What next?

Jeremy Huntwork jhuntwork at linuxfromscratch.org
Wed Feb 27 17:48:47 PST 2008

George Makrydakis wrote:
> In what way(s) could you do this that diy-linux and clfs do not do already? 
> How is it going to compete with the other two? Or three.. Or four... Or 
> Infinity?

I'm sorry for being dense, but I'm not sure I understand what you're 
asking here.

> Combining from the other projects? How? Why? They already _have_ it combined. 
> Why not work on _merging_ the communities into a single project? Doesn't that 
> make more sense since the goals are apparently the same since you are 
> choosing an evolutionary approach for LFS ?

Unless I misunderstood Gerard's proposal, that is what he is suggesting. 
We don't have (seemingly) the manpower and community interest any more 
to keep the current structure in place. I think the projects would have 
to be merged in order to continue.


> Isn't it a weakness in the social structure of LFS that it could not hold 
> these resources together? Educational use is no excuse imvho.

Very probably. And part of the issue, I think, has always been that 
different people see LFS from different viewpoints. This will always be 
the case to a certain extent, but perhaps, with a redesigned project, 
the potential for social problems can be taken into consideration as 
part of the re-design.

> Again, how can it be different from Gentoo, Sourcemage, T2, clfs, diy-linux, 
> Archlinux, GoboLinux etc... the list is endless on the meta - distribution 
> front. Package management is not going to help saving, if at all, anything.

How it will be different is something that will have to be discussed.

> I do not think it is geeky, it should be more "geeky" because there are MORE 
> THINGS TO LEARN than how to build a toolchain, but i am a "bystander" who has 
> no reason to doubt your intentions and is probably unimportant to you as a 
> contributing opinion (for the time being, as some other people, I do not like 
> some of your policies when it comes to "combining" things so there is really 
> not much to contribute *here* but an opinion).

I don't know what you refer to when speaking of 'policies for 
combining'. In any case, many of former LFS "policies" are probably moot 
at this stage, due to the recent heavy stagnation.


> Merging efforts (no matter to what you are referring to, you have a point). 
> Now this is the first and only thing that should be really considered. I read 
> that Mr. B You are absolutely correct on this one. Would you care to explain 
> if you are actually referring to attempts like LeafOS (you and a couple of 
> people where doing this long time ago, but it never lifted up), how things 
> should be done so that they do not end up in a standstill?

I doubt many people here were aware of the existence of LeafOS. But, 
since you bring it up, it is a shame that it didn't achieve its goals. 
It failed mostly because on a personal level, I didn't any longer have 
the consistent time to give it.

Anyway, LeafOS only existed because LFS did not seem to be moving 
forward or doing much of anything, really. And now, much of what is 
being discussed on the LFS lists are core concepts and goals of the 
abandoned LeafOS project. I would have preferred for it to happen with 
LFS in the first place, so I am happy now to hear these suggestions.

> ---------------------------------------------------------
> As to automation, package management ... give it a couple of days... Really. 
> You are hardly expecting this. Hardly. You will have much to discuss about 
> this in the following days. MUCH. And you will understand why some things
> developed patiently and unannounced before they ripe, create "glue points".
> ---------------------------------------------------------

Of course. I still have a few suggestions and ideas to put forward, but 
I am holding back a lot on these threads, because I'm waiting to see how 
the community as a whole responds first. And, I want to see Gerard take 
active action to make the decisions happen - to show that he is serious 
in reanimating the project, instead of just talking about new ideas.

> To me, the only issue that is holding back LFS and fragmenting it, is its 
> social structure. You are unlikely to have LFS-NG without taking into 
> consideration this factor. Until you do, you will be bleeding out people 
> elsewhere, or trying to "combine" things into branches etc. Other projects 
> are not supposed to be component or conceptual supermarkets.

You may be right that the social structure needs help. Suggestions on 
what needs fixing would be helpful, although, you may find that things 
which bothered you previously about LFS aren't a serious issue at present.


More information about the lfs-dev mailing list