From: Reason (firstname.lastname@example.org)
Date: Sun Jan 26 2003 - 18:56:58 MST
> -----Original Message-----
> From: email@example.com [mailto:firstname.lastname@example.org]On Behalf Of Gordon
> > HTML is a good example of the direction programming should be heading.
> > For example, if I forget to type a closing
> HTML is a convenient example that everyone is familiar with. Yes, HTML
> goes to far, because you can type just about any damn thing you want
> and it still works. Not to mention the many compilers (rendering
> engines, aka browsers) render HTML in different ways. So yes, HTML is
> fault tolerance chaos. HTML is in the right general direction, though,
> in this case.
I think that HTML is absolutely the wrong direction to be going in. As
someone who codes in tag- and element-based high level languages as well as
standard high and somewhat-less-than-high languages, HTML is far and away
the one that gives me the most problems in debugging at a given level of
complexity. Code maintenance, updates and expansion are by far the costliest
part of TCO.
What you want as a starting point for improvement, IMHO, is a language or
entry method that makes it as hard as possible to tell the system to do
something other than what you mean it to do and still pass muster (compile,
interpret, etc). In other words, that close meanings in underlying
algorithms are in fact very distance from one another in the
language/entry-space. I think that some form of modern high level language
that solves this problem in a (slightly) analagous way to the original
introduction of qwerty keyboards would be a good thing.
This archive was generated by hypermail 2.1.5 : Wed Jul 17 2013 - 04:00:41 MDT