[net.news.group] Net.software.projects group idea

aburt@isis.UUCP (Andrew Burt) (08/09/85)

Being a software developer like so many of us, it strikes me that one of
the golden opportunities of having the net is going unnoticed -- distributed
projects.

In other words, many people around the net might

	o  be working on the same or similar project; or
	o  be willing to work on it but can't do it alone; or
	o  have worked on it but ran into problem X and didn't
	   know how to solve it; or
	o  can't do it themselves but are willing to port it to
	   system X

and so on.  For instance, note the relatively high number of 'grep'
postings to net.sources and mod.sources over the past year or so.

Rather than having lots of individualized efforts going on, why not
create a newsgroup, say, net.software.projects, whose sole purpose is to
post ideas on software you're planning to write / are writing / would
like to see written.

Then, anyone interested in a given idea could respond to the author
with information on how s/he could help.  That is, create a project
effort where the participants aren't in the same room but communicate
via e-mail.

Now, before anyone says "Yeah, great idea, but let's only make it for
postings that require multiple people involved," or "not for novices,"
or "only for volunteer efforts," or any other restrictions, I would like
to clarify my intent.  I would want such a group to exist for anyone who
felt the need to post to it.  So if a novice posts saying he wants help
writing a checkbook balancer, fine.  If some kernel hacker wants to
do terminal paging at the driver level, fine.

I further think it shouldn't be restricted to public domain or volunteer
efforts, but should include efforts put into software for profit (at
either the individual or big business level).  Perhaps there might be
some crossover from net.jobs, since it would be paid for, but when was
the last time you saw anything in net.jobs asking for someone just to
help write a new grep?  Net.jobs appears to have only full time
positions advertised in it.  This group would work equally well for full
time positions with specific duties or just for small contract type work.

In an effort to keep it as free of restrictions as possible, basically
any project idea should be postable, even things as banal as "A friend
of mine wants someone to help bring up news software on an Xenix
system".  For that person, this is a software project (maybe not at the
development end, but still a software project).

The only groundrules should be that

	o  no followups are posted -- all replies should go directly to
	   the author who would be responsible for coordinating the
	   project.  Even if this isn't enforced in current news
	   software, the charter of the group should include "no followups".

	o  the problem and the environment are clearly stated:  e.g.,
	   what flavor of Unix it will be for, any hardware
	   dependencies, volunteer or not, deadlines if any, ...

	   Perhaps a standard form should be created a la

		Synopsis:
		Version(s) of Unix targeted for: (e.g., "all", "4.2")
		Hardware targeted for: (e.g., "all", "Vaxen")
		Special hardware or software needed:
		Deadline:
		Remuneration: (e.g., "warm fuzzies", "hourly rate")

Granted, the net.software.projects sort of idea doesn't apply only to
software projects, but equally well to hardware, mathematics, fiction
writing, or anthropological studies, ... ad infinitum.  If it turns out
that there's a desire for more groups (net.math.projects...) so be it
and long live the net.

It's possible the whole idea of cooperative ventures like this will flop,
and I suspect it will (pessimist, me?), but it might be worth a try.

				Andrew

[A final note on my choice of newsgroups for this posting (people seem
so touchy on these matters):  I included net.sources since that's where
the people most likely to use this group will be found, and there's no
net.sources.d; net.jobs since employers might want to know about it;
and of course net.news.group, to which all followups should go.]
-- 

Andrew Burt
University of Denver
Department of Math and Computer Science

UUCP:	{hao!udenva, nbires}!isis!aburt
CSNet:	aburt@UDENVER	(NOT udenva, as above...)
ARPA:	aburt%udenver.csnet@csnet-relay.arpa

control@almsa-1 (William Martin) (08/12/85)

This appears to be a rather good idea... One point, regarding the
proposal for having a "no followups" rule: I think that might be
counterproductive. Suppose I post a note saying that I am working on
idea "X". Across the net, there are twenty other people also working on
idea "X" (the usual simultaneous occurrence of ideas when the time is
ripe for them -- the history of science is full of examples of this). If
they only mail to me, it is then up to me to put all of these people in
contact with each other. I may not do this -- I may not know how, or may
not have time, or maybe I want to get their ideas for myself and not let
them collaborate. However, if followups are allowed, and all twenty
people post notices that they, too, are working on idea "X", the info
and the identities are public, and it is also apparent to others, who
might have thought about devoting time and resources to idea "X", that
it is being handled already by a lot of different people and/or
organizations, and mybe they would be better off putting their resources
into some other area. 

Of course, nothing would keep a "no followup" rule from being avoided by
people simply posting newly-originated notices with similar subjects (or
even putting a "Re: <old title>" in manually, so it would not be
enforceable in any event -- it just seems that making that be an
explicit goal might not be wise...

Other than that, seems like a pretty good idea. Guess that means it'll
never happen... :-)

Regards,
Will Martin

ARPA/MILNET: wmartin@almsa-1.ARPA    UUCP/Usenet: seismo!brl-bmd!wmartin

umdhep@eneevax.UUCP (Todd Aven) (08/13/85)

I am in favor of something along the lines of a project clearinghouse. I think
that it should be moderated, though. For an experienced net user cum moderator,
directing more novice traffic directly to already established projects or
finished software would make the newsgroup a more worthy subscription, such
as mod.sources vs. net.sources. I am not an experienced net user nor moderator
(barely familiar with Unix*) so I don't intend to volunteer myself except in
the unlikely case that formation of the newsgroup fails for lack of a moderator.
I hope everyone with any interest in the group will show support now to get 
this off the ground.

tsa

============================================================
|Todd Aven               MANAGER@UMDHEP.BITNET             |
|Softwear Sweatshop         AVEN@UMCINCOM (arpanet, bitnet)|
|High Energy Physics      UMDHEP@ENEEVAX.UUCP              |
|University of Maryland                                    |
|College Park, MD 20742      (301)454-3508                 |
============================================================

terry@neurad.UUCP (Terry L. Ridder) (08/13/85)

> Being a software developer like so many of us, it strikes me that one of
> the golden opportunities of having the net is going unnoticed -- distributed
> projects.
> 
> In other words, many people around the net might
> 
> 	o  be working on the same or similar project; or
> 	o  be willing to work on it but can't do it alone; or
> 	o  have worked on it but ran into problem X and didn't
> 	   know how to solve it; or
> 	o  can't do it themselves but are willing to port it to
> 	   system X
> 
> and so on.  For instance, note the relatively high number of 'grep'
> postings to net.sources and mod.sources over the past year or so.
> 
> Rather than having lots of individualized efforts going on, why not
> create a newsgroup, say, net.software.projects, whose sole purpose is to
> post ideas on software you're planning to write / are writing / would
> 
> UUCP:	{hao!udenva, nbires}!isis!aburt
> CSNet:	aburt@UDENVER	(NOT udenva, as above...)
> ARPA:	aburt%udenver.csnet@csnet-relay.arpa

	I support this idea and the given guidelines. I have been
	thinking of posting a request to the net concerning a project
	which I have been working on, but did not feel that it would
	be accepted by the net. Now my thoughts are different, so I 
	will now post my project and request any flames and or offers 
	of help be e-mailed to me.

	PROJECT:

	Write a C language library of Common-Lisp primatives.

	REFERENCES:
	
	Lisp - 2nd Edition by Horn and Winston.
	Common-Lisp - By Steele
	Symbolic Lisp Reference manual - By Symbolics Inc.

	PURPOSE:

	The goal of this project would be to have a public domain
	Common-Lisp compiler, and a Lisp primitive library.

	The Lisp library would be used in C programs at first and
	later by a Lisp compiler.

	OUTLINE:

	Phase 1: Write a Lisp library that can be used in C programs.

	i.e  Implement the 'car', 'cdr', 'print', 'princ', etc.

	Phase 2: Using the Lisp library begin work on Lisp compiler.

		I  would like to see this done using lex and yacc.

	Phase 3: Using the compiler from Phase 2 write a new compiler
	in Lisp itself.

	Phase 4: Continue to refine the compiler.

	I am open to a discussion concerning the project.

 	It must be stated that the result of the project will be 
	public domain software.   

	
-- 


  ===========================================================================
  |UUCP: seismo!neurad!bilbo|wiretap!{root, tlr}			    |
  |UUCP: seismo!neurad!bilbo!{root, tlr}				    |
  |UUCP: seismo!neurad!terry						    |
  | 									    |
  |U.S.SNAIL: Terry L. Ridder, 401 Cherry Lane E301, Laurel, Maryland 20707 |
  |									    |
  |Ma Bell: Home: 301-490-2248 Work: 301-859-6271 Work: 301-859-6642	    |
  |									    |
  ===========================================================================

bc@cyb-eng.UUCP (Bill Crews) (08/13/85)

I think it is a great idea, although I have reservations about the "for pay"
idea.  I've coordinated projects before involving independent software
contractors, and it seems to me that it is hard enough in one city, person-to-
person, to ensure that everyone feels that fairness is being maintained.

> Granted, the net.software.projects sort of idea doesn't apply only to
> software projects, but equally well to hardware, mathematics, fiction
> writing, or anthropological studies, ... ad infinitum.  If it turns out
> that there's a desire for more groups (net.math.projects...) so be it
> and long live the net.

Why isn't it net.projects.software, net.projects.hardware, etc?  Seems to me
to make a LOT more sense.
-- 
  /  \    Bill Crews
 ( bc )   Cyb Systems, Inc
  \__/    Austin, Texas

[ gatech | ihnp4 | nbires | seismo | ucb-vax ] ! ut-sally ! cyb-eng ! bc

bdw@drutx.UUCP (WelkerB) (08/14/85)

****** eater? oh, eater..... here eater....BAM! *********************

I also approve the project clearing house idea. It would be a good way
(it seems to me) to increase the quantity of quality public domain
software available to netnicks, And an excellent way to implement a
public forum on design as applied to specific projects. One of the
first subgroups could be "net.project.newsware" 8-).

		Bruce @ The Denver Works: AT&T-ISL
		ihnp4!drutx!bdw

umdhep@eneevax.UUCP (Todd Aven) (08/14/85)

The motion to make the group net.projects.? is a good one. It allows the
group to start out as net.projects, and when (not if |-) ) sufficient
interest in projects makes diversification reasonable, subgroups .software,
.hardware, .cooking, .whatever could be formed to meet demand, whereas
starting out as net.software.projects implies a backwards hierarchy.

tsa
============================================================
|Todd Aven               MANAGER@UMDHEP.BITNET             |
|Softwear Sweatshop         AVEN@UMCINCOM (arpanet, bitnet)|
|High Energy Physics      UMDHEP@ENEEVAX.UUCP              |
|University of Maryland                                    |
|College Park, MD 20742      (301)454-3508                 |
============================================================

itkin@luke.UUCP (Steven List) (08/15/85)

In article <134@isis.UUCP> aburt@isis.UUCP (Andrew Burt) writes:
>Being a software developer like so many of us, it strikes me that one of
>the golden opportunities of having the net is going unnoticed -- distributed
>projects.
>...
>Rather than having lots of individualized efforts going on, why not
>create a newsgroup, say, net.software.projects, whose sole purpose is to
>post ideas on software you're planning to write / are writing / would
>like to see written.

I think the idea is a good one.  I would make two suggestions: make it
a moderated group and name it mod.projects.software.  Then if Andrew's
suspicion that there is applicability outside software materializes, the
other names fall on naturally.

As for it being a moderated group, there are several potential
advantages:

    - a moderator could maintain an up-to-date list of ongoing projects
        and their contacts.  This list could then be organized along all
        sorts of useful lines (by category, region, language, and so on)
    
    - a moderator could ensure that ideas of similar type get grouped
        together in the postings

I would suggest that Andrew consider becoming moderator if there is
indeed a desire for this group.  The only immediate question I have is
where's the traffic?  According to recent reports, the idea is to create
a newsgroup to relieve an existing group of excessive, special-interest,
traffic.  Where is it?
-- 
***
*  Steven List @ Benetics Corporation, Mt. View, CA
*  Just part of the stock at "Uncle Bene's Farm"
*  {cdp,greipa,idi,oliveb,sun,tolerant}!bene!luke!itkin
***

peter@baylor.UUCP (Peter da Silva) (08/19/85)

> I would suggest that Andrew consider becoming moderator if there is
> indeed a desire for this group.  The only immediate question I have is
> where's the traffic?  According to recent reports, the idea is to create
> a newsgroup to relieve an existing group of excessive, special-interest,
> traffic.  Where is it?

Net.sources, net.sources.bugs, net.micro... lots of places.

Oh yes, I like the idea of it being a moderated group. Someone with a good
DBMS, so they can lookup who's working on what.
-- 
	Peter (Made in Australia) da Silva
		UUCP: ...!shell!neuro1!{hyd-ptd,baylor,datafac}!peter
		MCI: PDASILVA; CIS: 70216,1076