Slashdot Mirror
OSDL Announces Desktop Initiative
rhetoric writes "Earlier today at the LinuxWorld Conference & Expo in New York, nonprofit Open Source Development Labs announced the creation of a "Desktop Linux Working Group initiative focused on greater use of Linux on desktops throughout the enterprise." A press release is available on OSDL's website, in addition to this Register article." It's all part of their non-secret plan.
We all know how good committees are at deciding things. Compromise usually leads to the lowest common denominator. Do we really want a Linux desktop that's designed by a committee? Isn't that against the whole spirit of free software?
>>esr>>
The Register article says that the OSDL is setting out to crate a specification for what an enterprise Linux distribution should be made up of. Yet the Register article also implies that the OSDL is not going to receive much direct user input on the spec itself. Is this going to turn out to be another UnitedLinux?
SCREW THE ADS! http://adblock.mozdev.org/ Proud user of teh Fox of Fire - Registered Linux User #289618
No, the spirit of free software is that we don't have to use it if it sucks, because we can just build something similar, but better. (Or something completely different, of course)
Programming can be fun again. Film at 11.
I agree with you - if there's no user input on how the distro should be laid out, how it should interact with the user and how similar it should be to Mcrisoft operating systems, then how is this going to succeed?
SCREW THE ADS! http://adblock.mozdev.org/ Proud user of teh Fox of Fire - Registered Linux User #289618
Needless to say, as long as Linux distributions and desktop managers continue to proliferate, the average user's requirements will never be met. I say this as a *fact* not a *prescription*, so spare me the Linux-strength-in-diversity comments. I just think you can't have your cake (freedom/diversity) and eat it too (Linux on average desktop).
/one/ desktop (or suite of software options) to make it easy for these users -- perhaps it doesn't exist yet, but when it does, it can be used, even by the majority of users, regardless of whether or not there are other options.
I'm sure I'm not going to be the only one to disagree with you on this point. Although I agree that the current Linux desktops may not be ready for the users you describe, I don't agree with the leap of logic that diversity necessarily means it will never be ready. That's just silliness IMO. All it takes is
The "unify or die" logic has always seemed like a non sequitur to me.
The big problem isn't that Linux is particularly bad or anything, it's that many, many people already have Windows. As bad as Windows might be, it's really annoying and fear-inducing for bosses to imagine taking down all their machines and installing a different O/S on them. The meeting should be on how to get people to switch o/s's, not how to implement linux. Once people aren't scared to switch an o/s, then all will be well.
stuff |
No-one has said that the CLI has to go away. MS Windows CLI may not be as full-featured and may be hidden in the depths of the start menu now, but its never disappeared. What we need is for the CLI to not be essential, and for a GUI to be available to the end-user for most if not all tasks.
(Mods: this is on topic - bear with me)
So off I went to France. It was beautiful, perfect snow, lovely mountains, perfect pistes. I put on my snowboard and started to learn.
The problem I came across was that I couldn't do anything I wanted to. I could see where I wanted to go (I wanted to hit the slopes dammit!) but I completely lacked the skill required to get there.
After half a day, I'm ashamed to say I gave up. I was only there for 3 days and i'd wasted some of that precious time getting absolutely nowhere. So I put on my ski's, hit the reds and blacks and had a fantastic 2.5 days.
Linux is like that for me. I like it, I want to use it, the problem is that I think of it as a tool to do something else and I just end up getting frustrated because i can't do the boring things really quickly because i'm too ingraned in the Windows way of doing it.
I can change the display resolution quickly in Windows. I have to faff about in Linux. I can install items in Windows with a few point and clicks. Everything i've tried to install under Linux has botched up through my own general incompetance. The very basic of things takes 5 times as long and I get frustrated and eventually switch back to Windows (I still can't dial up under Linux, it refuses to recognise my external Hayes modem and KPPP dies horribly with some error message - the Gnome one hangs on startup).
Whilst Linux on the desktop might not be totally there, it's biggest problem is not that, but of people like me who don't have the patience to learn how to do the things (that they can do really quickly under Windows) differently.
Avantslash - View Slashdot cleanly on your mobile phone.
The average user wants to do everything grandma wants to do, but they also want to be able to install or upgrade software and hardware *easily*. There are lots of package management tools out there that make installing software on most distros very easy (emerge, apt-get, red-carpet etc). Most of them have nice GUI front-ends as well, allowing you to point-click-install.
Linux is ready for *some* desktops only, namely ones where users won't be constantly tweaking and installing new software and hardware.
in other words, linux is ready for the healthier installations, and not ready for sick computer use.
constant tweaking and installation of new hardware and software is not using a computer productively. using a computer productively means that the computer is operational, in the sense that it doesn't actually -need- anything further to be done to it in order to function as intended. it just works.
this 'just works' state is fairly easy to get to with linux, and other open operating systems, and stay there. but i can count on one hand the # of windows users i have personally known who can point at their aging computers and say 'that system just works, so i leave it alone and just use it'.
15 years of Windows use may have given you a neurosis, a false standard, with which you are comparing other platforms where such inflictions aren't really a priority. of -course- you can tweak and install software in linux; indefinitely in fact.
but the point of computing, and of computer use in general is to -USE- the computer to do something, not be continuously servicing it. this is a fact that seems so simple, yet for most window-dwellers, it often appears to be inseparable from 'actual use' as a concept.
i blame microsoft of course. tweaking and upgrading and re-installing and installing and 'software choice' is just a way of getting you locked into a constant crackhouse fit.
The average user wants to do everything grandma wants to do, but they also want to be able to install or upgrade software and hardware *easily*.
why should they? ubiquitous, cheap computing (s/cheap/inexpensive) means that once you've set it up and got it working, you can leave it alone and just use it.
people are starting to see that the windows treadmill is a trap. once you get sucked into a windows way of life, upgrades and re-installs and tweaks and fixes all seem to be 'normal' ways to use the computer, but in fact this is really a detraction from the core issue of computer science, which is 'how can i use this computer to do the job in front of me?'.
microsoft, and others in the industry who have been around long enough to have weathered countless waves of API and hardware technological changes know that computers are a constantly-changing product. its like a lump of magic matter which never maintains its state long enough for it to become a fixture.
but this is not the linux philosophy. the linux philosophy is: get it working, and once its working, use it.
; -- the corruption of government starts with its secrets. a truly free people keep no secrets. --
We all know how good committees are at deciding things. Compromise usually leads to the lowest common denominator. Do we really want a Linux desktop that's designed by a committee? Isn't that against the whole spirit of free software?
We are talking about industry standards. Outside of the software industry, they work very well. The software industry is still in it's immature stages - like the engineering industry was before there were standard sizes for nuts and bolts - manufacturers actually deliberately made their nuts and bolts incompatible because it gave them lock-in, just as the software industry does now. Incompatible nuts and bolts seems crazy now, closed/incompatible file formats will seem crazy in the future.
Yes, the engineering world has discovered that it's far more profitable to standardize bolts but proliferate drive types.
The consumer never feels they're locked in, but has to buy mulitple sets of tools, if only to remove the patented head bolt and replace it with a standard one. Very profitable for the patent holder.
The software industry seems to be learning this trick.
Witness XML, a standard for creating standards. You can claim XML compliance and yet extend it in propriatary ways. It is plain text, but the file sizes are truly gargantuan, so you need to compress them, for which you can use your own propriatary compression method. Certain outfits are now even starting to create propriatary XML parsers.
The bolts are all standard, but you still have to buy the tool.
There's one essential difference between file formats and bolt design though. Bolts aren't given extraordinary protection by the DMCA.
KFG
From the Register:
... meaning, no pop-ups, no more browser hijack? (I sure will miss 'em)
OSDL says it isn't out to create a Microsoft Windows replacement.
I don't get it. though there are some things for ODSL/Linux vendors to learn from Windows, there are very many things that they would NOT want to learn or copy from design of the monopoly OS to replace it. Desktop Linux should not become a Widows replacement for god's sake.
Linux is superior to Windows in many aspects, while Windows has some advantages in desktop use over Linux. For business computing, security can be the primary concern. I don't say either OS is more secure than the other, but the history shows that Windows is more likely to become a target of vulnerability attacks by hackers/ spammer/ ad agencies, and that some attack attempts successfully created mess. Even though there are far more applications available in the market, Windows wouldn't be a choice of OS if I were to make decisions.
Let's face it, how difficult would it be for a person of "computer literate" (according to his/her resume) to learn how to maneuver KDE/Gnome? I don't think it would take a year. If applications are network-installed, employees on the terminal system won't have to worry about installation of application. Let the IT dept. take care of it.
If you have hundreds of Windows apps to run, use wine, codeweaver(also wine), or vmware. Running Windows on vmware/virtual PC gives you access to Windows apps and ease of security control under Linux at the same time. It's certainly better than getting hacked and filling your monitor with a bunch of pop up ads and crap because you are using Windows, or Windows replacement.
Don't make a replacement for the 'every-user-has-root-access-by-default' OS. Just let people learn and replace.
Whilst I agree to some point, there are still plenty of people who use Windows 95 and 98 and shouldn't be considered stuck in the "windows treadmill".
The real treadmill is the hardware one for games. If you want to keep up with the latest and greatest games you have to fork out the cash for the faster better hardware.
People spend far more on this than they do on their operating system (which is generally the one they got with the PC and the one that is on there when the PC dies or is canned for a newer, faster one).
So whilst i agree with you that there is a treadmill, it's not really as much of a trap as you make it out to be.
Avantslash - View Slashdot cleanly on your mobile phone.
> We have RPM on Linux, and that should be okay to use here.
RPM's are awful. Distrubiting binaries instead of source in general is awful. The second you've linked it againest a different version of a lib it's usually unusable. The GNU build system takes care of that much better.
> An active-directory similar interface for distributing software packages.
Novell is in the process of porting ZenWorks which from what I understand can do this. Though I would find it hard to believe someone hasn't already written something to do this. It could be accomplished in a 200 line perl script.
>X crashes much too often
This I find VERY hard to believe. Xfree86 has definatly taken stability over features and has taken a lot of slack for it, but it definatly is stable. I'm running a RC (4.4.0 rc2) and don't even have problems with it. I ran 4.3.1 before that for a long time, never had problems with it. It's very agressivly compiled too. Some binary packagers have a way of turning a good program into an unstable binary (I'm looking in your direction Red Hat). Try compiling Xfree86 from source with just Make World.
>the time to restart X seem to match the time to restart a normal Windows XP computer
How are you restarting X? I can kill X with ctrl+alt+backspace and startx again in under 5 seconds.
That's just a random subset of things that we need in order to provide a quality desktop that most non-trivial/non-grandma users do. There are a million and one other things we need as well.
In short while a huge amount has been accomplished, there's still a huge pile left to do. Still, it's not as hopeless as it looks - the distance Linux has come since I started using it only 2 years ago is incredible. Beautiful fonts, cleaned up desktops, hugely improved artwork, maturing applications and powerful media players are just a few of the achievements I can think of.
To create a world-class desktop, an overhead vantage point is needed. I guess this'll be a stretch for the development model of free software.
Windows and Mac heads are used to a VERY strong cross-application cut and paste.
Windows has often - (no sarcasm) - exceeded expectations in this area by allowing all kinds of data to be intelligently moved from one app to another.
It's something designers rely on and use all the time.
Now is the winter of our disco tent
You make the case that the 'treadmill is great for games'.
And then you conclude:
So whilst i agree with you that there is a treadmill, it's not really as much of a trap as you make it out to be.
Sorry, but games are a trap too. You're lured into them for the purposes of 'entertainment and fun', and end up wasting countless, countless hours playing them. And when you are finished, you have nothing to show for it except memories. Of a completely artificial reality.
I'm not implying a value judgement here - it is for every human being alive to determine their own tolerance for the world around them. I'm just saying that both a) video game playing and b) windows upgrades are an industrial treadmill which produce very little except profit for their executors...
; -- the corruption of government starts with its secrets. a truly free people keep no secrets. --
We have to get more home users and more small businesses on, and that means improving the desktop, getting the applications they need built and 'marketing' it to people.
I would argue that focusing on larger business desktops would be a quicker route to the home users. People will buy a home computer to match what they have at work, not the other way around. In an enterprise, computers are chosen carefully for consistency and hardware compliance, so the desktop will run well. With a home user, you have to support every little webcam widget sold at BestBuy or Walmart, or else you lose.
If we get millions of corporate desktops established, hardware manufacturers will start supporting it, and people will start buying Linux for their homes.
And while I don't think we need to standardize on a particular window manager or desktop environment, at a minimum the Gnome/KDE environments should share lots of standards, such as clipboards, stanadard dialogs, themes, etc. I think this is what OSDL is trying to do, which is similar to freedesktop.org. Having multiple GUI toolkits is not necessarily a weakness; the same condition exists on Windows and that hasn't seemed to affect it negatively.
Hardware support?
I've just installed Fedora Core 1 on a 5-year-old PC at Work. The sound card doesn't work.
Hopefully as Linux grows in popularity, we'll have better hardware support and not just written by private individuals (as great as those people are). I mean actual drivers written by manufacturers to support Linux. New devices coming onto the market with an easy-to-install driver (which means no command-line installation, guys :-).
Easy addition of new hardware?
I bought a USB Flash drive a couple of months ago. I plugged it into my Win2k box at work and it worked without me having to do ANYTHING! I tried the same thing on Linux, I had to Google for instructions on how to do it which said something along the lines of "add x line to your /etc/fstab file, create a mountpoint then mount the device on the command-line."
Easy addition of new software?
OK, I'm using Fedora. I've got a nice computer so I like lots of nice software. If I'm lucky, the software I want to install will have an RPM available. So, I download the RPM and double-click it (yes, I don't use the command-line for installing most of my software!!!).
If I'm lucky again, the software will install. If I'm unlucky, the RPM will have dependancies I don't have installed in which case I have to go hunting for more RPMs.
apt has solved this problem somewhat so now I can just run:
... and the software should install. If I'm unlucky again, the installation will work but won't put any icons in the Gnome (or KDE) menu, so I have to go hunting around in /usr to find where this programme has installed itsself and manually add entries into the Red Hat Menu.
While Windows is not an OS I like, it does have things like software installation working reasonably well. While it still has problems with dependancies (this game needs DirectX 9, etc.), installations tend to be much smoother in Windows and you should almost never have to resort to the command line. Command line is great for /. readers, less good for the unwashed masses.
What I want to see is something like the Click-N-Run Warehouse but free, encompassing a much wider range of software, and taking care of all those nasty dependancies without me worrying about it in one click and no command line. In my opinion, Linux is worse than Windows in this respect. It's not going to be easy but Linux can (and should) be better than Windows for software installations. If Lindows can do it, why can't thousands of open source programmers around the world?
There are a huge number of reasons, and you demonstrate the primary one: "They have some sort of menu from which they can launch applications, they have file managers, they can print files."
At first glance you might not understand, but you demonstrate the problem precisly. Linux desktop enviroments are designed at the high level; there is a file manager, a way to launch applications and a way to print files. No one, and I really do mean no one, in the Linux desktop community has ever bothered to study and correct the lower level design. How should the user move a file from one place to another? How should applications be presented to the user? What does a user normally do with their file manager? What features does a user require from something like Kicker or the Gnome Panel? Even lower level; do we even need something like Kicker or the Gnome Panel?
99% of the Linux desktop enviroments come from a process of "Monkey see, Monkey do". Someone else is doing it, therefor it must be a good idea, therefore we'll do it too. There is very little apparent thought in the process, and I am certain there is very little understanding of the design and usability principles which underlie the original design which is being copied. Nor do those doing the copying apparently take the time to decide how the new feature will fit into their design. Come to that, most enviroments for Linux do not even have a guiding design!
As long as the various Linux desktops continue to imitate without understanding then they will offer a poor user experience. I'm not even claiming that they shouldn't imitate; it's just that they need to put much more thought into why they're imitating something and be able to understand the beneficial or detremental effects of their imitation.
Windows does a pretty good job of making general computer tasks easy to do. What drive me nuts, as a developer, is that they take this same dumbing down restrictive philosphy to their development tools. I really find that much of dev studio just gets in my way and slows me down. Linux has just the opposite problem. Everything is designed with the developer as the target user (not intentionally mind you). This results in all of us loving Linux, because it works naturally for "us". The problem with this is that the community that makes Linux is too close too the product to see/admit that it alienates end users.
What we need is for the CLI to not be essential, and for a GUI to be available to the end-user for most if not all tasks.
So, um, kind of like it is now, only more so, then.
Can you identify more than two or three common tasks for which there is not a GUI available in Linux? The only thing I can think of that I have to drop to a CLI for these days is compiling software myself, which is something that Joe User simply won't be doing.
"Hardware support? ... Easy addition of new hardware? ... Easy addition of new software?"
If we exclude home users for a moment, and think about linux on the desktop in businesses, all of these problems go away - normal users do not (or should not) do any of these things; their IT support department should.
The problem is that "RTFM" comes across as very elitist, regardless of what is intended. Instead, they should respond with "I think you will find the answer you're looking for here: " and provide a link to the FM, so people can R it. A lot of neophites don't even know where the manual is, so they can't read it (and we both know that manuals for Linux software isn't always easy to find).
"If English was good enough for Jesus, it's good enough for everyone else."
constant tweaking and installation of new hardware and software is not using a computer productively. using a computer productively means that the computer is operational, in the sense that it doesn't actually -need- anything further to be done to it in order to function as intended. it just works.
It is classic Linux-advocate style to redefine the user's problem to fit Linux's needs rather than the other way around. Some people like to get a new digital camera once a year. Some people like to install a new game once a month. Some people like to buy the latest and greatest MP3 players, video cards, wi-fi devices, photo printers, hand-held devices and all of it comes with software.
People want the capabilities of their computer to expand as the industry expands and new things are invented. It's a very closed mindset that says the "computer has a use and once it is set up it is static." My uses for the computer change every day (especially as a I am a progrmmer). Why should my less technical sister be restricted from a similarly expansive view of computing? If I call her up and tell her BitTorrent or iPod is the shit, it should be easy for her to install BitTorrent or an Ipod.
I'm not saying that Linux is intrinsically worse that windows at supporting dynamically changing systems: but for the average user today it is worse because of the driver and software support. That isn't Linux's "fault" but it is Linux's "problem". Not the user's problem. If you make it the user's problem they will stick with Windows and (frighteningly!) think of it as a more free and open system than the vision of Linux you are pushing.
When someone says RTFM they are saying, I'm not going to type the answer to you because it is already written down. Go read it.
Some people are saying that, but other people simply use it as a cop-out. Recent case in point: I'm having iceauth problems, so I've been digging through old newsgroup postings. Somebody replied to an iceauth problem post (you can probably find it on Google Groups; that's where I stumbled across it) by saying "RTFM." Look up the FM for iceauth, and tell me why this is a valid response.
The people who are saying "the answer is already written down; go read it" should give a link to the answer, and quote some of the relevant text. In other words, point the person in the right direction. Don't send them to a multi-page document without any direction.
There's a way to share specialized knowledge properly. If we're having a (verbal) conversation and you asked me what a word meant, I shouldn't reply "it's in the dictionary." Duh. Maybe you don't know how to spell the word. Maybe it's a specialized word that shows up in my medical dictionary, but not in my regular English dictionary. Maybe it's a slang term (hacker parlance, for example). Maybe you don't have easy access to the dictionary right now (you're not near one), maybe the issue is time-sensitive, and you don't have the option of getting yourself to a dictionary, etc., etc.
Assess the situation properly. Give the person enough info that they can make do in the short term, but know where to go to round things out properly. Using my "unknown word" example, tell the person what sort of word they're looking for ("it's a medical term") what the word means generally ("it has to do with XYZ") and tell them where they can get the additional info they need ("most anatomy books talk about it in detail" or "it might only be in medical dictionaries; it's spelled X-Y-Z"). Don't tell them the Latin origins of the word, what date it appeared in common usage, etc. They get that information when they go look it up.
MJC
"Linux has an infinite plethora of ways to solve peoples' problems, but peoples' problems -belong- to them, not the other way around."
That's a perfectly reasonable philosphical position to take, but if the Linux community follows it, Linux will never be mainstream on the desktop.
I disagree that the bulk of the linux world ignores usability. The first step in most open source projects is to get something out there that works. The second step is to get in as many useful features as possible. The third step is sometimes usability. Many many projects work on usability once they have something useful in the wild. Even the linux kernel has a nice menu system with good documentation to assist in configuring. Mozilla, OpenOffice, GNOME, and KDE have been focusing primarily on usability lately. There are weak areas, especially XFree86 configuration, but as projects become more mature there is often a focus on usability.
Developers: We can use your help.