RE: alternative to WebFacing -- MIDRANGE-L

> But the funny thing is that lately I've been told, quite frequently
> actually, that I rush in to things way, way too fast...  And I know there
> are a lot of folks, such as yourself, who have a h*lluva lot more
> experience
> at these things than I do.
>
> So, for the time being anyway, I'll wait and see who (if anyone)
> is willing
> to step up to the plate, and in what capacity.
>
> Now if I don't hear anything by midnight tonight...:-)

JT, over the years I've tried to cobble together a couple of efforts like
this.  They haven't been terribly successful.  For example, the free code I
give away on zappie.net (and its successor, www.java400.net) had only one
other contributor, Alex Garrison.  And while people are often (and I do mean
OFTEN) willing to help one another in remedying a specific problem, when it
comes to the amount of work required to actually develop some sort of open
source project for the AS/400, well, few of us have the time.  Most of us
are probably working more hours than we care to, with less job security than
we've had in a long time.  To attempt to start what would in essence be a
"side job" is asking a lot.

It's not that the idea is without merit, you understand.  It's just that
it's a whole lot of work for not a lot of perceived benefit, and with a
major stumbling block to boot.  I think in order to even attempt something
like this, you'd need one of the following:

1. A specific project.  Not simply a goal, but a well-defined target with
specific requirements and restraints.  Consensus here is not easy to
achieve.  For example, you have about five different factions here, some of
which have vested participatory interest.  Brad has e-RPG, Nathan has his
HTTP plug-in, I have revitalization, and I suspect that SPECIAL files, MI
hooks and some sort of TCP/IP redirection would each have their adherents as
well.  So, your first issue would be to decide which of those architectures
are the one you want to pursue.  You can guess that if I were to participate
at this level, it would probably only be to continue with the revitalization
approach, and I doubt that Nathan or Brad would be interested in helping.
Multiple agendas are unlikely to achieve results, unless you were able to
create:

2. An umbrella project.  This is the situation where the various ideas are
hashed out and reviewed.  Areas of commonality are identified, and
interfaces are designed to allow each of the various techniques to
interoperate.  Teams (a very loose term, as a team could conceivably be a
single individual) would choose to implement one or more of the various
components.  As they are designed, each would then go to a testing
environment.  This is the most democratic and the most productive
environment for long-term development, but it has some severe drawbacks.
For our community, the most critical problem would be that it splinters the
development effort.  We don't have a lot of spare hours as it is; focusing
in several directions dilutes those hours even further.  Another potential
problem is that you still would need an architectural overseer for the high
level infrastructure.  This sort of design actually cannot be done by
committee.  Ideas can be floated, but one individual, or at most a team of
two or three, would have to be given the full oversight of the umbrella
interface design.  Anybody working on the project would have to work within
those guidelines.  That would be quite a tall order.

IMHO, it's not the technical difficulties.  I'm pretty close to a functional
subset of a full technical solution, written entirely by me, in my copious
free time.  It can be done.  No, the far more difficult question for an open
source project is whether the community is willing to relinquish control in
any degree to an overseeing agency and work within that framework, even if
it might not directly promote their own personal agenda.  This group?  Um.
Well.  Read some of the posts over the past months and you may come to some
opinion as to whether that would work.

And to those who disagree with my assertion that software projects of any
magnitude require an autocratic rather than democratic model, please
remember that the most successful of the "open source" projects, namely
Linux, was primarily run by a single individual.  Only because people were
willing to contribute work with the understanding that the final, inarguable
decisions were his, was the Linux project able to work.  Closer to home, the
most successful AS/400 software company was System Software Associates, and
it was only successful when run by a single individual.  When that person
left, and management by committee began, the company folded.

IMHO.

+---
| This is the Midrange System Mailing List!
| To submit a new message, send your mail to MIDRANGE-L@midrange.com.
| To subscribe to this list send email to MIDRANGE-L-SUB@midrange.com.
| To unsubscribe from this list send email to MIDRANGE-L-UNSUB@midrange.com.
| Questions should be directed to the list owner/operator: david@midrange.com
+---