[
Lists Home |
Date Index |
Thread Index
]
On Wed, 05 Jan 2005 11:16:45 -0700, Uche Ogbuji
<uche.ogbuji@fourthought.com> wrote:
> I should also point out that this is an XML-DEV perma-thread, and as
> usual nothing new has been revealed in this round. I've been involved
> in at least two long rounds in the past, and I have no idea why I'm
> currently wasting time on it again.
I dunno ... at least Len's post
http://lists.xml.org/archives/xml-dev/200501/msg00144.html
can be bookmarked and pointed to the next time the permathread
emerges. I'm not sure I *completely* agree with it, but it hits most
of what I consider the high points:
- There is room for both static and dynamic typing in just about
everyon's toolbox.
- Neither provide a secret sauce to pour over real-world code to
prevent disasters
- There is a tradeoff either way -- dynamic/declarative tools tend to
have a higher run-time overhead, static tools impose a productivity
tax OR a tools tax.
- The quality of the code generated by declarative tools or
automagically invoked by dynamic tools improves as the technologies
mature. In the long run we can expect the dynamic/declarative stuff
to be more efficient, and the static checking stuff to be more
integrated into platforms (e.g. .NET and Java) and design/code-time
tools that eliminate much of the tedium.
- The trick is to optimize the mix of different approaches for a
specific project, not to grow so enamored of one particular Golden
Hammer that you use it to pound nails, drive screws, and cut lumber
:-)
- No tool or methodology will save you from human folly. The end of
Moore's Law is not that far away, but Murphy's Law and the Peter
Principle will rule forever, or at least until the Terminators finally
do us biological vermin in.
Each time around these permathreads I think we collectively learn a bit more.
|
