Deciding On The Future of Linux

A reader writes: The Free Standards Group has posted a request for feedback, now that they have completed LSB 1.2 and li18nux is also finished. Where should they/we go next? "

Comments

Looks like we need a poll

[Hi_2k]

Should linux next try for a) a standardized and integeated windows like shell b) new functionality c) faster preformance(its starting to get fragmented) d) cowboy_neal based penguin logo

Re:Looks like we need a poll

[mao che minh]

Linux needs to just keep going down the same path that it has been for the past three years: excellent development that keeps making it faster, better, and more usable from number of sources. Existing projects should remain modular, just like the operating system they serve - KDE and GNOME, PostgreSQL and mySQL, tcsh and bash, emacs and vi, Evolution and Kmail, et cetera. The strength of Linux has always been that it has no epicenter, no "base of operations", no one publisher screaming at the developers to do it one certain way or to satisfy or one particular user base.

This "United Linux" shit and other talk of solidarity scares me, frankly. Competition is good, multiple sources of development with differing vision is good. We have all seen what assimilation gets you. I don't even need to link to their website, you know who I'm talking about.

Drivers...

[}InFuZeD{]

It's all about drivers and compatibility with those gadgets...

For example, the reason I'm running Windows now is because I can't get my darn Palm m515 to work in Linux, and I don't even know where to start looking for help with my Minidisc Player...

So it's all about compatibility with those gadgets in my book :)

Re:Drivers...

[cpeterso]

I'm surprised someone hasn't created a driver abstraction layer that enables the same driver code to work on Windows, Linux 2.4, and older Linux kernels. New Linux drivers often get backported, but each driver developer must do the same type of backporting work. Why not consolidate all the backporting into a single, shared (and debugged) driver abstraction layer?

Re:Drivers...

[ffatTony]

Do you mean something like wine?

helper program calling

[blystovski]

I currently use Windows for ease of use. With it, you can specify what programs you want to handle certain types of files, and the operating system remembers your choices. This greatly aides with the multi-media functions of my home computer. The last time I tried linux on my desktop, that was the one thing that annoyed me the most about the OS in general. There seems to be no standard way for users to specify what programs they want to use for certain file types, which would in my opinion greatly increase user productivity and decrease user frustration when using Linux on the dekstop.

Re:helper program calling

[hey]

I agree...

$ cd /etc
$ ls -l mime*

-rw-r--r-- 1 root root 7559 Feb 27 2002 gnome-vfs-mime-magic
-rw-r--r-- 1 root root 36823 Apr 15 17:47 mime-magic
-rw-r--r-- 1 root root 99960 Apr 15 17:47 mime-magic.dat
-rw-r--r-- 1 root root 12758 Jul 21 20:10 mime.types

and none of these files say what the helper apps is.

Linux or RiscOS or OS/2 or...

[Anonymous Coward]

Not to start a flamewar, but what efforts should we put on establishing Linux as the number one OS of the future? The more we follow in a single file, the easier we fall. There are some superb alternative systems around thesedays -- I particularly like RiscOS for ARM boxes. Very, very sweet and usable desktop, extremely fast (written in assembler), and has a decent range of hardware.

As a Linux user since 1997, I've been impressed by the strides made in the user-friendly desktop arena, but it only takes one man to win the game before the cows come home. I think we should all consider this -- as they say in football, it's a game of two halves, and Linux isn't necessarily the OS of the future.

Linux is in greater need though of uniformity between the multitude of administration tools. It hasn't made a great deal of progress in the last 5 years in this department, sadly. Fortunately, things are getting better in this area and we may soon be in the position to conquer corporate desktop deployment on a large scale.

So let's not base our future on Linux as a whole, but on the wide range of free and Open Source software we can apply to our systems. Linux is the OS of the future, but we must maintain our dignity and versatility in accepting other systems.

Just my thoughts. Mod accordingly!

Re:Universal Copy/Cut&Paste

[GigsVT]

If you're more inclined to use the ctrl-c ctrl-v methodoly

Yeah, but it's not consistant. Users should not need to know which GUI toolkit their app was coded in. There are also consistancy problems, some apps try to implement their own way of doing the clipboard, and it's entirely possible to have something on the "clipboard" (for lack of a better word) that is different depending on where you paste.

And then there is the issue of doing on-the-fly text replacements. As far as I know, there is no way to copy some text, highlight other text, and paste, replacing the other text with the text you copied. This sounds like something that doesn't come up much, but it really does.

Suppose I want to copy a URL and open it in a browser window that already has a URL in the Address box. In Windows, one would just double click the old URL to highlight it, then paste the new one which deletes the old one. Another way would be to double click the old URL to highlight it, then hit backspace to delete it, then paste the new one, but you can't do that, because that copies the old URL.

So... I usually just resort to opening a fresh browser window, or clicking at the end of the old URL and holding down backspace. Maybe I am just an idiot, but this seems stupid.

I'm not saying cut/copy/paste should act exactly like windows, but it should at least be consistant, which it really isn't. I've adjusted to using X and various apps now, so I don't notice it as much anymore, but when I first switched from Windows, it was one of the most frustrating things I had to deal with.

Single UNIX Specification

[sdcmk]

I think a good move for Linux would be to keep heading for the Single UNIX Specification from the Open Group.
It would make it much easier to port all of the existing UNIX applications over to Linux. Also, being UNIX compliant would give Linux creditability in the minds of corporations who are looking for alternatives to Windows but do not want to pay or cannot afford a commercial UNIX environment.

I just put in my big 2...

[trims]

which are:

Unified System Documentation I want all docs in a single, standard format that all programs must write their basic documentation in. No more man, info, html, pdf, ps or whatnot. I'd prefer a fixed SGML DTD (docbook is OK, but I'd prefer a designed-from scratch one specifically to address the system documentation target). That way, we can can get good viewer independence with modern features (hyperlinks, fonts, in-line graphics). All of the current formats are lacking in at least two areas, and we don't have agreement on which to use. This is a big place for them to step up.

Standard Config Files No, this is not a request for a Registry (the merits thereof are for another discussion). What we want here is to get rid of the 80 billion different ways to write a config file. I'm sorry, but they all should be a nicely tagged XML (or similar) file nowdays. It sucks to have to figure out the idiosyncrasies of the various config files. This issue isn't simple, but is definitely a place where a good discussion is needed.

-Erik

Re:I just put in my big 2...

[Spy Hunter]

Amazing! An educated, informed, reasonable comment! I congratulate you, sir! And I echo your calls for standardization of config files and documentation (GNU info is an abomination and should be taken out and shot).

KDE has something that makes man pages a little more palatable. If you type a url of the form man:/command into a Konqueror window, you get a rendering of the man page for that command in HTML. Then you get colors, hyperlinks, nice formatting, the ability to dynamically resize the page, a nice search function, a back button, a scroll bar, mousewheel support, and all the other niceties of a modern browser. If the documentation was in a better format to begin with, one that had more ability to specify hyperlinks and graphics, this would be the perfect documentation browser.

Is there anything big still missing?

[SmallFurryCreature]

Most of the comments seem to be along the line of "Who the hell do they think they are". These comments are crap as the posters obviously have not read the post. They do not claim nor have ever claimed to own a particulair piece of software. They are just intrested in creating some sort of standard. If you like youre linux to not conform to that standard then that is just fine. Just as ANSI is not a law neither is neither is the LSB. For the rest of us it makes it easier to exchange between the different flavors of linux if all the files are in roughly the same place.

Others seem to want to turn linux into windows. If only (mime support/windows like shell/c:\Program Files like dir structure) was finally included I would start using it. Yeah right like anyone cares. I think that with the burst of the internet bubble the idea that linux should go to the masses has been left behind. If you saw the interview with Linus himself on the BBC you will have heard that he does noet even wish to compete with windows. MS has its market and linux has its own. That is real freedom of choice people. Those people that want linux to become like windows just want a gratis (not free) version of windows.

The FSG is a standards group, I presume therefore that their question is on what if anything needs standarization next. Standarization is not the enemy of freedom when standarizing on it does not put a brake on innovation. A standard desktop for instance would limit innovation and therefore choice. A standard directory layout does not unless I missed some special signifigance in keeping youre logs in /.[sic]

So what needs standarizing next? I have no idea. Software creators now are reasonably sure where to install the bits of their software and how they can achieve multi language support. Printing is also ridicously easy (could be because I only have access to HP printers). Is anything more needed, almost certainly, let the creators figure this out and not disturb them with a dozen wish lists by windows users who will never switch over because it will always be hard to switch to something wich is different. If it wasn't different then what would be the point of switching at all.

Use linux not because someone tells you to. Use linux not because you want to stick it to Gates. Use linux not because you want to be l33t.

Use linux because you like it strenghts and can forgive its weaknesses.

I couldn't agree more

[Dave_bsr]

Modularity is what makes this great. Being able to use any shell, or any WM, or any text editor...that is what makes this free software stuff so beautiful - you can run a great system that you have tweaked exactly how you like it. and so can I. and while I can run the same apps you do, my setup is Icewm with minimal graphics and very few "extra" applications running, and you'rs is KDE with every little graphical doohicky cranked up. I just think that modularity is beautiful.

Remember the industrial era? it was workable because each widget 5A was made exactly like every other widget 5A. you could assembly-line, you could swap out one part for another, and the final product worked the same. Maybe this is an idiotic illustration. But I think lots of little peices making up one great big OS is just nifty. that's all.