A modest proposal re. Unity
October 26, 2010 10:13 am community, freesoftware, gnomeHaving slept on it since writing my initial reactions yesterday I now have a proposal for Canonical & GNOME, which I hope the people concerned will consider.
Yesterday, I said “the best possible outcome I can see is that one of the two projects will become an obvious choice within a year or so”. So my proposal is this: let’s have a bake-off, Unity vs GNOME Shell, under the big tent of the GNOME project.
What needs to happen? Unity would have to agree to sync to the GNOME release schedule. Canonical will need to drop their copyright assignment requirement for Unity, and should ideally commit to using some of GNOME’s infrastructure. How much will need to be discussed. I’m sure that the Unity developers will want to continue to use Launchpad for bug tracking and bzr for source control, but perhaps the development mailing list could move to gnome.org, and the Unity website could be gnome.org/projects/unity or unity.gnome.org instead of unity.ubuntu.com?
GNOME will have to accept Launchpad as a platform for the development of GNOME software – there are potential integration issues, it is a headache using Launchpad & Bugzilla, co-ordinating Rosetta & upstream translation teams, and so on. But right now there is a general feeling that gnome.org is for “official” GNOME software, and Launchpad is for Ubuntu. We need to change that perception if we hope to be inclusive of Canonical and the greater Ubuntu developer community in GNOME. In fact, broadening the definition of what we call GNOME software was a key plank in the release team platform for 3.0 – resolving this question (and the equivalent question for projects hosted on Google Code and Sourceforge) will go a long way to growing the big tent. The GNOME project should also work to make it easy to switch from one shell to the other.
Developers who feel drawn to one philosophy or the other should work to make sure that their vision is the best it can be by September 2011. And at that point, presumably at the Desktop Summit in Berlin, GNOME, as a project, should choose one of the two, and put our full weight behind it.
This is potentially naïve on my part. Over the years, we have allowed a lot of animosity to build up between Canonical and Red Hat, among others. As a community, we’ve stood passively by while this has happened. Some will point to efforts to engage which were rebuffed. Others will point to a lack of real commitment to engage.
In situations like this, no-one is 100% right, no-one is 100% wrong. All we can do is look at the current situation, and ask ourselves: how do we get to where we want to go, from where we are? We have two choices – we can, like the Mayoman asked for directions to Galway by tourists, respond “If I was going to Galway I wouldn’t start from here at all”, or we can roll up our sleeves and try to make things a little better.
So – how about it? Is this a non-starter, or is it worth starting some conversations about it?
October 26th, 2010 at 10:44 am
I agree, I like the idea of having a tighter collaboration between GNOME and Canonical, and indeed, GNOME and it’s community should be open enough to receive the news of a new shell built for GNOME with a bit more enthusiasm, and not with disappointment. And it’d be nice to have this as you have said, to test both Gnome Shell and Unity for a time, and after that decide which one is better, and after the decision has been taken, drop the other. But I think the human pride is a major issue here, although it would be an unprecedented example of collaboration in the history of FOSS.
October 26th, 2010 at 10:56 am
[…] This post was mentioned on Twitter by Tara C., Zuissi. Zuissi said: Planet Gnome: Dave Neary: A modest proposal re. Unity: Having slept on it since writing my initial reactions yeste… http://bit.ly/bRiCxu […]
October 26th, 2010 at 10:58 am
How about, like KDE, you ack that there is no one true vision of the desktop, everyone wants something different, and have a really flexible UI that meets as many people’s needs as possible.
October 26th, 2010 at 11:05 am
Good idea but I doubt this will ever happen.
October 26th, 2010 at 11:08 am
A decision in September 2011, past the release freeze for any distro, won’t push GNOME 3.0 in the hands of the users only in spring 2012?
October 26th, 2010 at 11:18 am
You should look at KDE. They have vision, you do not have. Sorry there is no win-win way to solve this situation. There will be battle, the best will win. In meantime more people will use KDE, XFCE, LXDE and other DE.
October 26th, 2010 at 11:35 am
That’s brilliant !
I hope your voice will be heard. This is a very good compromise IMHO, and it would benefit to everybody (as a GNOME developer, I personally like Launchpad a lot).
Once you’ve settled this problem, could you make a proposition for the governement in Belgium ? We need someone like you
October 26th, 2010 at 11:40 am
I agree with this proposal. Seems fair.
I think, though, that the third option that people would have at the Desktop Summit would be to pick-and-choose the best aspects of each project.
For example, I love what Gnome-Shell is doing with app notifications. It’s innovative and useful.
But on the other hand Unity seems to be making good use of Zeitgeist, and the indicator applets are quite good for status icons (non-notification-related).
I would love to see the best of both worlds coming together.
October 26th, 2010 at 11:43 am
(PS: I still think both projects get one thing fundamentally wrong, though. The use of an overlay that covers the entire screen just to launch an app or switch windows screams bad usability to me.)
October 26th, 2010 at 11:55 am
Canonical/Ubuntu people should show more interest in collaborating with GNOME. And we are willing to receive their contributions in the open.
Maybe launchpad.gnome.org is the answer to this. I think they can feel more like home. Gnome.org is the main place where GNOME stuff is discussed. But, of course you can organize sort of hackests with presence of GNOME hackers at UDS if you want to discuss new features.
If you want to contribute to GNOME, please, you can be part of our community without resign of being part of Ubuntu.
And of course, we need to be sure that other supportive communities like Fedora, Opensuse and companies etc are confortable with the our collaboration infraestructure as well.
October 26th, 2010 at 11:57 am
Any effort to get people to engage rather than squabble has got to be good. Let people use the tools they prefer, encourage people to contribute, and a little FRIENDLY competition should get the free desktop shining in no time!
October 26th, 2010 at 12:06 pm
@foo: Providing one coherent and intuitive interface has its benefits and I have always supported GNOME in that idea. However, I see only positive sides when there are suddenly multiple different providers of those ideas, all based on GNOME! Realize it — Unity is GNOME diversity and will make us better and smarter if you use it or not.
October 26th, 2010 at 12:11 pm
Why does any of the stuff in the “What needs to happen” paragraph actually need to happen?
Why isn’t GNOME happy to consider Unity as an interesting experiment happening outside which, if it proves useful, may later be brought into gnome.org?
IIRC Nautilus for example lived (mailing lists, bugzilla) outside gnome.org, at least until the demise of Eazel.
I find it most peculiar. Ubuntu seems to cop a lot of flack for not producing enough of their own code and now there’s a whole project they are going to work on there seems to be an inordinate amount of complaining as to exactly how they go about it.
Why not just cut them some slack and see where the project goes?
As an outsider looking in the amount of angst I’m seeing splurged about almost vindicates Canonical’s choice to run it as their own project. If this “stop energy” is the norm why would you want your project anywhere near it?
October 26th, 2010 at 12:24 pm
@nicu: Thus my proposal to have a decision before then – the Desktop Summit is in the first week of August, just in time for GNOME 3.2.
@paul: GNOME has a long and considered stance on copyright assignment, born from painful experience. Ubuntu developers use Launchpad, and will continue to do so, but that makes collaboration with non-Launchpad-hosted projects more difficult. I’m proposing building bridges, and reaching resolution on an issue which could cause a lot of conflict in GNOME if left to fester. Let me turn the question on its head: what is the long-term benefit to the free software desktop in having Unity continue as a Canonical-owned “experiment”?
October 26th, 2010 at 1:09 pm
Well it all depends on Canonical’s plans I guess. Maybe they do want differentiation and there is not much we can do about it. Or maybe they dislike GNOME Shell and then are doing their own little things rather than influence GNOME Shell. Maybe the first step is to find out the real reason of such a decision?
October 26th, 2010 at 2:20 pm
I wonder if XFCE is just going to end up passing the gnome project as a whole. It’s stable, it’s small, it’s gtk based, it has the compositing bling, it looks wonderful and “gnome-like”, it integrates perfectly with Red Hat/fedora, and it seems to me to have more functionality than gnome+compiz.
After many years of using gnome I made the switch a few months ago and haven’t looked back. Obviously I still follow the gnome news since it’s an important project, but I can’t help but think XFCE is passing it by on its merits.
October 26th, 2010 at 2:23 pm
Sorry for picking on technicalities, but I hope your suggestion will be skipped by the gnome developers.
Launchpad is one of the most horrible bug tracking system I ever worked with. In fact it made me stop contributing bug reports to ubuntu and Exaile.
The difficulty of seeing activity and commits, maybe explains why there are so many unclosed bugs in Ubuntu.
Bugzilla on the other hand is also not my favourite, but at least it’s decent.
Another thing I have to say is, despite Ubuntu’s great contributions so far, I tend not to recommend users on it since their 8.04 version. They break stuff, and they seem to sometimes ignore users in a very blunt way. I feel like many others, that they are trying enclose Linux in a “Golden Cage”.
I think, that the Linux community and GNOME can survive with out Ubuntu’s sweet trap.
October 26th, 2010 at 2:43 pm
“As a community, we’ve stood passively by while this has happened.”
Yeah, we do that a lot. And then we wonder why we find ourselves in these situations.
Nice ideas, but yes, I think you’re being naive. The clan of Canonical haters are already on this, and I don’t think they’re interested in “getting along”. They want things their way.
As usual.
October 26th, 2010 at 2:45 pm
Hey Dave,
Interesting and thoughtful post.
Of course, I don’t make any decisions around the strategic planning of Unity (no-one should trust me with the keys
), but I wanted to offer a few notes from my own perspective.
Personally, I think GNOME needs to get over the “projects need to use GNOME infrastructure” requirement. I love the concept of the GNOME tent being inclusive of other development forges, because the key focus here should be awesome Free Software, irrespective of where it is produced.
Developers should have the freedom to write Free Software whereever and however they choose, and if their software is high quality, it should be part of the GNOME community and ecosystem. While GNOME provides some fantastic infrastructure facilities for developers, I don’t feel it is appropriate to demand that projects use it. Speaking personally, I have written a bunch of small GNOME tools (hackishly put together due to my feeble development capabilities), and I like to roll with Launchpad and Bazaar – my development forge of choice is Free Software, the tools are Free Software, and all the staple parts of my projects (code, bugs, blueprints, translations etc) are all open and apt for participation.
My recommendation would be that GNOME defines a policy for software that is part of GNOME but uses non-GNOME infrastructure. It could include requirements such as:
* the forge must be Free Software and use Free Software tools (e.g. version control).
* a sys-admin point of contact must be provided.
* the forge must be accessible and localized.
* . . .
You could have a set of “approved” forges (e.g. Launchpad, SourceForge, GitHub) that people can participate in.
My worry is that if GNOME maintains this infrastructure requirement, it is going to make GNOME irrelevant as I believe developers will prioritize using their tools of choice rather than the benefits of being part of the GNOME project. This seems like a totally solvable problem that benefits both GNOME and developers having the freedom to produce their software in a place that is preferential to them.
Sorry for the long post.
Jono
October 26th, 2010 at 3:23 pm
@Jono Bacon: you seem to have glossed over the issue of copyright assignment for Unity. I don’t know the full details on that, but would be curious to hear the other side.
October 26th, 2010 at 3:28 pm
@jono: My major concern is to ensure that access is open, and that communication channels are open between different GNOME modules. It’s easy to reassign a bug from Rhythmbox to GTK+ or GStreamer, not so easy to pulseaudio, and close to impossible to the kernel (I don’t know if that’s ever come up, it’s just an example). To me the conversations are what makes the project. Conversations can happen on Launchpad, and they can happen in GNOME, but conversations between the two don’t happen much. If we can figure out how to increate the size of the communication channel, and have a team working on software hosted on Launchpad work well with an upstream dependency, then I’m all for it. I haven’t seen it happen until now (thus my suggestion that maybe using gitbzr & a mailing list could work). But it should go both ways – we’ll need to make it easier for changes in upstream translations to be tracked in Rosetta, upstream bug comments & changes to get synced with Launchpad (& vice versa), etc. The whole point is to enable conversation.
How would you feel about dropping copyright assignment & engaging GNOME with Unity?
Dave.
October 26th, 2010 at 3:36 pm
As much as people are giving unity and Cannonical grief (i’m a little indifferent, as a KDE guy), I don’t think this split is really awful. The thing that really bugs me is the copyright stuff, especially since the opinion seems to be side-stepped at every confrontation (even Jono avoided it above ;). )
By the way, last I tried (about 3 days ago on a 2.32 fedora machine), gnome-shell is not even close to being in a shippable state. I don’t think people can really point their fingers at this stage.
October 26th, 2010 at 4:34 pm
Hi Dave,
re. Copyright Assignment – that is not something I make a decision about.
As for engaging GNOME with Unity, Unity is Free and Open Source software, and I think collaboration between the two projects would be fantastic. As with GNOME, contributions are welcome, and I know the Unity team have always been keen for technologies such as the app indicators and notify OSD to be used by GNOME. The challenge is this infrastructure issue – if you are asking could we bring Unity over to GNOME infrastructure, I think that is unlikely.
Totally agree on the points surrounding conversation – we have some technical facilities such as bug linking which can help, and I think we should ensure that the communication channels (both conversational and issue focused, such as bugs) should be open. I think what we need right now is an acknowledgement that both GNOME Shell, Unity and other shells are all welcome in the GNOME community. We have an awesome platform, we just view it, start apps and access system content in different ways.
Jono
October 26th, 2010 at 5:40 pm
“How about…you ack that there is no one true vision of the desktop, everyone wants something different, and have a really flexible UI that meets as many people’s needs as possible.”
There is no such utopia, as any UX designer will tell you. The more people you try to please, the worse your UI becomes.
October 26th, 2010 at 5:56 pm
@Jono
Sure you don’t make a decision about copyright assignment.. but you can have a personal opinion about it right? Right? Or does are Canonical employees equiped with some sort of shock collar that gets activated if you say something contrary to corporate policy in public?
The copyright assignment issue can not be ignored. Stop trying to ignore it.
@Dave,
Canonical has shown that they can in fact work outside of Launchpad to do upstream development. They do it for their kernel work. We just saw Canonical finish the long process of getting AppArmor into the kernel. We see them interacting with git repositories as part of the Ubuntu-spinoff Arm Linaro.
Speaking of Linaro, we even see Canonical agreeing to assign copyrights over to the independent mult-vendor supported non-profit Linaro org (and the FSF) as part of standard Linaro practise. Would the GNOME Foundation be willing to take a copyright assignment from Canonical at a entity level like Linaro does?
Everything you say GNOME needs to compromise over to appease Canonical, Canonical is already willing to do in other contexts when working with other upstream projects.
Canonical’s own reluctance to conform to GNOME procedures is clearly strategic.
I’m not going to tell you whether GNOME should change policies or not. Though I will say that there is clearly a long history of corporate entities aiming at the consumer market doing the same sort of differentiation on top of gnome that Canonical is doing right now. Canonical feels like hot button, but are they really doing anything strategically different that Nokia didn’t do with the creation of Hildon several years ago in support of Maemo based devices?
There is something systemic either in the consumer market dynamics or in GNOME policy..or both…that encourages vendors who are taking a run at the consumer market to differentiate. Personally I think its more the market than GNOME policies, but I can’t easily quantify that into a sober data based analysis. The same tendency seems to impact kernel development too..but the linux kernel project hasn’t changed policy based on this vendor tendency to need to differentiate.
With that in mind. I’m also telling you that Canonical will conform to similar policies that GNOME already has in the context of other projects has when Canonical feels its in their best interest.
-jef
October 26th, 2010 at 6:27 pm
Jef:
“Sure you don’t make a decision about copyright assignment.. but you can have a personal opinion about it right? Right? Or does are Canonical employees equiped with some sort of shock collar that gets activated if you say something contrary to corporate policy in public?
The copyright assignment issue can not be ignored. Stop trying to ignore it.”
Jef, that is unfair. Who said I was ignoring it? I just said that I don’t make the decision surrounding whether we have it or not, and Dave was asking me to remove it – I am just pointing out that I don’t make that decision. It is also unfair that you assume that Canonical has a policy to silence it’s staff of their views – I wish you sometimes remember that while we may not always see eye to eye on software policy, we are all actually good people with good intentions and not out to screw Free Software.
My view is simple: I think copyright assignment policies serve a valuable function in being able to easily re-license code, and solving the problem of people-inaccessible-or-dying-and-therefore-unable-to-respond-to-emails-to-relicense-code problems. I think they are valuable tools under the premise that they are used to protect the re-licensing of Free Software and not block appropriate re-licensing based on the unavailability or dead contributor issues.
Jono
October 26th, 2010 at 6:36 pm
@Jono
And you believe that Canonical’s current copyright assignment is a _fair_ and _equitable_ way to solve the issues associated with good faith re-licensing needs? It’s not the FSF’s agreement. You’ve read Aaron Seigo’s critique of Canonical current policy? Do you personally thing safe guards against proprietary re-licensing should be added to Canonical’s current policy?
-jef
October 26th, 2010 at 6:57 pm
Jef,
Personally speaking, and not speaking for Canonical, I think that it would be better if the contributor agreement removed the ability to re-license under a non-OSI license.
October 26th, 2010 at 7:58 pm
I agree with your previous post wrt the lack of collaboration.
But actually I think Unity is a better project. So from this perspective, that is, imagine you’re a Canonical employee, which thinks your project is better designed than GnomeShell, how would you feel? You would feel like “who has to collaborate with who?, Gnome could have collaborated with us developing this because it’s better than GnomeShell”.
But I understand a bit also Gnome’s policies.
However, let me introduce a 3rd point of view: a technical party (me, a developer) which will not contribute to it, but which prefers Unity. For those, I think Unity should use Gnome infrastructure (bugzilla, git) and be a first upstream citizen like GnomeShell. Wasn’t Unity accepted by Gnome upstream in the first place? Let’s reconsider that answer that was given to them, I’m sure they will accept it, and everyone will be happy.
October 26th, 2010 at 9:04 pm
Jono: “Jef, that is unfair. Who said I was ignoring it? I just said that I don’t make the decision surrounding whether we have it or not, and Dave was asking me to remove it – I am just pointing out that I don’t make that decision.”
That’s not actually true. Quotation follows:
bolsh: “How would you feel about dropping copyright assignment & engaging GNOME with Unity?”
He didn’t ask you to remove the copyright assignment; he asked how you ‘would feel’ about doing so. i.e., he asked for your personal opinion. Which Jef then excoriated you for not giving. =) I know Jef’s a bit…enthusiastic sometimes, but he wasn’t actually being unfair.
Anyway, you gave it now, so never mind. Your personal position on copyright assignment seems very reasonable to me, and is one of the main reasons for the assignment to non-profits which I think both sides agree is generally beneficial. In that case, what do you think of bolsh’s proposal of assigning copyright for Unity to the GNOME Foundation, presumably under similar terms? It seems reasonable to me.
I am, of course, like you, speaking personally and not on behalf of my employer (or GNOME, for that matter). I have no official standing with RH or GNOME in regards to this whole thing, I’m just an interested bystander.
I do quite like Dave’s idea, I think the other commenter who pointed out the timeframe issues has a point, though. Both Canonical and GNOME are planning to push their shells around six months from now, which doesn’t leave a lot of time for Dave’s process to run without delaying both sides a lot from their schedules.
Maybe the best we could do is the pragmatic option Paul suggested. I might look at building Unity packages for Fedora just to throw the cat among the pigeons. It seems fair, if Ubuntu’s going to have GNOME Shell packages…
October 26th, 2010 at 9:27 pm
Adam,
The only potential problem you’ll come to with regard to Unity packages in Fedora..I fear…is the extent of vendor specific patching to gnome components to make Unity work as designed. If there are vendor patches that must be applied to gnome elements to make Unity work as expected..that could be a non-starter.
-jef
October 26th, 2010 at 9:52 pm
there’s ways to cope, but yes, it’d get ugly.
October 26th, 2010 at 9:53 pm
…so, uh, Jon, if you could prevail upon the Unity folks to avoid that (or get any necessary underlying changes upstreamed) as much as possible, it’d be aces =)
October 26th, 2010 at 9:57 pm
@Dave
I think the “long term benefit” is in seeing whether the experiment works.
If it does and Unity ends up being a must have component then bring it under the GNOME umbrella. Canonical would either have to relax their control or a fork would result.
I understand how some individuals (and corporates) may be put off contributing by Canonical’s Contributer Agreement and I think that is reasonable decision for them to make. I don’t really see how that is any different to any other reason why someone may choose not to work on a project. At the end of the day it is the license granted that allows GNOME to operate freely irrespective of who owns the copyright.
You say that “GNOME has a long and considered stance on copyright assignment”. I have read it, it says:
What is different here?
It seems to me that similar “experiments” to this one have led to two of the most significant user-facing parts of the GNOME desktop.
October 26th, 2010 at 10:05 pm
Why not forking Unity into a project called, for example, Diversity (as opposed to Unity
and host it in Gnome’s git and bugzilla? Everyone could contribute to it without copyright assignment, and you could still bring Unity’s fixes into it.
This way you solve 3 problems by one move: the copyright assignment, the use of Gnome infrastructure for it, and the inclusion of it as an important role of Gnome’s upstream in parallel with (GS)hell.
October 26th, 2010 at 10:16 pm
BTW, the move about creating “Diversity” would actually be very equivalent as the move about creating LibreOffice
October 26th, 2010 at 10:17 pm
Andres,
That would require a heroic effort by a group of people who are deeply invested in building a collaborative environment around Unity.
There are multiple copyright controlled codebases that make up the Unity interface…its not just the thin UI layer..but its also the utouch libraries and libappindicator and libzeitergist etc.
Realistically that sort of commitment to a forking effort is only going to be viable down the road if Unity gains significant traction. It’s way to early in the lifecycle of Unity as an interface to see a fork of that magnitude. But stranger things have happened.
-jef
October 26th, 2010 at 10:40 pm
@Andres and @Paul: I had considered a GNOME fork of Unity as a possibility. Hardly what I’d call a best-case scenario.
Actually, Aaron Seigo pointed out the most likely scenario: This adversely affects GNOME to the point where KDE becomes the de facto desktop for Linux. Ubuntu calls the Unity experiment a failure & moves to KDE 4 for 12.04 (or sets off to bring to life a Qt based desktop shell based on the GNOME design philosophy. GNOME will live on in Fedora, becomes a second class citizen in OpenSuse. This is not a good scenario for GNOME, but perhaps not a bad one for the free software desktop.
October 26th, 2010 at 10:54 pm
Dave,
Aaron’s doomday scenario is a little self-serving obviously. But also consider for a moment that Unity is no less than the 4rd differentiated interface that Canonical has fielded…maybe he’s got a point. Here’s my official count of Canonical UI “experiments” to date:
Ubuntu MID..officially dead…related to Moblinv1..also dead.
Ubuntu UNR..replaced by Unity.
Ubuntu UNR based on ELF for Arm…dead?
Mutter based Unity..now dead.
Compiz based Unity..the new hotness.
This does not include the ill-fated Mi interface Canonical helped create for HP which was never released under and Ubuntu brand.
-jef
October 26th, 2010 at 11:40 pm
interesting thought, Dave. Actually, every time I load KDE to do validation testing for a Fedora release, I wind up thinking ‘you know, this is actually really good’.
October 27th, 2010 at 12:45 am
Perhaps Gnome could adopt Apache’s rules for revolutionaries, as a means of encouraging innovation through forks while retaining overall control and momentum.
October 27th, 2010 at 2:48 am
I’d have thought the ‘risk’ in that sense would be greater if Unity were a GNOME.org project.
As a Canonical project Canonical owns its success or failure. If it were a GNOME.org project then it’s failure might make GNOME look worse.
In reality if the Unity project fails and Gnome Shell is good I imagine that would be an easier migration path for Canonical to take than a jump to KDE.
A far greater risk (as far as choosing to move to another environment) is posed if people and companies think that GNOME.org isn’t an environment in and around which they can try interesting things.
GNOME’s inception in part involved a choice of the LGPL compared to the KDEs GPL, essentially admitting the fact that even at a licencing level some parties might see value in different approaches to writing code. GNOME exemplified the benefits of Free and Open development without shoving it down anyones throat.
It would be somewhat ironic if GNOME becomes seen as the “do it our way” mob and KDE seems a less restrictive environment by comparison.
October 27th, 2010 at 6:29 am
I don’t understand why infrastructure is an issue. I understood the whole idea of Launchpad was to make it so that different projects could interact without having to be hosted all on the same system (http://blog.canonical.com/?p=192 says “Launchpad’s strength is in cross-developer and cross-project communication, including communication even with projects hosted elsewhere”). The GNOME infrastructure is basically the place where the community hangs out; splitting it so that some people hang out in the community centre and other people hang out in some local commercial offices (to strain an analogy) seems a really bad idea.
Dave’s assessment of the likely impact doesn’t seem too wide of the mark.
October 27th, 2010 at 8:26 am
Paul: “GNOME’s inception in part involved a choice of the LGPL compared to the KDEs GPL”
You probably meant to write GTK+ and Qt respectively, KDE’s libraries have always been LGPL or BSD style licenced
October 27th, 2010 at 10:10 am
Nice idea, but…
> Canonical will need to drop their copyright assignment requirement for Unity
I think we can safely say this isn’t going to happen right here!
October 27th, 2010 at 12:42 pm
@Jef: “Aaron’s doomday scenario is a little self-serving obviously”
it wasn’t actually my doomsday scenario: it was my guess at what would happen if the worst-case scenario Dave envisioned happened.
Dave suggested that if Unity and GNOME Shell both fail as a result of keeping both projects going separately, Free software would get knocked backwards by years.
I simply suggested that if that were to happen (and I carry no opinion on the likliehood of it) that Free software would not be knocked back, but that other Free software projects would pick up the slack. The Free software eggs aren’t all in one basket, and that’s not a bad thing. It allows different groups to perform radical experiments in the hopes it works out tremendously but which may ultimately fail catastrophically while posing as little risk to the Free software ecosystem as possible. Which is pretty impressive.
IOW, even if this Unity thing really is a horrific turn of events, Free software overall is at least well positioned from a risk management perspective.
October 27th, 2010 at 1:03 pm
[…] has run into its troubles, in part because building a stable desktop is not an easy task. In an update to that post, Neary recommends that the GNOME shell and Unity be run head-to-head to determine a clear […]
October 27th, 2010 at 2:46 pm
[…] A modest proposal re. Unity Having slept on it since writing my initial reactions yesterday I now have a proposal for Canonical & GNOME, which I hope the people concerned will consider. […]
October 27th, 2010 at 3:05 pm
I think GNOME community should just continue to develop and do stuff as before, why bother at all.
We all know how it should collaborate with others, gnome mailing list, bugzilla, there is your communication, all distros do it that way, all free software advances like that.
Considering that we saw none contributions from Canonical, all their projects are developed behind closed doors and then released, now we have copyright assignment also, what do you risk? Nothing, just continue with good work.
And what good are those projects for, a new fancy locked notifications, a new font, a new icon theme and bla bla… I always wonder where are those innovations that ubuntu fanboys always mentions.
October 27th, 2010 at 5:01 pm
Aaron,
The doomsday thing was a little bit of fluff.
I totally get your point. In fact I went to a talk about complex systems recently that really drove home the value of redundancy for me in highly complex system.
Complex systems run at optimal short term performance have increased catastrophic failures. Increasing the reliability of any component in the system enhances that tendency to failure. The only thing which reduces catastrophic failure risk is redudancy in the system at the component level. Doesn’t matter if its the power grid, the stock market trading, IT network infrastructure, NASA missions or salmon fisheries.
And I daresay (without doing the quantifiable analysis which determines such things) that FOSS as an ecosystem is one of these very complex systems. Redundant efforts in the short term feel like inefficient use of resources to everyone involved (and more importantly those paying the bills)…but in the long term its absolutely a hedge against catastrophic failure of the entire system. Components will fail but the system lives on.
But even then too much redundancy has diminishing value as a hedge against statistical failure assessment. There is a science to knowing how much redundancy is optimal and how much is absurd. And I don’t really think we have a working toy model of how this system works that captures the complexity of it.