Re: Bits 'o business -Reply

[Robert Holmgren holmgren@xxxxxxxx]

On 27 Jan 97, Bob Brody nearly lost it:

> I really can't stand DOS programmers and their sloppy ignorance
> for documenting while they sling their bug infested slop out
> into a world where computer science is taken seriously even by
> people who abide the misfortune of having to use a PeeCee.

Whoa, get a grip on, Bob. Gee. As for the "bug infested slop", you were
warned. XyWin is virtually a beta release. There was one minor bug fix.
(The OS doesn't exactly help, either.) Whereas Xy4DOS went through ten
sunstantial fixes (v1.00, 1.01, 1.02, 4.011 through 4.017), and lots and
lots of incremental builds (e.g. 4.015G, 4.017B). Moreover, by the time
v4.017 was finished, they'd retrofitted Xy4DOS with almost all the new
features that they'd subsequently invented for XyWin, i.e. which had not
been part of the original Xy4DOS product (Signature). That's why the .DLG
and .MNU files for XyWin and Xy4DOS are 100% interchangeable. And that's
why there are a lot of Xy4DOS users here, versus few XyWin users. (Well,
maybe affection for text-mode has something to do with it too...)

Secondly, don't blame the programmers (who shouldn't be documenting anyway
-- sheer waste of their productivity). The truth is, docs exist, and
they're quite thorough -- you just didn't receive them in the XyWin
distribution. An appalling omission. XyQuest always managed to publish
documents and supply human help; TTG is ... indescribable.

I mentioned a couple of months ago (when I posted DIALOG.XW4 for you) that
I have a bunch of text files obtained here & there. Here's one of the best
ones, at least for a XyWinner. A bit of preamble: Xy4DOS's original
release in 12/91 was a very messy event. The product (Sig) preceded some
of the manuals by about four months or so. And that wasn't all. XyQuest
promoted the product as something radically new, operated by menus; I
vividly recall the first few Emails from the SysOp at the BBS, which didn't
even acknowledge that Sig could be customized, or made to operate almost
like Xy3. Took months of tooth-pulling to extract the facts, and always in
bits & pieces -- never a coherent view.

At some point Christine Madsen (or Steve Crossman?), in response to really
LOUD howls & protests, posted to me a file that describes, in detail, how
the Menu system is constructed (HLP and MNU; DLG U2 etc probably aren't
discussed, because they hadn't been invented yet; but the frame's
the same, only the names change).

Caution! This is an early, almost prehistoric explanation. There have
been elaborations (but no alterations, really) to the way Help works. The
doc may contain obsolete info here & there (haven't looked at this in a
while); but 99.9% of Signature is still alive in XyWin, so this should
supply much of what is (unaccountably) missing from the XyWin manual.

You're truly the first person in a very long while (or the first to talk
about it) who has tried to extend the menus per se. Takes real nerve! I
stayed as far away as I could. Good luck.

Afterthought: This is a big file, so rather than burden the mailbox, point
your browser at:

 http://chelsea.ios.com/~holmgren/temp.theory.mnu.shtml

I'll take this page down after a few days. Hope it's useful. (Although
it's searchable as a file, Xy4DOS users -- who have this substance, and
much more, in print form -- probably won't find it earthshaking or even
illuminating.) Note that the programming and frames don't actually "work";
the text was set up for illustrative purposes only, apparently.

-----------
Robert Holmgren
holmgren@xxxxxxxx
-----------