| View previous topic :: View next topic |
| Author |
Message |
thewtex
Tux's lil' helper
Joined: 22 Jun 2007
Posts: 93
|
 Posted: Wed Jan 16, 2008 4:32 pm Post subject: |
 |
|
| Quote: |
With this kind of model, it seems bazaar would work better. One thing I wanted gen2 to avoid is having to edit a sources.list immediately upon installation to get a workable desktop/server/workstation. I also would not want to have to modify the sources.list during the installation as people had better ebuilds. A default list would have to be maintained and then you are basically getting into the same power structure as gentoo right now. Please correct me if I am missing something. I do agree there needs to be a way to choose individuals but that should be secondary. By specifying a desired stability level, the system should be smart enough to get ebuilds from everyone who registers thier computer as a repository and pick the best ebuild at that time. I have no idea how well this will work in practice but thats what testing is for! |
You are right about the default list. It must always exist. Also, your automatic extension idea seems to be much better.
Something to keep on the priority list is to make the core easily and completely replaceable so all the the community resources remain available on replacement. Here, we apply the Gentoo philosophy of choice in on itself. If someone wants to make a new core for an esoteric embedded or supercomputer architecture offsite, they can do so. If they don't like how gen2 core is going, they can easily go their own way.
| Quote: |
git's scaling is widely known to be great, and I would be open to a git implementation. It just seemed for certain features that bazaar fit the overlay model a bit better (and seemed easier to use for someone who has limited SCM experience). |
With paludis you get rid of the fat python requirement and a huge performance increase. With bazaar, both advantages are both lost  . Git may not be as distributed as bazaar, but it was certainly designed as a distributed framework. When cloning a git repository, you git;) an full equivalent to the original, yet it stays very small. Git was also designed to be very hackable, so it is easy to make it do what you want. I am a newbie to SCM, but found git easy to understand and get up and running.
I watched the google video by Randal Schwartz, http://video.google.com/videoplay?docid=-3999952944619245780
Then looked at the tutorial for git http://www.kernel.org/pub/software/scm/git/docs/tutorial.html
and everyday git: http://www.kernel.org/pub/software/scm/git/docs/everyday.html
Documentation for git is quite good.
| Quote: | | I feel like its easier for a person who doesn't want to report successes to edit thier config, than it is for a person who doesn't care to do the same. The feature will be described in the main installation docs so it's easy for people to opt out. |
People have a bad experience at this point with the collection of their personal information without their consent. The government does phone taps and internet taps. Microsoft and software on Windows collect personal information. The program will have more participation if people are given trust and respect.
Another option is to not make it opt-in or opt-out at all, but a necessary question that must be answered during installation. This doesn't help the quick/easy install idea, but it is not that bad, and their are exceptions to every rule.
Will keep in touch:) |
|
| Back to top |
|
 |
thewtex
Tux's lil' helper
Joined: 22 Jun 2007
Posts: 93
|
| Posted: Wed Jan 16, 2008 9:38 pm Post subject: |
|
|
There are others with the some mindset; see zeroum's comments here
http://blog.funtoo.org/2008/01/tale-of-two-cultures.html
| Quote: |
May I propose a solution to the scalability issues, and the user-developer distance problem?
Gentoo and the Portage Tree is like CVS and SVN, they are centralized content management systems made of code and people.
I believe we should go distributed, like Git, Bazaar, Mercurial, etc.
Every part of the tree would go to overlays, and every overlay would be treated like first class citizens.
KDE for example, would be an overlay, and even KDE extragears could be an overlay of KDE.
There could also be competiting overlays, why stop them? Two KDEs? Yes, the Free Software community usually tackles the same problem from multiple fronts, how many distros do you know, how many inter-process communications systems are there? And they often merge back to an unified and larger than the sum of its parts solution, DBus is that. Just don't cross the same namespace and you'll be fine.
Gentoo-Core would just be an [obligatory?] overlay of the main tools needed to build this all. Use GWN to tell "Danger, Will Robinson! Expat 3.0 is coming in a month!"
b.g.o would take everyone's bugs, commit access? Just like Git. You trust people, you merge their changes, you'd sync to many overlays, not just one. Documentation? Make Gentoo-Wiki a first class citizen too, port the documentation over there. The same with many Gentoo utility sites. f.g.o's Unsupported Software forum would thrive.
No elitism, bring everyone to the same level.
Are we a meta-distro? Yes, not only that everyone builds their own distro, but that we make Gentoo and Gentoo make us back.
2:46 PM
|
Also, I registered gen3.org. Would reduce name confusion due to the different pronunciation, but has the same clever relationship. Short and sweet. Let me know if you want it. |
|
| Back to top |
|
|
micr0c0sm
Tux's lil' helper
Joined: 29 Oct 2005
Posts: 148
Location: New York
|
| Posted: Thu Jan 17, 2008 8:47 am Post subject: |
|
|
| thewtex wrote: | | Quote: |
With this kind of model, it seems bazaar would work better. One thing I wanted gen2 to avoid is having to edit a sources.list immediately upon installation to get a workable desktop/server/workstation. I also would not want to have to modify the sources.list during the installation as people had better ebuilds. A default list would have to be maintained and then you are basically getting into the same power structure as gentoo right now. Please correct me if I am missing something. I do agree there needs to be a way to choose individuals but that should be secondary. By specifying a desired stability level, the system should be smart enough to get ebuilds from everyone who registers thier computer as a repository and pick the best ebuild at that time. I have no idea how well this will work in practice but thats what testing is for! |
You are right about the default list. It must always exist. Also, your automatic extension idea seems to be much better.
Something to keep on the priority list is to make the core easily and completely replaceable so all the the community resources remain available on replacement. Here, we apply the Gentoo philosophy of choice in on itself. If someone wants to make a new core for an esoteric embedded or supercomputer architecture offsite, they can do so. If they don't like how gen2 core is going, they can easily go their own way.
|
The core will be like any other repository in paludis - if you don't want to use the core repository, just delete /etc/paludis/repositories/core.conf. It will sync /etc/paludis/repositories/*.{conf,sh,py,rb} so the core will be easily replacable.
| Quote: |
| Quote: |
git's scaling is widely known to be great, and I would be open to a git implementation. It just seemed for certain features that bazaar fit the overlay model a bit better (and seemed easier to use for someone who has limited SCM experience). |
With paludis you get rid of the fat python requirement and a huge performance increase. With bazaar, both advantages are both lost . Git may not be as distributed as bazaar, but it was certainly designed as a distributed framework. When cloning a git repository, you git;) an full equivalent to the original, yet it stays very small. Git was also designed to be very hackable, so it is easy to make it do what you want. I am a newbie to SCM, but found git easy to understand and get up and running.
I watched the google video by Randal Schwartz, http://video.google.com/videoplay?docid=-3999952944619245780
Then looked at the tutorial for git http://www.kernel.org/pub/software/scm/git/docs/tutorial.html
and everyday git: http://www.kernel.org/pub/software/scm/git/docs/everyday.html
Documentation for git is quite good.
|
Great resources I will be watching and reading them once I get home.
| Quote: |
| Quote: | | I feel like its easier for a person who doesn't want to report successes to edit thier config, than it is for a person who doesn't care to do the same. The feature will be described in the main installation docs so it's easy for people to opt out. |
People have a bad experience at this point with the collection of their personal information without their consent. The government does phone taps and internet taps. Microsoft and software on Windows collect personal information. The program will have more participation if people are given trust and respect.
Another option is to not make it opt-in or opt-out at all, but a necessary question that must be answered during installation. This doesn't help the quick/easy install idea, but it is not that bad, and their are exceptions to every rule.
Will keep in touch:) |
Good compromise - have the variable commented out and require it to be uncommented before executing paludis --install or what have you. |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
Gentoo Chat
