[clc-devel] CRUX/CLC website proposal
Hi, I've been working on the new CRUX/CLC site lately, here's some screenshot of a working wikka setup: http://www.varlock.com/tmp/wikka/ Questions time: 1. Ok, apart from Wilber which is a temp thing, do you like the current style? I've also worked on punbb integration (last image) since there's been some request for a forum. 2. Regarding content, I've ported most of the documents on the current CLC and CRUX websites, but there is more to do. Should we setup a working temp site somewhere to fill together the pages before we go live? (then we can transfer the db to the actual site). 3. Integration I briefly spoke with Johannes regarding the integration of the a bug tracker into the website. Will we use cvstrac for this? Other options? Another question is the ports database page: I suppose we should change it a bit to fit the new contrib (do we still want to list the external httpup repos not in contrib?) Comments, ideas and contributions are welcome as usual ;) Regards, Simone -- Simone Rota WEB : http://www.varlock.com Bergamo, Italy MAIL: sip@varlock.com
Hi Simone, On Sun, Apr 17, 2005 at 01:05:31 +0200, Simone Rota wrote:
Hi,
I've been working on the new CRUX/CLC site lately, here's some screenshot of a working wikka setup:
http://www.varlock.com/tmp/wikka/
Questions time:
1. Ok, apart from Wilber which is a temp thing, do you like the current style? I like it a lot!
I've also worked on punbb integration (last image) since there's been some request for a forum. I wouldn't go for a forum before any trusted person agrees to be forum admin, and answers questions reliably. I wouldn't know who this could be at the moment.
2. Regarding content, I've ported most of the documents on the current CLC and CRUX websites, but there is more to do. Should we setup a working temp site somewhere to fill together the pages before we go live? (then we can transfer the db to the actual site). I think this would be great. I'd also like to see a list of things we'd like to put there, and maybe some structuring thoughts before we even start.
Also, I'd like to suggest that we think about subprojects (webpage, ports, tools etc.) first/now, just to make sure no one feels bad for not being involved. Also, I'd like to see a single responsible person for the website, which can take a decision if it has to be. Finally, I'd also suggest that we try to find motivated contributors which join the "web" subproject, in order to have them aboard already now. Thoughts?
3. Integration I briefly spoke with Johannes regarding the integration of the a bug tracker into the website. Will we use cvstrac for this? Other options? What's really cool about CVSTrac is the reporting based on cvs commit messages (think: security), even though this lost a bit of importance now that we have the security warnings via e-mail.
(Light) alternatives could be: http://www.mantisbt.org/ http://flyspray.rocks.cc/?p=Main_Page Or trac http://www.edgewall.com/trac/ which has Subversion support :-)
Another question is the ports database page: I suppose we should change it a bit to fit the new contrib (do we still want to list the external httpup repos not in contrib?) I spoke with Till about moving his ports db to the (future?) CRUX/CLC server. I think some *SQL based system would make a lot of sense.
Also, we could consider setting up a CVS/SVN/whatever repository for such "tools" projects, to allow more than just one person to contribute (this also applies to many other projects). Anyway, thanks for your work so far! Kind regards, Johannes -- Johannes Winkelmann mailto:jw@tks6.net Bern, Switzerland http://jw.tks6.net
On 04/17/05 11:18 Johannes Winkelmann wrote: Hi,
I wouldn't go for a forum before any trusted person agrees to be forum admin, and answers questions reliably. I wouldn't know who this could be at the moment.
Hehe, nobody here is a great fan of forums (me neither); we'll leave it out for now.
Should we setup a working temp site somewhere to fill together the pages before we go live? (then we can transfer the db to the actual site).
I think this would be great. I'd also like to see a list of things we'd like to put there, and maybe some structuring thoughts before we even start.
Since I believe it's good to have something to work on as discussions take place, I've setup a temp site; CLC members interested can send me a mail and I'll create an account and communicate the url / username / password. If there is general opposition against having a temporary site online at this point, I have no problem taking it off.
Also, I'd like to suggest that we think about subprojects (webpage, ports, tools etc.) first/now, just to make sure no one feels bad for not being involved. Also, I'd like to see a single responsible person for the website, which can take a decision if it has to be.
I think these subproject entities we're talking about can be divided into two groups: - already existing projects that so far has been developed more or less inside CLC, many times with no real plan or coordination ;) - external projects that has no or little relation with CLC (here I also include prjects not yet started) If I understand well some of our objectives would be to achieve better coordination and major partecipation to the first group, and offer some service (web/wiki, access to cvs repository) to the second one. To make a long story short, I think we fist need a list of services we can provide to subprojects, then identify and recruit for projects of group one, and finally publicize what we have for other project to join.
Finally, I'd also suggest that we try to find motivated contributors which join the "web" subproject, in order to have them aboard already now. Thoughts?
Yes, this is one of the first steps we should take.
Will we use cvstrac for this? Other options?
What's really cool about CVSTrac is the reporting based on cvs commit messages (think: security), even though this lost a bit of importance now that we have the security warnings via e-mail.
As Juergen suggests on this thread I think it's a good move to continue using cvstrac, since it has worked so well so far. This way we can concentrate on other things at the moment since there's a lot of other stuff going on ;)
I spoke with Till about moving his ports db to the (future?) CRUX/CLC server. I think some *SQL based system would make a lot of sense.
Also, we could consider setting up a CVS/SVN/whatever repository for such "tools" projects, to allow more than just one person to contribute (this also applies to many other projects).
Agree; as I wrote before once we know the services we can provide and the subprojects, than we're ready to go. So if someone has some spare time, a portdb that can be integrated into the website is a good candidate for a subproject. Regards, Simone -- Simone Rota WEB : http://www.varlock.com Bergamo, Italy MAIL: sip@varlock.com
On Sun, Apr 17, 2005 at 01:05:31AM +0200, Simone Rota wrote:
Hi,
Hello Simone,
I've been working on the new CRUX/CLC site lately, here's some screenshot of a working wikka setup:
http://www.varlock.com/tmp/wikka/
Questions time:
1. Ok, apart from Wilber which is a temp thing, do you like the current style?
Yep, looks very nice.
I've also worked on punbb integration (last image) since there's been some request for a forum.
To be honest, I don't like forums.
2. Regarding content, I've ported most of the documents on the current CLC and CRUX websites, but there is more to do. Should we setup a working temp site somewhere to fill together the pages before we go live? (then we can transfer the db to the actual site).
3. Integration I briefly spoke with Johannes regarding the integration of the a bug tracker into the website. Will we use cvstrac for this? Other options?
As long as we are using cvs as our RCS, I'd suggest so, because of the tight integration of cvs. For free we get an cvs browser and a nice timeline and the possibily to link commits to tickets.
Another question is the ports database page: I suppose we should change it a bit to fit the new contrib (do we still want to list the external httpup repos not in contrib?)
Comments, ideas and contributions are welcome as usual ;)
If I can help with something, please let me know. Greetings Jürgen -- Juergen Daubert | mailto:jue@jue.li Korb, Germany | http://jue.li/crux
Hi, Simone Rota wrote:
1. Ok, apart from Wilber which is a temp thing, do you like the current style?
Looks very good.
Another question is the ports database page: I suppose we should change it a bit to fit the new contrib (do we still want to list the external httpup repos not in contrib?)
I suggest to keep the external httpups in the portdb - so there is a central overview for every port avalible for CRUX. Johannes Winkelmann wrote:
I spoke with Till about moving his ports db to the (future?) CRUX/CLC server. I think some *SQL based system would make a lot of sense.
I will start this week with adapting my database to integrate into the wiki. Regards Till -- http://www.tbmnet.de
On Sun, 2005-04-17 at 01:05 +0200, Simone Rota wrote:
1. Ok, apart from Wilber which is a temp thing, do you like the current style? I've also worked on punbb integration (last image) since there's been some request for a forum.
I like it, looks very nice. As far as the forum is concerned, I don't care one way or another. In my opinion it's just a matter of presentation. After all, we have mailing list archives, which would provide the same access to past information, just in a different format. I can go either way. Since the ML works well and the consensus seems to be anti-forum, we might as well stick with the current setup there.
2. Regarding content, I've ported most of the documents on the current CLC and CRUX websites, but there is more to do. Should we setup a working temp site somewhere to fill together the pages before we go live? (then we can transfer the db to the actual site).
Seems reasonable. I also like the subproject idea, probably needs a fair bit of planning.
3. Integration I briefly spoke with Johannes regarding the integration of the a bug tracker into the website. Will we use cvstrac for this? Other options?
I like cvstrac and we're all used to it. Might as well keep using it. It's easy to move, certainly, since it's mostly contained in an sqlite db. The only alternative that really catches my eye is trac, since it supports subversion, but cvstrac will do if we don't go that route. Trac has some annoying dependencies, if I recall correctly.
Another question is the ports database page: I suppose we should change it a bit to fit the new contrib (do we still want to list the external httpup repos not in contrib?)
Whether we use Till's or not (which seems nice, I've no objections to it), it would be nice to adapt it to sql or something since there's quite a bit of disk searching involved right now. If there are external repos that don't go into contrib, perhaps it's best not to list them on the CLC pages. After all, isn't part of the point of the new contrib to sort of centralize and list everything? That would be something like a new unmaintained or a second contrib outside of contrib... Perhaps it could be argued that maybe subprojects are needed there as well, like gnome or mythtv or xfce4 or kde ports, etc., but perhaps they'd be better of listed only in their subproject pages, if we go that route. Matt (jaeger@freenode/#crux)
On 04/18/05 16:25 Matt Housh wrote:
3. Integration I briefly spoke with Johannes regarding the integration of the a bug tracker into the website. Will we use cvstrac for this? Other options?
I like cvstrac and we're all used to it. Might as well keep using it. It's easy to move, certainly, since it's mostly contained in an sqlite db. The only alternative that really catches my eye is trac, since it supports subversion, but cvstrac will do if we don't go that route. Trac has some annoying dependencies, if I recall correctly.
Regarding cvstrac, I think I managed to integrate it pretty well into the new website: http://www.varlock.com/tmp/wikka/05_cvstrac0.png http://www.varlock.com/tmp/wikka/06_cvstrac1.png http://www.varlock.com/tmp/wikka/07_cvstrac2.png (well, apart from the light blue that I'll have to change one day or another ;))
Another question is the ports database page: I suppose we should change it a bit to fit the new contrib (do we still want to list the external httpup repos not in contrib?)
If there are external repos that don't go into contrib, perhaps it's best not to list them on the CLC pages. After all, isn't part of the point of the new contrib to sort of centralize and list everything? That would be something like a new unmaintained or a second contrib outside of contrib...
I prefer to only list the opt / base / contrib ports too. But this way we could "loose" ports from people not interested in partecipating to the new contrib. I'm perfectly ok with it, but just in case we can consider a dedicated (wiki?) page listing just the urls of additional repos, to show that there's life outside clc/crux :)
Perhaps it could be argued that maybe subprojects are needed there as well, like gnome or mythtv or xfce4 or kde ports, etc., but perhaps they'd be better of listed only in their subproject pages, if we go that route.
I'm for the dedicated page approach; additionally, since there's the optional "group" tag into the clc Pkgfile header, this could be used for selecting a group of related ports. Just a combo-box away. -- Simone Rota WEB : http://www.varlock.com Bergamo, Italy MAIL: sip@varlock.com
Simone Rota wrote:
I prefer to only list the opt / base / contrib ports too. But this way we could "loose" ports from people not interested in partecipating to the new contrib. I'm perfectly ok with it, but just in case we can consider a dedicated (wiki?) page listing just the urls of additional repos, to show that there's life outside clc/crux :)
The advantage of including all httpups in the database is that you have one point where you can search for ports in the whole CRUX-universe. I suggest something like a checkbox "exclude httpups" in the search mask. Or favoring CLC-ports in the search result list. I agree with the idea that the portdb on the CRUX/CLC website should only list CRUX/CLC ports. But from the point of view of an average user it is annoying to have two databases or actually none for httpups. Regards Till -- http://www.tbmnet.de
On 04/18/05 19:04 Till Biedermann wrote:
The advantage of including all httpups in the database is that you have one point where you can search for ports in the whole CRUX-universe.
yes, this is also true.
I suggest something like a checkbox "exclude httpups" in the search mask. Or favoring CLC-ports in the search result list.
Good idea, I suggest presenting the feature as an "also include external repositories" checkbox, maybe disabled by default.
I agree with the idea that the portdb on the CRUX/CLC website should only list CRUX/CLC ports. But from the point of view of an average user it is annoying to have two databases or actually none for httpups.
If including external repos gains general consensus, I think we'd then need a cron job that fills the sql database from both the local and remote sites. I say this not to go into iplementation details, but to highlight the fact that the sql source should be unique. While I don't think I'll personally have much use for external repos, it's ok for me to include them as well. The only negative aspect I'd like to highlight is that this could encourage "lazy" maintainers not to join the new contrib, with all the unpleasant consequences we have now (many dup ports, incompatibilities, etc.) -- Simone Rota WEB : http://www.varlock.com Bergamo, Italy MAIL: sip@varlock.com
participants (5)
-
Johannes Winkelmann -
Juergen Daubert -
Matt Housh -
Simone Rota -
Till Biedermann