Re: Re Word/Weird

[Chris Kowalchuk <chris -at- bdk -dot- net>]

Re: Rebecca Rachmany's post

I have heard this type of argument before, and it is valid in its way:
if you jump through all the hoops required to get performance out of the
tool, you will get the performance. If you have a large team, you can
assign one or two hoop-jumpers to make everyone else productive. If you
are a single-person outfit, or if you are structured such that your
writing is not properly supported so that you are effectively a single
person outfit, then it is hard to both make a living and devote the
required resources to fixing the tool.

But consider Bill Hall's main argument: that you spend too much time
fooling with the program, and not enough time using it. So, instead of
wasting time fixing crashes and so forth, Rebecca's team "subscribed to
newsgroups, [learned] to use the macro language, wrote macros which
overcome most of the major irritations..."

The major irritations list just happens to be everything anyone wants
formatting software for: numbering, cross-references, headers and
footers." If I don't want those things, I can use a bloody text editor
in DOS, or CPM for that matter.

Now that's great that Rebecca's team solved their problems, and did all
this work, but if I recall the marketing correctly, we were not sold "an
ongoing experiment which, with sufficient custom re-programming, should
work most of the time"; we were sold a "solution" that was supposed to
work out of the box.

Nowhere in the documentation that ships with the product does one read:

"to automatically number pages: subscribe to a Word users list, order a
third-party book on programming Visual Basic, design a workaround...:"

To follow the argument that if you rewrite the code, the product works,
I don't know why you wouldn't just stick with shareware and Unix or
Linux. It's cheaper, just as much work, and you get that same glow of
satisfaction from having written successful code to solve major problems
you were experiencing (the ones listed as "features" on the outside of
the box).

The trouble is that most of the people who make the power purchasing
decisions are impressed by Microsoft's marketing claims and market
position. "We make the operating system, and of course, we are best
positioned to make tools that integrate with it seamlessly: pick it up,
plug it in, and your whole organization is off to the races." Sounds
great. Most managers never have to write anything more complex than a
memo or a letter, so Word never fails for them.

As for the "other products are just as bad" argument. Yes and no. Frame
and Word are apples and oranges. I don't care how well Word ever works,
I would argue that, optimally, you don't use those two programs for the
same purposes. To compare apples to apples, the bug disaster that was
WordPerfect 5.2 for Windows virtually killed that product forever. Too
bad, because with each release, the owners of WordPerfect eventually fix
the bugs. Word carries them from version to version. There is actually a
good structural reason for this. WP nests all it's formatting commands
and graphics in the text, as did the word processors of old. It executes
instructions linearly, as it encounters them, and has clear "begin" and
"end" markers for each instruction.  That is why it lets you edit those
codes (because it can). Essentially, it works similarly to the "tag"
system used by programs like Frame. Word, on the other hand, apparently
stores all its formatting information and graphics in a single hidden
location at the end of the document, so that it must constantly
"remember" where everything is for the whole document at once. No wonder
the styles and graphics placements etc. get flaky on big documents.
Having a powerful computer helps, but the essential problem can't go
away. I have battled with both these programs for years, and in the end,
I was always able to us WP more reliably, although it often shipped with
worse up-front problems. One of the advantages that I truly appreciated,
however, was that the WP people typically admitted their mistakes, and
would actually indicate in the manuals what the program could not do,
and suggest workarounds. I have never seen anything like this in an MS
manual. If their product doesn't do something, they never mention it.
All you get is more hype. I think that is part of what infuriates people
trying to use the program: they feel they have a right to believe the
marketing; that they paid their money for what they were _told_ they
were going to get. As has been noted elsewhere, in few other industries
would these low standards be acceptable, and they won't be acceptable in
this one forever either. And yes we do need a standard mark-up language
like XML, and no Microsoft should not be allowed to keep redefining it.

So there.

Chris Kowalchuk