Exherbo – my goals for the remainder of 2011 & 2012
(Disclaimer: The following is just my personal opinion and what *I* want to achieve for Exherbo. It’s fairly likely not everyone will agree with everything I’m going to state. 🙂 )
We’ve come a long way with Exherbo since we started:
– We have a pretty much complete set of packages to use Exherbo on desktop, notebook and server systems.
– With exheres-0, we have a very powerful and useful EAPI.
– We’re now a rather “stable” Linux distribution in that we don’t usually break things every other week anymore.
– We’ve stopped actively discouraging people to try Exherbo.
That said, I still have a few goals I’d like to see achieved:
– Join the Open Invention Network (OIN).
For some months now, I’ve been in discussions with the OIN and I’m happy to say that we’re going to join them. I hope (and am confident) we’ll get this done in 2011.
– Set up a structure that would allow us as a project (in contrast to individual developers) to accept donations.
Currently, Exherbo doesn’t really have any means to accept donations as a project. Sure, as individuals, we can accept them (e. g. a generous donator (who wants to remain anonymous) gave me a very nice server which we’re now using for Exherbo) but Exherbo as a project can’t.
There are several ways to accomplish this:
– Join an umbrella organisation that accepts donations on our behalf.
– Create an organisation for Exherbo ourselves.
Personally, I’d strongly prefer to create an organisation ourselves, e. g. as a German “eingetragener Verein” (“e. V.”), a (non-profit) registered association.
We’d be independent from weals and woes of any umbrella organisation and we’d have much more freedom to deal with anything concerning donations, laws, etc.
Since becoming a registered association would require some effort and actually cost money (in the case of an e. V. about EUR 75 to 120), I’d be willing to do the paperwork and pay for it myself.
(To be completely honest, I would not be willing to work and/or pay for joining some umbrella organisation.)
We’d be independent from weals and woes of any umbrella organisation and we’d have much more freedom to deal with anything concerning donations, laws, etc.
Since becoming a registered association would require some effort and actually cost money (in the case of an e. V. about EUR 75 to 120), I’d be willing to do the paperwork and pay for it myself.
(To be completely honest, I would not be willing to work and/or pay for joining some umbrella organisation.)
– Creating a “Developer’s Manual”.
We do have exheres-for-smarties and it’s really good but it can’t really stand alone. We still expect people to be at least somewhat aquainted with that thinG… That needs to change!
Thus, one of these days I’m going to start working on a real, full-blown “Developer’s Manual” which will incorporate everything you need to create and work on exheres.
Thus, one of these days I’m going to start working on a real, full-blown “Developer’s Manual” which will incorporate everything you need to create and work on exheres.
– Getting multilib/multibuild to work
We’ve had multilib branches for what feels like aeons. People have worked on multilib (spb, myself, compnerd & others) and they seem to have given up.
I’m not sure what’s wrong with our current approach (if you do, please comment on it here) but I know I really want to see the day when we can finally merge the multilib branches back to master.
Should that turn out to be impossible, we should get rid of them for good but we need to finally do something about multilib/multibuild.
I’m not sure what’s wrong with our current approach (if you do, please comment on it here) but I know I really want to see the day when we can finally merge the multilib branches back to master.
Should that turn out to be impossible, we should get rid of them for good but we need to finally do something about multilib/multibuild.
– A stable exheres-1.
As I said in the introduction, exheres-0 is fairly stable by now and, I think, pretty much feature-complete (maybe with the exception of multilib/multibuild).
Thus, I’d really get to the point where we formally declare that and make exheres-0 into a stable exheres-1 EAPI.
This is mostly a symbolic goal but I think it would give us, Exherbo, a slight increase in credibility and confidence.
Thus, I’d really get to the point where we formally declare that and make exheres-0 into a stable exheres-1 EAPI.
This is mostly a symbolic goal but I think it would give us, Exherbo, a slight increase in credibility and confidence.
– Participation in Google’s Summer of Code.
For three years we’ve been trying to get accepted into Google’s Summer of Code. This year, I’ve lead the third attempt to get accepted and, according to Google, we almost got there.
In 2012, I’m definitely going to try again by employing the improvements Google suggested. I do want Exherbo to participate in GSOC – and if it’s the last thing I’ll ever do! 😉
In 2012, I’m definitely going to try again by employing the improvements Google suggested. I do want Exherbo to participate in GSOC – and if it’s the last thing I’ll ever do! 😉
(This, too, is more of a symbolic achievement but it’s important to me and it would potentially help Exherbo achieve some sound technical goals as well.)
– Exherbo entry on Wikipedia
Entirely symbolic, this goal is just here because i think we’re nearing the point where Wikipedia’s guidelines about notability, etc. will be met.
– Working on patches
My personal pet peeve. We really need to improve on working on patches.
I’m not sure we’ll be able to achieve all these goals but I’m going to do my best to make all of these become reality for a bright future for Exherbo!
| I am and have been working on quite a few F/OSS projects: |
Can’t promise anything, but I’m hoping to do some work on multilib next time I get some time off work (around Christmas at the latest), assuming no-one else fixes it before then.
I’m really glad to see a post about exherbo’s progress here in p.e.o. I think it’s worth to blog something more about exherbo, because it’s a very good project and explain to others what is going on in the backstage it’s always a good thing(Considering the #exherbo-dev channel is just for few, ml is quite silent, no forum, the only information’s source is #exherbo).
So please keep bloging about exherbo!! 🙂
Said that, I agree with all points, but I especially feel the need for the developer’s manual(I asked for it months ago) and the multilib stuff (of course).
* Use debian style multiarch with /lib/triplet, /usr/lib/triplet
* Multilib as in lib32 lib64 is bad
* Move towards newer FHS and ignore some brain dead recommendations
* Move towards more machine logic and highly restructure paludis as the code is ugly as can be (fsmerger is hideous).
To expand on my last post:
* parse build systems for dependencies – m4, etc
* use svm, etc to classify flags
* create an optional integration testing framework – something like rspec for integration testing of different components
* create a dsl for modification of defaults
* use a default daemon monitor such as monit or bluepill
Goal number one – joining the Open Invention Network – has now been reached!