Re: experts

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

At 09:57 AM 3/29/01 +1000, Marcus Carr wrote:
>The fact that they are experts doesn't mean that
>they have somehow evolved past what made them good developers in the first 
>place.

That's precisely what I'm claiming it does, perhaps especially when they 
are gathered into committees.

I'll also claim that most experts don't realize it happens, and will deny it.

> > To over-generalize, here's what the problem looks like:
>
>[Points about defining, optimising and combining layers deleted.]
>
>One could over-generalise in the other direction as well. In many 
>circumstances,
>the inexperienced are prone to over-complicate a system, whereas someone 
>who really
>understands the issues has the confidence to implement what might look 
>light an
>embarassingly simple solution.

Yes, given a pile of tools, the inexperienced sometimes try to use them 
all.  Or, like me in my woodworking shop, they try to buy a tool to solve 
each and every problem.  (My tools to talent ratio is pretty awful right now.)

However, that hasn't been my experience.  I've seen some grotesque hacks 
when developers use the tools they know to solve inappropriate problems, 
and some ugly pileups as requirements for projects changed and code reuse 
rather than refactoring was the rule.  More typically, though, I've seen 
people work through small projects on their way to bigger ones, learning to 
use tools only when they need them.

Individual experts do produce far simpler things than inexperienced users 
on a fairly regular basis.  I just don't see this happening in the 
specifications field.

>Given the unquestionable complexity of the XML
>suite, should we infer that the creators were in fact not expert enough? 
>Probably
>not, and besides, what real value is there in this sort of deconstruction 
>based on
>hindsight? It won't even assist in deciding who should participate in 
>future WGs
>unless one of us can produce some concrete evidence supporting our position.

No, but it might suggest to more 'ordinary' developers that 
specification-building isn't work reserved for Olympians.  By pretending 
that such tasks are purely the work of annointed institutions and 
individuals, we continue to tell people that they're only consumers.

I don't expect to change who gets on WGs in the short term.  I very much 
hope to change who participates in standards building over the longer 
term.  Perhaps the greatest achievement of XML is that it makes it possible 
- even easy - for a much wider group of developers to participate in 
"standards creation" on a wide variety of levels.  To me, that's really 
cool.  I'd hate to bury that potential under a landslide of features, even 
if they come with an "experts' seal of approval."


Simon St.Laurent - Associate Editor, O'Reilly and Associates
XML Elements of Style / XML: A Primer, 2nd Ed.
XHTML: Migrating Toward XML
http://www.simonstl.com - XML essays and books