Riad,
(I knew this would come to this point) OK, I am not angry or insulted, and
let me apologize for the tone of my last letter. It was written in haste -
it seemed like I missed the first round of corespondance so the timer
started. Additionally I was speaking only of what I have done when I
said that it wasn't something I just slapped together.
I'm sure there are different things in each module that would be
advantageous under certain circumstances. I would definitly like to hear
about any problems roadblocks you've come across and how to fix them. I
can share my findings with you aswell.
I think I'll go do some work now on pickle.
Cheers,
aryeh
------
Aryeh Goldsmith
Director of Interactive Programming
CheckMate Internet Strategies & Solutions
http://www.checkmate.net/
On Sun, 16 Jul 2000, Riad Wahby wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> >The project hasn't died, it simply hasn't been updated. I initially took
> >the Net::IRC module which I was familiar with and have worked on and
> >rewrote the connection class to suit the AOL TOC Servers. Therefore
> >although the readme states the module is beta, it has been taken from
> >Net::IRC which has gone through many revisions and bug fixes. And this
> >module itself hasn't seen many bugs. (I uploaded a working version)
> >
> >I do not know how many people have downloaded the module and currently use
> >it, but I do get correspondence lately quite a few times every other week
> >about the module, which seems to mean that people are using it and
> >working with it.
>
> Obviously, then, this module is being used and is not outdated. Thank
> you for making this clear.
>
> >Anyway the module was put together for the TOC server, if your
> >module uses the real AIM protocol which I started coding but never
> >finished (TOC was the easy solution) you can make another connection
> >class and integrate this into the module - which I was intending to
> >do at some point, so you can choose which protocol to connect with.
>
> Mine is actually also an implementation of the TOC protocol. Since
> AOL is not allowing MS et al. to create clients that use the 'real'
> protocol, I felt that any module based on it would be quite
> unstable--it seems that AOL has no problem breaking everyone's client
> by making small changes to the protocol that do not effect their own
> code base, as they've done just this many times.
>
> >I don't know how much time you spent on you AIM module but I think it might
> >have been more productive if you would have followed the instructions on
> >CPAN concerning creating modules.
> >
> >I really hate to come off sounding like a bitter ass but I'm insulted that
> >someone wants to take over a namespace because they went and started a
> >project before simply reading the instructions on cpan.org or checking for
> >its existence.
>
> I did check for the existence of similar modules and found that the
> existing modules were unsatisfactory (or, at least, their
> documentation suggested this to me when I began my project). That I
> decided to name mine Net::AIM is somewhat accidental. I began by
> calling it Net::InstMsg, which was, as you can see in the archives of
> this list, objected to (and rightly so) on the grounds that instant
> messaging is not limited to AIM. Forgetting that Net::AIM actually
> existed, I followed a suggestion originating on this list and renamed
> my module to Net::AIM. I readily admit that it was pure sloppiness on
> my part that I did not first check to make sure that no such module
> existed. I relied on my memory of the names of the modules I had
> examined; it obviously failed me.
>
> I'm sorry that you feel insulted. I meant you no offense in naming my
> module Net::AIM. In fact, with one exception, I did follow the
> protocol suggested on cpan.org concerning the creation of modules.
> The oversight I committed, not making a final namespace check before
> deciding on a name, was unfortunately a grevious one.
>
> >This is of course regardless of the performance of the 2
> >modules. Maybe I should have made the version 3.02 ;). I didn't slap
> >something together. I worked with modules that were already written and
> >tried to design a module that would be able to grow properly (ie the
> >different connection classes).
>
> I feel that my module was also carefully designed to make instant
> messaging easy for the module user. The fact that my module does not
> rely on previous work does not in any way indicate, in my mind, that
> it was, as you say, 'slapped together.' The final project is, to my
> eyes, a fairly elegant solution, managing permit and deny lists, buddy
> lists, and the like, as well as giving the module user complete
> control if necessary while still maintaining the ability to abstract
> away all socket communication if the module user wishes to do so. I
> feel that the module is a useful contribution to CPAN, and it is
> certainly one that I will be actively using and updating.
>
> I do realize that some of my actions were wrong, and I apologize for
> the trouble; I will be renaming my module to Net::AOLIM, so the
> namespace collision issue will be resolved within the hour.
>
> Riad Wahby
> [EMAIL PROTECTED]
> [EMAIL PROTECTED]
>
> 5105
>
> -----BEGIN PGP SIGNATURE-----
> Version: PGPfreeware 5.0i for non-commercial use
> Charset: noconv
>
> iQA/AwUBOXEz+4h6K+kZMcamEQJGcwCgiUVEoA8wvi9CpIuXgcstTJDfkn8AmgNQ
> q71ZyQ0PgvB5bewzs194ZxE1
> =wIiX
> -----END PGP SIGNATURE-----
>
