I’m doing a main stage talk at [this years LugRadio Live](http://www.lugradio.org/live/2006/index.php/Schedule) on “Packaging and autopackage”.
This talk isn’t written yet, but it’ll probably quickly cover the technology (ie, a demo of it in action) and then go onto talk about the philosophies behind it. The interesting thing with autopackage isn’t the technology, it’s the distributed vs centralised debate. That debate is key to everything, and is worth spending some time on.
Towards the end I’ll probably touch on a few other Big Picture things that I’ve been pondering lately. Specifically how a distro might go about getting Firefox levels of market share.
If anybody has suggestions for what should go into it, let me know in the comments below.
June 19, 2006 at 10:58 pm |
Hey Mike – nice going. Do you need some sponsorship for some of your expenses when heading South? I am more than willing to contribute.
June 20, 2006 at 1:48 pm |
Mike, i’d be interested in reading the manuscript after it’s been on air. Just cannot focus myself on “audio input”.
June 23, 2006 at 5:05 pm |
I’m seriously thinking of going to the UK for this talk…. It would be very interesting to go and see this…
June 24, 2006 at 3:20 am |
Well, it’s a cool event but you’re putting me under a lot of pressure here ! I doubt it’ll be earth shaking or anything
June 27, 2006 at 3:18 pm |
Concerning the debat about centralized vs decentralized I fear that that can be put like this : what is a core componment and what is a software componment ? How to find the limit between them from the user´s point of view ?
For example the latest Abiword package doesn´t provide some libs anymore. I suppose that these libs aren´t especially required for the software, but considers the contrary : you can think that a developer would be fool to do that, but in his mind the developer just relies on the distro´s repository and he can think that even for some non-essential libs.
Finally, the end user would certainly not understand why his “desir” isn´t included in the package, why he would have to do an other download.
Of course, I wasn´t speaking about the Rob´s job in particular, I was just pointing out that someone could ask to you how to determine a core componment from a software componment ?
Albeit the distro´s repository is a centralized offer, the offer and creation of libs is not. Hence, this end likely with some missing libs for the user since the distros don´t centralize everything.
In this case, how a distro can be successfull without a clear delimitation or if the distro´s maintainers and the developers rely on each other ?
What would be the usability of the autopackage´s cross-compatibility if a package relies on the distro´s repository, in other words, if it isn´t autonomous ?
You certainly have though about that and I would resume it in this way : distribution is a incomplete centralization of the decentralized whole development of the open sources.
In itself, that sound like a contradiction, but how an OS can avoid it ?
best
luc
June 27, 2006 at 5:04 pm |
This comes back to the “what is the platform” question, which every vendor has a different take on and has never been fully resolved. Where the line is drawn is pretty much arbitrary and unimportant, the important thing is that it’s drawn in the same place everywhere … (or at least in each distribution)
June 29, 2006 at 3:04 pm |
Yes, in each distribution indeed but that implies a commitment of all the developers and all the distribution maintainers.
Otherwise a given distribution could never be a platform, and thus, unlikely to be as succesfull as FF ?
I say that because in your post, you´re speaking about “a” distribution, where it seems rather about Linux on desktop.
best
luc
June 30, 2006 at 5:06 am |
Right, exactly, needs standardisation between the distributions there. We really need a freedesktop.org/xdg-list equivalent for shared distro engineering IMHO.
June 30, 2006 at 3:11 pm |
I agree with you, though, do you realise that drawing such a line will distribute the tasks and the workload between the maintainers and the developers who would be willing to commit ?
Considers the Rob´s case, such a line may not afford him to give up the packaging of the concerned libs aformentioned. I don´t know what was his reasons, but if it´s a matter of ressource and time, he could not hold his commitment.
This involves that the packagers work for/with the developers rather than for the distros. In itself, it´s like a revolution : the current paradigm is put wrong side up.
Simple to heard through a radio.
One can also say that the current paradigm allocate the ressource wrongly. (though, sounds a little be too economic, but same meaning)
best
luc