[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

RE: The myth of 80/20

• To: 'Robert Koberg' <rob@k...>
• Subject: RE: The myth of 80/20
• From: "Bullard, Claude L (Len)" <clbullar@i...>
• Date: Thu, 4 Mar 2004 13:02:22 -0600
• Cc: xml-dev@l...

And the paradox is resolved by understanding 
that it is not a measurement (data) but a function 
that is being evaluated. The myth of unnecessary 
granularity.

If we do the least possible, just for argument's 
sake, would a vector graphics language need 
any other shape elements beyond path and glyph?
Should we have a framework that only manipulates 
the string values for vectors, or should we have 
collections with indexed access?

Well, more and both.  One aspect of spec writing 
and features is to be very clear about what level 
of the system/subsystem one is specing.  We always 
seem to have a devil of a time getting people with 
different backgrounds to understand that an XML 
language spec in and of itself doesn't provide 
much in the way of interoperability without a 
runtime spec to go with it.

Will we spec XML languages for XAML primitives?

The browser created opportunities for competitors. 
It also limited them because the boundary between 
the browser and the operating system isn't real. 
It is politically convenient but not necessarily 
technically advantageous.  It created a myth of 
superior engineering that was and is a fraud but 
it made such a fine free condiment that lots of 
people came to the party for the freebies.  Now 
it is just a nuisance because like using free 
food to open a bar, once one has a steady and 
very large crowd, it is unnecessary overhead but 
the crowd believes they have a right to it. 

One reason a spec goes out of control is the 
politics of consensus (who understands what 
when and how loud or soft are they).  80/20 
is a crowbar and a cudgel when that becomes 
politically difficult.

len


From: Robert Koberg [mailto:rob@k...]

Bullard, Claude L (Len) wrote:

> Then the problem becomes if by using that approach, 
> you are stuck in the place you get to, 80/20 can 
> hose you.  The HTML browser is a good example of 
> a cul de sac.  SVG may be another one.

The last 20% reminds me of something a teacher threw out in a freshman 
philosophy class. And this goes to support Eric's point that is a myth.

[hoping I remember correctly]

A philosopher (Xeno?) put forth a arrow could never reach its target 
because before it could hit it, it had to go half the distance. Before 
it got half way it had to get halfway to the halfway point -- ad infinitem.

-but it does get there-

Oh well, back to work.

• Prev by Date: Re: The myth of 80/20
• Next by Date: Re: schematron rules anormal behavior (off-topic)
• Previous by thread: Re: The myth of 80/20
• Next by thread: schematron rules anormal behavior (off-topic)
• Index(es):
  • Date
  • Thread