[fpc-devel] Offer to repair and maintain the FPC community website (repeat msg, no HTML)

Cephas Atheos cephasatheos at gmail.com
Thu Sep 27 21:04:04 CEST 2012


Guys, I'm sorry if this has re-opened old memories of the "NNTP wars" and
the "Great Listserv Battle", and caused flashbacks and
"post-http-stress-disorder"! :)

That wasn't my intention, I swear!

Of course, it would be naïve of me to think that any urge to fix and
update (or "change", as it's called :) things wouldn't cause some conflictŠ

Having said that, if I can inject my own experiencesŠ I remember long
nights with my mates running uucp and 2400 baud modems and writing connect
scripts and resetting ttys frantically trying to recycle serial portsŠ And
that was leading-edge technology, oh my!
That was so much fun, and it was "only" a hobby, I had to go and do paid
work the next day.
The one thing I do know for sure is that technology has got to the point
now where it isn't just one thing that will be what we (and by "we" I mean
you guys, and eventually, hopefully, me as well) will use to do what we
need to with fpc and lazarus.

So I think there's a good chance that we'll use listserv technology to do
what we need to communicate to people, and a choice of nntp and http
clients for anyone who's more comfortable that way.

We don't need to go completely one way OR the other, and most modern forum
software acknowledges that. Bottom line, we call sendmail() to notify,
isamdb() (or whatever) for storage, and probably http for ui and
management. My hope is that we get to meet and maybe even exceed people's
expectations for help with problems, looking after the data in the
archives and making that available to anyone who needs it, regardless of
the interface we provide.

But, in reality, people these days find listserv difficult to understand,
set up, and keep on top of. Especially when messages are more easily and
safely stored where they'll do the most good - in a globally-accessible
searchable relational database (which is still pretty old technology -
maybe not smoking jacket-and-pipe old, but it's been around for a while).
You can still download and store the data you need if you still want to do
that.

The same goes (I'm sorry to say) for nntp servers and clients. Just saying
clients and servers takes me back to the good ol' days of users
respectfully requesting bits of information from the Acolytes and
Maintainers of the ArchiveŠ There are better and far more easily and
securely available protocols that no-one has to install clients for, or
configure passwords and ports and ssl hashes, or try and come up with a
hierarchical structure that makes sense to them, and only themŠ

The bottom line is, what we're used to (and I still have my CP/M and
Borland news server lists updated every week, trust me!) isn't necessarily
what new users are going to expect. They're not even going to ask for that
information - they're going to expect it to be provided for them, in a
format that is secure, infinitely searchable, fast, nice-looking (that
always helps) and that they don't have to download, install, configure,
test, and maintain. That's most definitely what we're used to, I agree,
but it's not how things work these days, except for a valiant few. (That's
us).

One thing I've learned from these interesting discussions is that whatever
form the site morphs into, it will need to be able to at least allow a
couple of different ways of getting that information. If that means we
have to build an nntp servlet interface for people most comfortable with
nntp clients, let's at least try to include that as part of the service. I
don't know of any modern server software that provides multiple protocol
access to the same datastore, but you guys know more about that than I do!
If it's possible, why not build it in? Then at least you get to compare
access types and see what really works for you.

I'm truly not just a forum-promoting hyper-unrealistic change monkey. I'm
too old and way too tired for that these days, trust me! :)

But the thing that's drawn me in to this whole situation, is that the way
it works now *can* be improved (hopefully without degrading any part of
the community interaction), and by improving it, we make it easier and
"funner" for new users of fpc to come onboard and ask, contribute, whine,
and discuss. I suspect that's a good thing, but it's not up to me, I'm
just the spanner in the works!

I sound like a real estate developer. Kill me now, please.
-Pete















More information about the fpc-devel mailing list