[Mot-general] MOT development

[Phillip S. <te...@al...>]

Greetings.

I have been using MOT and following its development since it was
released back in early 1998.  For me, MOT filled a niche that no other
software, commercial or otherwise, came close to at the time.  It still
does, and I still use it today because nothing else has stepped up to the
plate (although there is something very similar now.. read on below).

Over the last few years, I've traded E-mails with Hal Roberts and
discussed the possibility of taking over the project.  Back in March 2000,
Hal indicated to me that he didn't have as much time to work on MOT, and
probably wouldn't be developing on it anytime in the future.

Given the state of the project as is evident on SourceForge, the
project appears dead.  The most recent postings in the forums are by
me, where I went through and answered the unanswered questions under
Help.  I haven't seen any messages on the MOT list in a long time..

Anyway, this E-mail serves a number of purposes.  The primary purpose
is to get an idea for how many people are still using MOT, or have an
interest in what it does (what are you using now?).  The other is to get
feedback on how MOT was/is being used (specific real-world examples),
collect any existing patches/code enhancements from would-be contributors,
build up a new feature list, and finally, to find out how many people
would be interested in helping me revive the project.

I've been looking around at MOT alternatives to get a feel for whether I
should proceed with a massive rewrite.  I came across something called
IRM, short for inventory resource manager
(http://sourceforge.net/projects/irm/), which comes pretty darn
close to what MOT does, except that it's more specific/non-generic than
MOT.  It has all the functionality and more of my current MOT setup as is
used for project tracking and hardware/software inventory.  I like a lot
of what IRM does, but I still like the flexibility potential that MOT's
idea encompasses.  None of these do any LDAP auth/integration yet, and I
want something that can easily export to CSV, XML, etc. (MOT does some of
this currently).  Also, hooks/imports/exports for PDAs (such as WAP/WML
type stuff) would be useful.  IRM already has a knowledge base,
reports generation, an interesting templates system for faster entries,
and other cool things I wish MOT had.

IRM is under heavy development it looks like, which is promising.  My main
concern in all of this is the code base.. I'm really interested in having
DBI abstraction, as well as HTML abstraction using templates and OO
classes.  I would like to have something more like MOT that can be easily
extended in a modular way to do other things (i.e you grab a pre-definied
XML file and import it.. boom.. you have some new useful tool).  But
certainly, MOT's existing job tracker schema could benefit from the types
of things you can do now in IRM.

Given that both MOT and IRM are PHP/MySQL based, I could probably use
ideas/concepts from both and merge them into a new framework.  Basically,
the code base I want to create would use a DBI abstraction
layer, like Metabase or ADODB, and a HTML/code abstraction template system
such as Smarty.  Ideally, the framework to use would be BinaryCloud
(http://www.binarycloud.com/) which already incorporates these things and
more, and is very powerful.

I have been in contact with the BinaryCloud developers, and they seem to
think MOT could also be used as a dynamic front-end to BinaryCloud,
capable of tying into and controlling their Entity Managers.  This would
make for a *very* powerful content management system, for instance.  This
is but one avenue development could take or branch into.

The immediate goal would be to fix any remaining problems with the current
MOT, and release a non-beta.  Then, concentrate soley on a rewrite.

So, to all MOT users, give me some feedback on how you're using or were
using MOT, and what would make it better for your specific environments
(please don't send bug reports, or other help requests directly to me
right now.. use the tools on http://sourceforge.net/projects/mot/ to
report those kinds of problems).

To any potential MOT contributors and/or developers, I would like a sanity
check on what I'm proposing, and to find out how many people would be with
me in this endeavor.  I envision that if I alone took on this project, it
would take 6-8 months to complete an initial rewrite (possibly longer, as
I'm a fulltime sysadmin, part time consultant, and will be moving to a
new job in the near future).  I really do not want to put much effort
into the existing code base (note again that I value the ideas/concepts
MOT has, just not how they're currently implemented).


-phillip