[icq-devel] Protocol Databases (was Re:Offline messages info)



Rüdiger Kuhlmann icq-list-oWUpAQCWC2r4 at ruediger-kuhlmann.de
Mon Dec 2 23:09:55 CET 2002


>--[Zoe Smale]--<skyerat at hotmail.com>

> > There are packets that have links to other packets; those links are
> > simply HTML, which can be put into the comment field (most fields don't
> > allow HTML, but the comment field and the packet description does).
> That's better than nothing, but it's not a particularly feasible solution.

It is.

> Consider the example I mentioned. The user data structure has more than a
> dozen items of information it and it's shared between 7 different SNACs

The correct thing to do would be to create a packet in the generic section
for it, and link it from everywhere with appropriate comment. No structural
changes necessary.

For the specific thing, I'm not sure which data structure you actually mean.
See, a mailing list posting can be horribly unprecise. Like yours.

> However, can you really imagine someone adding a bunch of HTML comments to
> try and link those all together? Can you imagine anyone that wanted to
> update one of those fields following the links to all the other places it
> occurs to update it there as well?

You don't need to link all together. You need links from the packets to the
structure, and maybe some links back from there to the most important ones.

> time to try and fix that mess when they can just post a simple explanation
> of the structure to the mailing list with a list of the SNACs where it is
> used.

Cool - and how is that cross-related? It isn't as well. It's just a list,
buried somewhere in the list archive. Have fun finding it when you actually
need it.

> What I'm trying to get at here is that it should be the responsibility of
> the database maintainers to add the data to their databases.

I don't have any responsibility at all.

What you're trying to get at here is trying to make your life as easy as
possible.

> If that means restructuring, or cross linking or adding new tutorials then
> that's what they should do (and that's something that nobody else can do).

There's no need to restructure here; cross linking and new tutorials can be
created by everyone (though the latter cannot be added to the web site by
everyone).

> Let the people researching the information post it to the mailing list or
> whatever is most convenient for them.

Look, I do try to gather all information into one place. It just would be
better if everybody would add it to databse, or fix the oh-so-many errors in
it that everybody claims there are without being specific (or simply fixing
them).

> Everyone is immediately aware of the new info,

Every member of the list, maybe. Considering the emails I get from people
grateful for the database, I can tell you that the information is not
readible available on the mailing list for everyone, in particular if you
haven't been subscribed since the beginning and have it in your own mailbox.

> anyone else can comment on it,

Sure. _That_'s what the mailing list is for.

> and they can leave the problem of getting the info into the database
> up to you.

Which means: dump it on the mailing list and leave the work to keep
everthing in one place to someone else. Y'know, I have my own client to fry,
creating the infrastructure is already some work, and there are already
enough places where to hunt for information.

-- 
         100 DM =  51  € 13 ¢.
         100  € = 195 DM 58 pf.
  mailto:ruediger at ruediger-kuhlmann.de
    http://www.ruediger-kuhlmann.de/



More information about the icq-devel mailing list