As long as a given design has trade-offs, a second design 
will trade-off the opposite features.  No, there is nothing 
wrong with that as long as the reasons are technical.  No 
that isn't always the case.  So smart people tolerate 
multiple approaches even if they don't learn them.

Then the question is, did the subject line technologies 
fail, or did they simply come out as the choices of 
minorities of users? Smaller doesn't mean worse.  Worse 
doesn't mean better.  Yet for that reason, the architects 
make choices for large projects.  One has to determine 
the local costs for global coherence.

The phenomenon of lock-in is well-understood.  It doesn't 
mean early adopter or first.  It means a majority percentage 
with a feedback loop that corrects perturbations so the 
percentage is maintained.  HTML has that.  XML has that. 
DOM has that.  A standard without that isn't one but a 
standard is inside a bounded set.

If XOM works better inside Apache, sure.  Ask if it is 
good for the local members before joining a union.

len


From: Rick Jelliffe [mailto:rjelliffe@a...]

People are fixated that standardization means unification: that there 
should only
be one way to do something ever standardized. But is the Java world 
really worse
off with Eclipse using SWT and Java supporting Swing? Is the world 
really worse
off that MS has Win32 and .NET?  Is the world really worse off that 
there is HTML
and PDF? Or rather HTML and XHTML? (Now I don't see that there is much need
for two standards where one is a spoiler, you saying tomato when I say 
tomato. Different
issue.)

However just because two technologies largely overlap is surely no 
reason to deprecate one: DOM/JDOM/XOM do the same things perhaps but 
approach the problem from different angles. In the meantime, if Apache 
is the only brand at the moment, is the answer for  Elliotte to move XOM 
to  Apache?