swarm-modeling
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Looking at Swarm's ftp


From: Alex Lancaster
Subject: Re: Looking at Swarm's ftp
Date: 25 Oct 2001 03:42:37 -0700
User-agent: Gnus/5.09 (Gnus v5.9.0) Emacs/21.1

>>>>> "PJ" == Paul E Johnson <address@hidden> writes:

[...]

PJ> I would like to gravitate toward a setup that separated the Obj-C
PJ> and Java applications, and was like so:

[...]

I think the idea is great, the "anarchy" directory is getting a
little, well, anarchic.... ;-)   

One suggestion, let's rename "anarchy" to the more standard "contrib"
(a la Red Hat and Debian), and eliminate/combine the "evaluated"
category, i.e. move all currently maintained packages into "supported"
and unmaintained packages (i.e unmaintained by the SDG) into "contrib"
and then rename "supported" to "sdg" (or somesuch), since "supported"
perhaps implies that we actually have a phalanx of staff to do real
support.

So we then would have simply:

apps
     java
          sdg
          contrib
     objc
          sdg
          contrib

PJ> Marcus got me permission on the ftp server to rearrange things,
PJ> but I'm not doing it yet until I wait for criticism/feedback of my
PJ> proposal.

PJ> As a separate initiative, I want to make it easier to figure out
PJ> what these packages are. From here on out, all contributed
PJ> programs should include a text file with a name like
PJ> "ProgramName.txt" that at least includes this information:

PJ> Author Name: Author email: Author http: Title of Program: Date:
PJ> Most Recent Swarm against which this runs: Language in which
PJ> program is written: Abstract describing purpose of model:

PJ> I will provide a template, authors can fill in.

I would only suggest that we stick with the *.lsm format as far as
possible, so other software tools can still parse it, and simply add
extra fields that correspond to more Swarm-specific stuff.  That way
we can leverage third-party apps that read .lsm, and I can use the
info to turn each .lsm into an XML fragment that can be rendered on
the website.

Also the <packagename>.lsm should have a matching tarball
i.e. <packagename>-<version>.tar.gz so that the script to generate the
website link and blurbs can be more or less automated...  If it
doesn't have an .lsm file at all, that's OK, the script could just
generate a link with no blurb.

PJ> In the long run, our ftp archive will have tarballs accompanied by
PJ> these little text ProgramName.txt files, the same way you see ftp
PJ> archives on ftp://sunsite.unc.edu/pub/linux/utils/ have *.lsm
PJ> files.

PJ> I may have to generate these .lsm files myself for some old
PJ> applications. I love sss and arborgames so much I'm delighted to
PJ> do it.  Others, well, I don't know that I'm qualified, and the
PJ> original authors might receive some harrassment.

And so they should! ;-)

PJ> Now is the time for you to rip on me about this, otherwise I'm
PJ> going to start doing it.

Sounds great!  Can't wait!  Go Paul, go!

Alex
-- 
   Alex Lancaster * <address@hidden> * www.santafe.edu/~alex 
Dept. of Integrative Biology, UC Berkeley (ib.berkeley.edu) * +1 510 642-1233
    & Swarm Development Group, Santa Fe, New Mexico (www.swarm.org) 


                  ==================================
   Swarm-Modelling is for discussion of Simulation and Modelling techniques
   esp. using Swarm.  For list administration needs (esp. [un]subscribing),
   please send a message to <address@hidden> with "help" in the
   body of the message.
                  ==================================


reply via email to

[Prev in Thread] Current Thread [Next in Thread]