Lists Home |
Date Index |
On Wed, 2002-03-27 at 16:09, Bullard, Claude L (Len) wrote:
> "How I wish, How I wish you were here..."
We're just two lost souls swimming in a fishbowl, year after year.
Or I am, at least. I work for a fairly well-know tool vendor, so some
of the questions aren't entirely apropos. (You haven't heard of
Talsever? *laugh* Talsever is my domain, so that I can make rude,
raspberry like noises on public mailing lists without reflecting badly
on my employer).
> Better late than never. If RNG is a dark horse:
> 1. How have you incorporated it into your workflow
> (eg, tools?)
> 2. Was that a reasonable thing for a production manager,
> web author, etc. to do?
> 3. Are you pleased enough with it that you would support
> a tool vendor that supported it over one that didn't?
As noted, we *are* a tool vendor. There's currently significant debate
about whether to add support for RNG; the problem is that RNG allows
some things that other schema models do *not*. So some of the
fundamental particles used to model DTD, Schema, SOX, XDR, whatever,
would need revision. Is RNG worth the effort? I make the case
(internally) that it is, but I'm not too concerned about it.
One of the great advantages of RNG (for personal-project workflow) is
that one doesn't *need* tools. It is, in my opinion, easier to work
with than either XSDL or DTD, particularly in that one can write a
"sample instance" and then start decorating and reworking. The
impression that I've had is that it is also easier for folks to
understand when they are ambushed with the finished product (that is,
it's easy to understand a schema model, no matter what language, if
you've spent months designing and refining and arguing and nuancing ...
harder to comprehend, and very dependent upon presentation (meaning
language) when faced with nothing but the finished draft).
> 4. Given alternatives, which do you use most and which
> would you recommend to your successor were you to be promoted
> to their manager? (IOW, you have to live with your choices.)
RNG hasn't been an alternative until quite recently, and I haven't had
to kick off initiatives for schema design in that time. When it happens
again (I've done two such mini-languages within the past year and a
half, I think), then I'll opt for RNG, no question.
In private code (that's so seldom worked on as to be unlikely to ever
see the light of day, although it's "open source" in concept), I've
started using it, both to learn it through harsh experience, and
because, frankly, the experience is nicely un-harsh.
> It doesn't matter as much who owns or creates the specification
> as it does that we get undeniable value from it and that
> we are willing to demand it from our tool vendors by
> voting with our feet or our purchases.
Get one of the vendors to support it. I don't think that there's been a
lot of demand (not for us, anyway, although our profile is a good deal
lower in the last year or so, I think).
Amelia A. Lewis firstname.lastname@example.org email@example.com
You like the taste of danger, it shines like sugar on your lips,
and you like to stand in the line of fire
just to show you can shoot straight from your hip.
There must be a 1000 things you would die for;
I can hardly think of two.
-- Emily Saliers
This is a digitally signed message part