Mark Shuttleworth writes: “We are a somewhat chaotic crowd, the software libre army. Thousands of projects (hundreds of thousands, if you consider Sourceforge as a reference point). Hundreds of thousands of contributing developers from virtually every country and timezone. We are a very loosely coupled bunch. But sometimes I wish it were easier to keep track of changes and have a slightly clearer view of progress across that whole galaxy.” Eugenia agrees.
To assist “regular users”, a website which has been advertised and stickers on goods in stores should also help with getting more vendors to support open source.
When i purchase hardware, i usually look for stuff that i know works with Linux. So these hardware vendors which are “Linux friendly” get my business. Even if you are going to use the hardware on a Microsoft box, i’d urge you to purchase goods which you can also use on a Linux box as well. I’m sure over time this would start to level out the playing field, and send a message to hardware vendors.
The best thing we can do (and i agree with Mark’s blog) is work together a lot closer than before. This will make the collection of projects into a better OS. For the power and regular users out there.
It’s still herding cats though. Some will get uppity because they don’t want to do it ‘the Ubuntu way’. Others will throw a fit because they think Mark is taking away the freedom from linux. Others prefer KDE.
Whilst I think this is wise and will help, you can bet your bottom dollar another two distros will come out of it; because people would rather fork, than swallow their pride.
Others prefer KDE.
So does Mark Shuttleworth, just so you know.
Kubuntu is as Ubuntu as Ubuntu itself. I run Kubuntu, and to me it is no different than running Ubuntu. They’re two flavors of the same distro.
I agree with the rest of your points.
BTW, there’s a typo in the title.
“KDE… So does Mark Shuttleworth”
He never said he preferred it, he stated that he was using it on his desktop. Such a move could be a simple gesture of showing that he is dedicated to making sure that Kubuntu is a quality distro that is suitable for everyday use. “Eating your own dog food” so to speak.
For reference see: http://www.kubuntu.org/announcements/kde-commitment.php
He may prefer it, then again he may not. Whether or not he does, lets not put words in Mr. Shuttleworth’s mouth. I’m sure he would prefer it that way.
Edited 2006-12-28 20:06
Indeed, he has stated he uses Ubuntu (therefore GNOME) on his primary system – which happens to be a laptop.
My bad, I thought he had switched to Kubuntu for his laptop.
In any case, to me Kubuntu *is* Ubuntu – if only because you can install Ubuntu and then install kubuntu-desktop on top of it. As I said, they are different flavors of the same distro.
it’s hard to take Shuttleworth seriously when they can’t even agree on a standard desktop. What will it be, KDE or Gnome? QT or GTK+? Pick the best and stick with it so that developers will know what to code for.
? What do you mean they cant agree on a standard desktop?
Ubuntu is Gnome. So where is the confusion?
Kubuntu, Xubuntu etc are sub distros for people who like those other desktops. But Ubuntu proper is Gnome.
Seems simple to me.
The problem at the moment is not the desktop but the fragmentation of services that underly the desktop. I think there has been some momentum in rectifying this issue like with the likes of Dbus and Compiz but there are still some areas that are weak like in Audio where Esd is used by Gnome and the Arts is used by KDE or in file layouts between distros.
It would be nice to have more uniformity under the hood which is then utilised by what ever flavour of desktop one wants to use. My ideal would, in the Audio side, be the utilisation of jackd as the audio server in Linux with some cleaning up and standardisation on it.
With some underlying uniformity then software developers would be better able to target Linux as a whole when writing applications.
Also the Portland project is working on issues such as these.
I disagree. People should use the desktop they prefer. Choice is good, especially when apps made for one desktop work on the other. In any case, it seems to me developers *already* know what to code for. Are there any programs out there that are unreleased because the devs don’t know which toolkit to pick?
I really don’t understand this quaint notion that everybody should use the same thing…honestly, this is a false problem.
it’s a false problem unless linux seriously wants to go anywhere on the desktop.
I disagree with you. I don’t think it has any incidence on Linux desktop adoption whatsoever, and you haven’t provided one shred of evidence that it does.
As I said, I don’t know of any developer who really complains about this, or of any user being affected by this in any significant way. All we have is more naysaying from you.
So I challenge you to give me concrete real-world examples showing that the existence of multiple Desktop Environments is detrimental to Linux, because so far you have not made that case by any measure.
OK, here’s an example. Everyone knows that linux needs an alternative to MS Office. But instead of everyone working on the same project, we just *have* to have a whole team on koffice. Why this duplication of effort? because KDE *needs* special apps of its own. Now if KDE was the standard desktop, there wouldn’t be a need for a special KDE office. OpenOffice would be koffice and kde would put its resources to better use. Instead, we have 2 subpar office alternatives instead of 1 that would make people dump MS Office 2003.
You forget that a lot of the critical code is shared between KOffice, OpenOffice.org and GNOME Office. Not to mention other projects, something like KOffice uses a lot of the same, well tested code as the underlying KDE platform (same with GNOME Office – yay Abiword!).
“OK, here’s an example. Everyone knows that linux needs an alternative to MS Office. But instead of everyone working on the same project, we just *have* to have a whole team on koffice. Why this duplication of effort? because KDE *needs* special apps of its own. Now if KDE was the standard desktop, there wouldn’t be a need for a special KDE office. OpenOffice would be koffice and kde would put its resources to better use. Instead, we have 2 subpar office alternatives instead of 1 that would make people dump MS Office 2003.”
Personally I agree with this statement. The duplication of effort unfortunately is due to the fragmentaion of the different distros, viewed as Linux, although Linux proper is only the kernel. KDE just insists on having a ‘Keverything’. I use KDE myself and it has some great features, but to his end I agree. If they concentrated on OO.o instead, maybe even bigger and better things would come.
That is not *at all* what you were arguing before. You were saying that the mutlitude of DEs *prevented* people from making software.
There’s also a logical fallacy in your argument. There’s absolutely *no proof* that someone working on Koffice right now would world on OpenOffice if KDE didn’t exist (or if it was the main desktop). Absolutely none. That’s utter bullcrap.
Why are people working on alternative Office suite for Windows? They should all just give up because there’s already a MS Office, right? They should stop working on OpenOffice, or on WordPerfect, or on Gobe Productive Office…they should all work on MS Office instead, right?
What do you have against competition? It’s the founding element of a market economy, and yet you seem to consider it as a bad thing.
Well I disagree with you. I think competition is good. I think it pushes people to improve their product. I also think that Koffice and OpenOffice are very good products, and getting better all the time.
And, most of all, *none* of this has anything to do with low dektop market share numbers for Linux.
So does Mark Shuttleworth, just so you know.
Oh, come on! This is pure flaimbait, and you know it. What do you know about his personal preferences, other than what is “said”?
Oh, come on! This is pure flaimbait, and you know it.
Why would that be “pure flamebait”? Is it a crime to prefer one desktop environment over the other?
I have already been corrected that Shuttleworth uses Gnome on his laptop (his main PC) and KDE on his Desktop (his secondary one). I myself use mostly KDE but I do think that Gnome is a fine desktop as well.
What do you know about his personal preferences, other than what is “said”?
I distinctively remember him saying that Kubuntu needed a push and that he had been using KDE recently. I didn’t realize it was on his secondary PC, and I’ve already been corrected about that. Please pay attention.
It wasn’t my intention to start a flameware, however I’m not sure the same can be said about you. I suggest you take two minutes to take a deep breath and chill.
Actually, it’s not impossible. The trick to herding cats is to get them to herd themselves (i.e. place a nice juicy mouse at one end and lead them forward).
We’ve seen this happen countless times. Look at RSS feeds on all the planets. There are tonnes of blogs out there written on hundreds of blogging applications, but it’s possible to create planet sites like http://planet.gnome.org or http://planet.ubuntu.com or http://gstreamer.freedesktop.org/planet because all these blogging applications use a few RSS standards. The “cats” (a.k.a. developers) herded themselves to these standards because most “cats” belong to more than one community (e.g. GNOME, Ubuntu, Gstreamer, FSF, etc) and want to have their blogs available to all these communities with the minimum of hassle.
This isn’t much different from the Bugtracker situation. If all bug trackers supported one of several Bug Tracking schema (e.g. http://xml.coverpages.org/ni2001-12-20-e.html ) then it would be possible for bug trackers to query each other and point to each other or depend on each other.
launchpad == software libre army tool ????
I think he doesn’t want to restart the “open launchpad” flamewar, he just wants to focus in the idea. Having the same bug reported in every distro’s bugzilla does not work. It’d be nice to add a “decentralization” feature to bugzilla.
We need to solve two problems, having single sign-on for the bug trackers, luckily OpenID does that we just need to hook it up to the user handling and login. Then we need an easy way for downstream to clone a bug to upstream and again we have the solution, xmlrpc in bugzilla.
Bug-buddy already makes it trivially easy to file a bug (Some might say to easy: http://uwstopia.nl/blog/2006/12/bug-reports). With the recent scripting support we can get every bit of information we need, hell since we know the exact packages thanks to our friendly neighborhood distros packagers we can even fill out backtraces so there’s no need to install debug symbols on the users machine.
So I would say if we continue to bet on the freeness of the entire stack, we will be just fine.
People have already brought a bug up about OpenID support in launchpad ( https://bugs.launchpad.net/products/launchpad/+bug/1169 ), though it’s been a year with no change!
As for notifying downstream, this is what Malone does, it co-ordinates between launchpad and external projects.
I don’t especially care about Launchpad untill it’s freed it will never get adopted outside of Ubuntu and related projects. Would you let a blackbox controlled by other people be your bug tracker.. I can’t say I know many people who would be willing to do that, yet that is what Mark is asking the world to on nothing but his word. I doubt that will work to be honest, at least not anytime soon.
Thus the option we have is to integrate free solutions.
OpenID integration was debated on the Fedora Summit and Red Hat already has support for xmlrpc in their Bugzilla (it’s just a matter of getting upstream to agree to it’s integration and everyone will have it). I could see this working a whole lot sooner than the long promised but never acted upon liberation of Launchpad.
It says “Shuttwworth” instead of “Shuttleworth” in the headline. Could someone fix it?
Aaron Seigo has blogged about a related topic some time ago:
http://aseigo.blogspot.com/2006/12/patchserver.html
OK, the subject is somewhat of the mark, actually it is good observation, good example, ugly finish.
Mr. Shuttleworth certainly has made the right (and obvious) observation. Development is mostly disconnected right now and taking the bugtrackers as example is a good choice to start to tackle this problem. Making bug- and more importantly patch-tracking easier over the majority of free and open software is a worthwhile goal.
This could have been a good opportunity for Mr. Shuttleworth to say, see, we use a proprietary bugtracker for Ubuntu (and siblings) and we have no intention of opening this piece of software, but we would like to sponsor an effort to create a free and open standard to offer a protocol for bugtracker communication and migration. Unfortunatly Mr. Shuttleworth dropped the ball here and just pimped his proprietary solution. That’s bad. And ugly.
Maybe a group or individual with less vested interest in one piece of (proprietary) software will pick up the ball and go where no-one, incl. Mr. Shuttleworth, has gone before?
Something as complicated as what Launchpad sounds like propably isn’t nessisary and probably wont gain momentum. I think “Trac” with the proper modifications is more likely catch on for this role. It already understands “bug links” and it already understands Inter-Wiki, so some kind of an “Inter-Wiki Bug Link” would be the next logical step. Given a choice people will generally choose the technology that disturbs their existing work flow the least amount for the perceived benefit.