Lists Home |
Date Index |
> get some perspective, and this usually doesn't take long at all. I
> had much occasion working with developers who curse and splutter all
> time at all the little tripping points of XSLT that Mike Kay points
> But at least twice, I remember a developer saying, after a few days of
> this, something to the effect of "wow, XSLT is certainly a different
> of thinking, but I must say that I accomplished my XML processing task
> using XSLT in a fraction of the time it took for me to do the same
> DOM and Java". I have also heard variations of this (i.e. APIs other
> DOM and languages other than Java) several times.
> In the end, productivity is what makes people accept XSLT.
Yes, XSLT is like XML in this respect. Seductively easy to get snared
in its embrace, but painfully unforgiving when someone unknowingly
violates one of its expectations. The specs try hard to minimize the
number of death-traps waiting to swallow the inexperienced, but people
are bound to get stuck from time to time. But even with the current
situation of "muddle along and get bloodied until you find
productivity", I think XML is an enriching institution. And since only
two percent of the travelers read the map before heading off across the
field, it is up to the book authors and tool vendors to put big orange
flags near the death-traps that we discover, so more people make it to
the land of productivity without getting bad attitudes.