-------- Original Message --------
Subject: Re: Develop OpenForum? Inspired by both what I use and also
OpenBSD daily
From: Adam Wolk <adam.w...@tintagel.pl>
Date: Thu, December 21, 2017 5:47 am
To: cpb_p...@bennettconstruction.us
Cc: ports@openbsd.org

On Tue, Dec 19, 2017 at 11:11:09AM -0700,
cpb_p...@bennettconstruction.us wrote:
> I read about the problems with OpenBSD code reading and why it shut
> down.

I assume you are referring to #openbsd-daily code reads.


Yes



> Problems with the methods to study existing code and possibly submit
> diff's.
> 

I don't recall any problems with diffs? IRC is just a bad format for
live code
reading in a group but I don't recall having any problems with diffs
that
resulted from our reads on the channel.

I simply meant that an actual, usable diff could be sent as an
attachment.
HTML just doesn't support tabs, etc.


> Right now I am using Mwforum, which doesn't seem to have any security
> problems,
> beyond the very rare trolls that actually register.
> 

I don't see how a forum would help with live code reading. When we
discussed the
issue on IRC most people trended towards screen or terminal session
sharing or
optionally audio/video streaming. I could see something like
http://bxr.su/OpenBSD/ working for live reads if it had a built in live
chat and
allowed inline annotations that other people would see live.


For me, I see a certain gracefulness in being able to see an entire
thread dealing
with code in progress and also a discussion that was held months ago
where I could
easily bring up questions about the past discussions.

> I have been using it to host LedgerSMB's forums.
> One of the really nice features it has is an addon to convert emails
> into posts.
> I have been doing this quite nicely for a good period of time. These
> emails go
> to read only boards.
> 
> The forum supports attachments, (diff's would be a good usage).
> It also supports addons to change the forum controls into other
> languages.
> I wrote the Spanish addon.
> 
> The forum is written in Perl and supports multiple databases, PostgreSQL
> preferred.
> It works in Perl or mod_perl.
> 
> The downside is that the author is no longer working on the project or
> accepting
> any diff's for the code.
> 

That doesn't sound encouraging.
No, which is why I bring up the idea of starting from scratch and doing
both better work and adding to the existing capabilities.
Having a copy of say the tech@ on a separate forum board also makes
discussing current work a nice possibility without getting in the way
on tech@
Mwforum also supports private chats.

> I would like help developing a port to make our own forum software that
> works with
> our httpd and also perl and mod_perl.

See what I mention below.
If writing this is going to be done, then it should be portable.
mod_perl accepts modules written in C.


> 
> I would like to continue having the ability to post emails, such as
> ports@, tech@,
> misc@, etc.
> 
> Possibly also being able to send emails back for the existing mailing
> lists.
> 

I don't see the connection between that functionality and the code reads
we did
on #openbsd-daily. Perhaps I misunderstand your intention?

I find it rather difficult to go back to older mailing list threads and
go
next in thread, previous in thread instead of being able to see the
entire
thread at once. it's nice to be able to look at one message at the same
time
as reading a set of responses to that message.

> Personally, I would like to use PostgreSQL and Perl, although C would
> also be
> acceptable. Others would be better able to comment on the security of
> Perl vs C.
> 
> Is anyone already working on this or interested?
> I think this might be a good way to continue the code studying that was
> started.
> 
> 
> Thoughts?
> Chris Bennett
> 

If you want to discuss how to continue with the daily code reads, then
#openbsd-daily on IRC is a better place to have it.

Regarding the port of Mwforum, I personally have no need for it and
would rather
not advocate anoyone hosting an unmaintained forum that isn't receiving
security
updates. However I won't stop/discourage you from making a port.

No, I don't want to port existing software that may need major rewrites
due to
problems the one and only one author had.
We have all done big projects only to discover we screwed up much
earlier and
have to either rewrite everything or cobble together a crappy
workaround.
As a group project, all would also learn a lot about coding.
Maybe it would be best to add this as a project written in C to go along
with
the daily code reads.

I, personally would find code discussion with an on going project in C,
just
for us newbies to be rewarding. (Plus a proper license would be great)
So at this point, I think using Perl would not be the best way after
thinking
about it.

Reply via email to