Sorry for the delay.
They are currently read only at the moment.
They are not intended to remain as they are now. The intent is to get
things running again, so that upload is operational and new packages can
go back to being submitted.
This is all stalled on me while I never seem to finish completely
restoring the system as it was.
--
jcap
> They are currently read only at the moment.
>
> They are not intended to remain as they are now. The intent is to get
> things running again, so that upload is operational and new packages can
> go back to being submitted.
>
> This is all stalled on me while I never seem to finish completely
> restoring the system as it was.
So, can anybody do something to help you, John?
--
Cosimo
> So, what's the current status?
> Will they remain as they are now?
> What about new packages?
John, I think it's time to make this happen, or to find someone else
who can make it happen. Thoughts?
Best,
David
I realize I have been remiss in both my attention to JSAN as well as to
keeping everyone up to date on what's going on. I'm sorry about that.
To reiterate, the state that still is, of JSAN, for those wondering, we
suffered a db loss many moons ago, and it was not recovered. The site
has been static since, and I did what I could at that time, to keep the
code in there, at least somewhat available to people.
Moving forward here's what I'm hoping to accomplish.
1. I wish to get the db rebuilt. The only missing piece will be
author's login passwords.
2. The site needs to be brought back up running dynamically off said
restored db.
3. A password retrieval mechanism needs to be added to retrieve said
lost passwords.
This will get us back to square one. A working (in as much as you wish
to view it) JSAN repo, as it was, before the crash. From there
discussion on future direction seem reasonable. It would be nice to
have the system back up as it was though, and not lose what's in there.
So, where does that leave the project?
Here's some good news. As of last night, I restored much of the rest of
the database. Namespaces, distributions, and authors are all in there
again, the latter with randomized passwords.
So that brings us to 2, putting the site back on the live db.
I am going to work on this tonight. I will without fail report a status
one way or another to the list by tomorrow night (~27 hours or so from
now at latest). Maybe sooner.
As soon as I have done the above, I want to get the codebase active
again. To that end, my thinking is to convert our svn repo over to Git,
get it up on GitHub, but I'm not married to this idea. Seems a good way
to get people looking at the code again though.
As soon as that's done, people can fix the password retrieval feature if
I can't get to it.
My personal goal here is try and foster this project back to life. If I
can get to the point mentioned above, I feel I can remove myself as a
roadblock, and then community can contribute in a more normal fashion.
I also of course have lots of ideas on how to move the project forward
from all the more creative standpoints, but for now I'll settle for,
make it work at least as it once did.
Short of how long it's taken me to get here, please let me know if you
feel I've made any poor decisions above, and you'd rather see things
done differently. I am available both on list and off.
John
--
jcap
> I realize I have been remiss in both my attention to JSAN as well as
> to
> keeping everyone up to date on what's going on. I'm sorry about that.
Thanks for finally writing, much appreciated.
> Here's some good news. As of last night, I restored much of the rest
> of
> the database. Namespaces, distributions, and authors are all in there
> again, the latter with randomized passwords.
jcap++
> So that brings us to 2, putting the site back on the live db.
>
> I am going to work on this tonight. I will without fail report a
> status
> one way or another to the list by tomorrow night (~27 hours or so from
> now at latest). Maybe sooner.
Excellent, thanks.
> As soon as I have done the above, I want to get the codebase active
> again. To that end, my thinking is to convert our svn repo over to
> Git,
> get it up on GitHub, but I'm not married to this idea. Seems a good
> way
> to get people looking at the code again though.
GitHub++. I've already moved Module::Build::JSAN and Test.Simple there.
> As soon as that's done, people can fix the password retrieval
> feature if
> I can't get to it.
>
> My personal goal here is try and foster this project back to life.
> If I
> can get to the point mentioned above, I feel I can remove myself as a
> roadblock, and then community can contribute in a more normal fashion.
Yeah, that will be very helpful. Especially if tests can be added and
the whole thing could be tweaked so that it's easy to set up and run
locally, to simplify development. Contributions I sent to Casey in the
past were largely blind, as I had no way to test them. He would fix my
syntax errors; a PITA, I'm sure.
> I also of course have lots of ideas on how to move the project forward
> from all the more creative standpoints, but for now I'll settle for,
> make it work at least as it once did.
You should put them into the project wiki on GitHub.
> Short of how long it's taken me to get here, please let me know if you
> feel I've made any poor decisions above, and you'd rather see things
> done differently. I am available both on list and off.
This sounds great! Thank you John.
Best,
David
You should put them into the project wiki on GitHub.
David, correct me if I'm wrong, but what I believe he meant was, after I
get the SVN repo converted to Git, and in GitHub, then start to fill out
the wiki for that repo.
--
jcap
Good progress actually. Got the basics of the site humming again,
building itself off the DB. Apache conf is mostly there. A few minor
issues, and then I can dive in and make sure Jause is functional and it
can be made live.
More to come.
For those that care, posting short blurbs here as well:
http://twitter.com/openjsan
John
--
jcap
> Do you mean your module-build-jsan project on GitHub, here: http://github.com/theory/module-build-jsan/tree/master
> ?
> I think starting a wiki for project ideas is a great idea -- but
> what about something more "central"?
No, I meant the wiki for JSAN, once John moved the source over to it.
It will be its own project.
David
Needs a few corners rounded but most major issues seem ok.
Have to do a few minor things, run a few test uploads, and then move it
into production.
--
jcap
No particular framework. Some Apache::Template, TT2, etc.
It needs to be cleaned up. Whatever we pick is fine, but I want it to
be easily deployed, as David was saying, so that devs can check it out,
get it running, and test their changes in a for real environment.
John
--
jcap
I have a development site setup to work on while what you see at
www.openjsan.org remains untouched. When I am done, I will get
everything cleaned up, checked back in, and make that accessible via
www.openjsan.org
The dev site is not really public at this time. It is more of my dev
instance, much like I hope all of us can have our own dev instances to
work on before committing changes.
--
jcap
Well, as of now, the live site was rebuilt a few minutes ago from the
restored db and redeployed codebase/templates.
I've made a few quick adjustments to the homepage too, and slapped a
twitter news widget on the front for better or worse to show some new
progress.
I have to do some testing on the import procedure, some cleanup of the
code, make sure all is in svn, migrate svn to git, and get it up on
GitHub.
I'll keep you posted.
--
jcap
> The actual site www.openjsan.org is rebuilt periodically from a set of
> templates, and the most recent distribution info when things are
> working.
W00t! I'll try to get a new version of Test.Simple uploaded this
weekend.
> I have to do some testing on the import procedure, some cleanup of the
> code, make sure all is in svn, migrate svn to git, and get it up on
> GitHub.
I suspect that there isn't actually that much in SVN, given the SVN
failure last year. Looks to me like the whole thing is just at r72. If
that's the case, then you might as well skip importing the history,
since it's negligible, and just create a new git repo and add
everything to it as a first commit and push that.
My $0.02, having just gone through the [hell] of a major migration.
[hell]: http://justatheory.com/computers/vcs/git/bricolage-to-git.html
Best,
David
Previously the openjsan repo consisted of many pieces including, the
openjsan webapp, perl modules that were deps of the web app, but also
functional in their own right, some python (planet js), some jsan libs
for a few users, some other perl modules.
I have taken the webapp (site & jause) and migrated them into 1 Git
repo.
The rest is forthcoming.
The site has been updated to reflect this change and link to our GitHub
issue tracker and wiki.
http://github.com/openjsan - all the projects will reside eventually
http://github.com/openjsan/openjsan - the current part discussed above
http://github.com/openjsan/openjsan/issues - todo, bugs, requests
http://wiki.github.com/openjsan/openjsan - wiki
--
jcap
Sweet, thanks John!
Best,
David