xml-dev - Re: XML Editors - Word 2000??

[Ketil Z Malde <ketil@ii.uib.no>]

Marcus Carr <mrc@allette.com.au> writes:

> Ketil Z Malde wrote:

>>> People who are drafting XML documents such as legislation are not
>>> working with the structure,

>> So they don't know about paragraphs, or references I take it?

> Wow, you are grumpy. 

Told you so, didn't I?

> Do I really need to rephrase that sentence?

Perhaps I misunderstood.  What are they working with, and why is a
WYSIWYG environment focusing on fonts and colors, better than one
displaying the document structure?

> It's not a matter of intelligence, it's a matter of focus. If you're paying
> a lot of money to obtain someone's expertise about the law, it's not
> practical (and may even be financially foolish) to pay them to fiddle around
> with tags. 

I maintain it is much better than pretending the structure doesn't
exist by covering it under formatting issues.

> What would you say to the legal expert who said that anyone smart
> enough to understand XML tagging should be capable of writing
> legislation?

I would say that while I agree that the world would be a better place
if lawyers were replaced by those understanding XML, writing of
legislation is a highly specific, complicated process, while tags are
a really simple concept.

> Why do we assume that just because we have a need for XML data, all sorts of
> different professions are going to accommodate us at their own expence?

Because you want the XML data to be meaningful.

If you want meaningful data, you need to let users see, work with, and 
focus on structure.  WYSIWYG doesn't give you that.

> I'm not saying that it's the right way to go in all cases, just that I'm
> sick of having people ask if they can author in Word and me telling them
> they have to use MultiEdit or Emacs.
> They don't want to hear it and I don't want to say it. 

Well, don't tell them then.  Give them an XML format with <b>, <i> and 
<u> tags hooked up, congratulate them on being XML-aware, and receive
a pat on the back.

> There is a strong correlation between what a document looks like on
> paper and what it is - as paper publishing is still inevitable in
> some industries, they are often nearly the same thing.

I'd say the most important lesson from SGML and XML is that they are
*not* the same thing. 

I've on occassion done some support for people using word processors
for writing HTML pages.  They tend to be rather annoyed if I complain
about various issues, from mulitmegabyte bitmaps to using obscure
fonts and extreme hard coded table sizes to downright illegal HTML.
The response is always that it looks all right on their screen, so it
must be my software that is doing something wrong. 

> Wunderbar! 

Although I get a reading on my sarcasm-o-matic, I must say I'm pretty
happy with it, yes.

> And you get valid HTML out the other side, you say? 

Of course, what else would be the point?

> Have you tried modifying it for DocBook?

I have written DocBook documents with it, yes, although I haven't
bothered to modify it for that purpose.  Why would I want to do that - 
except adding an initial template document, which is rather trivial?

(Thinking about it, I guess I could modify it to display e.g. heading
contents in a larger font, or italicize emphasized elements the way it
does for HTML, but I haven't really bothered, since a) it's not really
a big deal, and b) I don't use DocBook on a regular basis, and anyway,
I bet somebody, somewhere already have done that)

> I find it interesting that you are looking for an application that will help
> you create good data, but apparently begrudge others the same thing.

No, that's not what I'm saying at all.  I'm saying that pretending the 
structure is automatic and derived from formatting information is
bogus, void and in error.  If you want "good data", you need to
educate users about structure.  Yes, I know this is hard and
unrealistic.  It is still the right thing.

> Your preferred interface involves the use of raw tags; 

You don't *have* to display tags as characters, of course, but most
editors I've looked at, only puts some kind of fancy box or similar
instead.  No real improvement, IMHO, but whatever floats your boat.

> It doesn't have to be ugly to create correct data.

Of course not.

> They always understand what you're trying to achieve, they just want
> it to be easy for them to use.

Yes.  But my knee-jerk reaction was to the idea that word processors
- in particular MS Word, but the point would be the same for any
layout - and formatting-oriented interface - are ideal for producing
XML.  I maintain that they are too complex to be easy to use, and
their complexity drawing focusing away from what's important if
structure is what you want.

Look, Word can't even enforce the use of styles, I've waded in
documents where somebody took a lot of effort to set up the .dot
files, only to find users are just inserting formatting information
instead.  The point is that Word makes it easy to do just that, and
that is precisely the kind of thing you want to avoid.

>> Of *course* you should require validity.

> You're mixing up what I'm saying - well-formedness doesn't equate to format
> oriented documents and the subject still says "XML Editors".

But well-formedenss is kind of easy to get away with - in any format -
even if many tools for e.g. HTML authoring doesn't even manage that.
Personally, I think DTD-less XML document design is an abomination
(although I see the point of being able to parse documents without
access to the DTD)

> I'm saying that validity allows you to guide the user by restricting
> the available elements whereas well-formed doesn't.

Yes, and that guiding is a good reason to use XML.  Much better than
buzzword compliance, for instance.

-kzm
-- 
If I haven't seen further, it is by standing in the footprints of giants

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo@ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo@ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@ic.ac.uk)