Home
The Toolkit for Online Communities
17450 Community Members, 0 members online, 2306 visitors today
Log In Register
OpenACS Home : Forums : OpenACS Q&A : New CVS Instructions and Contribute Page

Forum OpenACS Q&A: New CVS Instructions and Contribute Page

Icon of envelope Request notifications

I have updated the contribute page and have included CVS guidelines. Feedback is welcome.
Collapse
Posted by Neophytos Demetriou on
Peter, could you also link to the package inventory document that you have created so that people know who maintains what?

[I have updated the document -- could you please commit the changes until others claim their packages].

Collapse
Posted by Neophytos Demetriou on
Peter, there should also be a page or a section in the contribute page that acknowledges people who have contributed *hugely* in the project (in the past/present). There is only one name in my mind right now: Ben Adida (co-founder of the project).

I'm curious as to what others may think.

Collapse
Posted by Peter Marklund on
Neophytos,
I committed the package inventory page and linked to it from the contribute page.

The contribute page should mainly present the people who are there to help today but I would be more than happy to give credit to past contributors. The difficulty of this task is that there are so many.

I think it might be appropriate though to add a note about who started the project - Ben, Don, and Roberto? I realize Ben's contribution has been enormous and I would like to work that into the page. Any concrete suggestions? I think those things might be more suitable on a history page that we could like to from the contribute page. I think there might already be a history page somewhere.

Collapse
Posted by Neophytos Demetriou on
Thank you Peter. I see that you've also updated the Team members page. Again, thanks. Perhaps, we could contact the people who might have not seen the relevant threads so that the document will be completed.

>>>>
I think it might be appropriate though to add a note about who started the project - Ben, Don, and Roberto? I realize Ben's contribution has been enormous and I would like to work that into the page. Any concrete suggestions? I think those things might be more suitable on a history page that we could like to from the contribute page. I think there might already be a history page somewhere.
<<<<

I think a history page linked from the "contribute" page will do. I think we could still provide some recognition to the people that have worked in the OpenACS project (I cannot come up with more names other than Ben right now but there are not many who have actively participated in the OpenACS project -- this recommendation is as much for the future). However, I agree with you that we could not acknowledge everyone who have helped say in the aD's original ACS -- a general statement would do, IMHO. What do you think?

Collapse
Posted by Lamar Owen on
Having been here at the beginning, when OpenACS was still ACS/pg, I believe a history page, particularly showing the three original core people (Ben, Don, and Roberto) that got things through at least 2.3 days (I was only involved peripherally with the project, with the driver and PostgreSQL itself) would be wonderful.

Going through the archives I have of the old acs-pg-dev mailing list is quite nostalgic, with the big transitions (at the time) of going from AOLserver 2.3 to AOLserver 3.0, with everybody having problems with the 3 betas (3b5 was the most stable at the time, IIRC); going from PostgreSQL 6.5.3 to PostgreSQL 6.6 (which became 7.0); and the biggie of going from ACS 2.x code to 3.x code.  'That time' being late 1999 and early 2000.  Getting ns_column and ns_table back in the driver is my 'claim to fame' at the time.  Sure made Ben, Don, and Roberto's day when I did it.  'It' being a graft of code that Mark Dalrymple did for the Oracle driver into the PostgreSQL driver, with some rewriting for PostgreSQL-specific issues.

And I still need to clean up that old graft job.... :-) And I need to get the driver using LISTEN/NOTIFY so that a connection being dropped while the backend is churning at a lengthy query doesn't happen any more (I'll post more details to the AOLSERVER mailing list when I know enough about what I want to do to get comments).

A lot of water has passed under the bridge since then, and lots of people have made much larger contributions.  But I'm glad I was able to help, even if just a little.  Dan Wickstrom helped out early on as well, IIRC.  And then he did the bind variable support that made 4.x possible.  The README file from the driver distribution has everybody (I think) mentioned.

Other names I haven't heard in a while showed up, such as Jamie Ross and Brent Fulgham.

Man, I've been running PostgreSQL for too long... :-) (AOLserver 2.2.1 with PostgreSQL 6.1.1 on Red Hat 4.1 was the first prototype (in 1997) -- things didn't really work well until Red Hat 5.0's glibc and PostgreSQL 6.2.1 teamed up with a glibc version of AOLserver 2.2.1 and Mark Dalrymple's beta postgres driver (Hi Mark...) -- which would have been the fall of 1997)

So, obviously, I'm all for a history page.  With permission of those mentioned in this post, I'd be glad to contribute e-mail archives to such a page.
Collapse
Posted by Peter Marklund on
Lamar,
thanks for that fascinating retrospect of OpenACS! Neophytos has expressed interest in helping us write a history page that we will link to from the contribute page. You should sync with Neophytos and others on that little project.
Collapse
Posted by mark dalrymple on
and Mark Dalrymple's beta postgres driver (Hi Mark...)

As much as I'd like to take credit for that, I can't. Someone else (I think it was Eric O'Laughlen on my AOLserver 2.x team) did that. I do take full credit for the oracle driver graft (and corruption) for aolserver 3 :-)

Collapse
Posted by Lamar Owen on
Mark, you are quite correct.  I wrote that from memory; I should have consulted my archives.  Although, messages that old are from the days when I still used Netscape Communicator, so I have to go hand search text files.  At that time I was using AOLserver 2.1 on Linux, and was testing 2.2 with PostgreSQL.  Eric sent me the beta, which enabled the extended_table_info stuff (the ad hoc query system of AOLserver 2.x is still one of the nicest ad hoc systems around).
Collapse
Posted by Neophytos Demetriou on
Lamar, Mark and the rest, as Peter has already pointed out I will prepare the history page and I would greatly appreciate if you review the document so that anyone who deserves credit gets it. I'll prepare a draft before the end of the week when I'm back in my office in Cyprus. Any help is more than welcome.
Collapse
Posted by Staffan Hansson on
I'm sure everybody knows this already, but since nobody has mentioned it I felt I should point out that there already exists a history page (it's hard to miss, since the OpenACS index page links to it). Ben's historical narrative seems to end with the creation of OpenACS. Hopefully the new history page will have something to say about the more recent history of the toolkit as well. Looking forward to reading all about it.
Collapse
Posted by Jamie Rasmussen on
I like the CVS guidelines document a lot.  Maybe it should mention what to do with binary files?  There are a fair number of them in our tree that were not added with -kb.
Collapse
Posted by Peter Marklund on
Jamie,
thanks for pointing out the special case of binary files! Since I don't know how to work with them myself, could you add a little note about that (you'll find an edit link at the bottom of the page)?

Thanks!

Collapse
Posted by Jeff Davis on
I fixed the cvswrappers file (in the /cvsroot/CVSROOT dir) a while ago but did not fix the existing binary files. Any new ones added should be -kb automatically (unless they don't match the existing cvswrappers extensions).
Collapse
Posted by Jamie Rasmussen on
Jeff's comprehensive cvswrappers file is the best way to handle this for the future. You can read about the problem and a solution for exisiting files in the Cederqvist manual. (You can check an individual file with the cvs status command.)

I've put a list of non-kb'ed files on my site. I didn't check them individually, but they all listed as "modified" when I cvs up on a Windows machine.

Collapse
Posted by Jeff Davis on
Jamie, I did a cvs admin -kb on all the files on your list maybe you could try another checkout and see that they come across unchanged...
Collapse
Posted by Jamie Rasmussen on
Looks perfect, Jeff.  Thanks!  And apologies for the tag goof in my last post.
Collapse
Posted by Don Baccus on
Since this thread's turned partially into a history discussion, here is the very first thread posted to an OpenACS/PG site, just after I'd spent a long saturday porting the bboard packages and all the pieces needed to support it. At the time it was running on an old P200 which until earlier this week I was still using as my firewall/ftp box.