Tools (was Re: Fwd: [xml-dev] Not using mixed content? Then don'tuse XML

["Simon St.Laurent" <simonstl@simonstl.com>]

On 4/8/13 4:53 PM, Jim Melton wrote:
> Your statement is very much akin to my stating that all use of hammers
> is actively harmful.
>
> The use of hammers when, say, polishing glassware is undoubtedly
> harmful.  Their use when building a house with a wood frame is
> indisputably good, even though a pneumatic nail gun might be better at
> times.

Hammers are not a great metaphor for this.  Using a hammer barely 
changes the way you look at the world, despite the popular saying that 
"to a man with only a hammer, everything looks like a nail."

If you want to talk about tools, I can talk about stepping away from my 
table saw.  It's not especially that the table saw was dangerous, but 
rather that in addition to its obvious power as a rough tool and a 
rarely successful temptation for use as a fine tool, it changed the way 
I looked at projects.

Planning projects around a table saw shifts the way you plan your 
projects.  Straight lines suddenly become much more appealing, as do 
certain widths of wood.  There are genuinely inventive people out there 
who cut curves (coves) in wood with a table saw, but I'm reasonably 
certain that manufacturers and the occasional publishers of articles 
promoting such use cringe at the prospect of a serious mistake.

(Please do not go out and try to cut coves with a table saw.)

Schemas are much more like table saws than they are like hammers.  A 
hammer will help you build a project, but it will rarely determine the 
way you build it.  The feeling of smooth power over materials (and saved 
time!) often disguises the impact the tool has on the shape of your work.

> Schemata (of whatever sort) have great value in some environments, and
> cause many problems in other environments.  A categorical dismissal of
> the value of schemata under any and all circumstances betrays either
> arrogance, intransigence, or inexperience.

Categorical dismissals are rarely popular with the people whose tools 
are being dismissed.  Unfortunately they are sometimes necessary.  And 
yes, dismissals frequently lead to the adjectives you just deployed.  We 
can revisit this conversation in a decade and decide whether that was 
appropriate.

> I've used XML schemata (DTDs and XSDs) for many years and they have
> served me very well.  Their use has helped me avoid many traps in data
> that I generate and in data that I accept from others, and they have
> amply paid for their use repeatedly.

Again, I don't expect people immersed in markup culture to immediately 
respond with "Oh, yeah. Of course.  I should have been doing this 
differently all along."  Thirty years of tradition grounded in a couple 
of centuries of information practice before that is difficult to discard.

> Your experience is different.  That's fine.  I would never claim that
> "any situation in which the use of schemata is avoided is actively
> harmful", because it's obvious that is false.  It is also obviously
> false that use of schemata is always harmful.  My personal experience is
> proof of that -- I have benefitted from using schemata and I have
> encountered absolutely no problems as a result.

This may be a question of perspective, but it is not a question of 
experience.  It's about which questions you deem worth asking.

I know that people hate to be diagnosed with problems when they are 
quite certain that everything is fine, but XML has already limited 
itself severely by spending so much energy on this particular path.  We 
just choose to focus on our successes and mutter about those strange 
people who chose other paths.

Thanks,
-- 
Simon St.Laurent
http://simonstl.com/