weather channel radar map to the background, etcetera.

No, it is not.

It's a book so named that Gnome advertises on their front page. That makes it official to me.

First off, working with Mozilla Firebird is a stroke of genius. There are a heck of a lot of man hours being put in on that project, we should utilise them rather than recovering ground already trod upon by the lizard.

I'm not so sure -- Epiphany and Firefox are very different projects. Certainly they have similar stated goals, but the execution shows that Epiphany is *serious* about them. When you see Epiphany you first notice all of the features that are missing. If you're like me, you'll quickly notice that you don't really miss any of those featues, which underscores their dedication to the goal of creating "the simplest interface possible for a browser." (more)

I use Firefox under windows at work and Epiphany at home, and so I have a lot of experience with both browsers. And they are both very good products, don't get me wrong. But Firefox is much more of a traditional browser, with the drop-down URL bar, the nested bookmarks, and the XUL theming. I can't really see them dropping those features as a Mozilla project so I'm not sure what they have in common with the Gnome desktop.

A solution that would be beneficial to everyone is to have a shared gecko engine betwen the Moz Suite, Firefox, and Epiphany, and a shared system for implementing plugins. Obviously only certain kinds of plugins would work on a non-XUL implementation like Epiphany or Camino, but other plugins like the flashblock or useragent-switcher should be possible to port.

With a common engine between the three you give users the option to choose how they would like to browse. Depending on their preferences some like the Moz suite, some like Firefox, and other like me prefer Epiphany. There is a lot of room for valid differences in preferences and I think all three browsers have a legitimate place.

Because of this I don't see how Firefox could become the default Gnome browser. They could HIG-ify the dialogs and create a Gnome theme, but like OSX the pinstripe theme is no replacement for actual integration. Collaboration is best left on the back-end of the project, let there be multiple different front-ends for different purposes.

A similar tool already exists - check out GNOME Blog.

Or even - better - Blog right from a panel applet. http://www.gnome.org/~seth/gnome-blog/

Although the official reason that GnomeBasic was dropped was because of "stagnation", the real reason that it died was because Mono was supposed to take it's place.

If that happened, I've seen no evidence of it.

While you can write Mono code in Java, PHP, Logo, Oberon, Pascal, Forth and Lisp, VB is still unavailable.

It's a pity such a popular language appears to be entirely ignored.

We decided nothing of the sorts. On the contrary we've gone great lengths in trying to make Emacs-keybindings usable for those who use them. Thanks to conflicting shortcuts in the HIG and the way the keypress events are handled by default in gtk+ it's not exactly nice.

Just see src/galeon-window.c(galeon_window_key_press_event) if you're morbidly curious.

We decided nothing of the sorts. On the contrary we've gone great lengths in trying to make Emacs-keybindings usable for those who use them. Thanks to conflicting shortcuts in the HIG and the way the keypress events are handled by default in gtk+ it's not exactly nice.

Just see src/galeon-window.c(galeon_window_key_press_event) if you're morbidly curious.

Oh, and FYI

GNOME DID IT FIRST!

We don't use the Apple HIG because we use the Gnome HIG.

Oh, and FYI

GNOME DID IT FIRST!

We don't use the Apple HIG because we use the Gnome HIG.

The item I like about Xfig is I can create template objects quickly and easily and add them to it library of objects. The last time I tried to make an object for Dia I just gave up.

Now if Inkscape could export to the .xfig format and Xfig to the proper .svg format that would be great! Using both tools would save me sometime.

You may enjoy memerizing key bindings for umpteen million different apps, but I don't and neither does the majority of the unix desktop's potential market.

As I have posted elsewhere, ctrl-c/ctrl-v is the standard for GNOME. I believe the same is true of KDE.

KDE and GNOME apps are what the "majority of the unix desktop's potential market" will be using. Your complaints about the state of copy-paste in unix are from 1995 and are disingenuous at best.