Re: saving AC development strength

[Akim Demaille]

| Akim Demaille wrote:
| > 
| > | > | 1.  changequote to something certain to not be confused (probably 
would
| > | > |     have been a really good idea way back when, but likely not now)
| > 
| > No, it has always been a bad idea.  It's just that noone was aware how
| > bad it was.  You're comparing a
| > 
| > ~/src/ace % wc -l [[...]]
| >    5589 total
| > 
| > with a
| > 
| > ~/src/ace % wc -l [[...]]
| >   14024 total
| > 
| > So yes, the problem is more visible now.
| 
| ???  Are you arguing this on the basis of *SPACE*???
| Where disk space now sells for US$ 0.000000002 per byte
| and DRAM goes for US$ 0.0000002 per byte.  Please tell me
| I am misunderstanding you.  Comprehensibility and usability
| are vastly more important than that!  Even if the issue
| is added processing time, I cannot see it outweighing
| ease of use.

No, I'm saying that with a small scale project, problems are less
likely to appear than with a full scale Autoconf.  In addition, the
figures are not faithful: 2.5x is *much* more factored than 2.13.
Some day I should do statistics on the average nested levels.

Also, 2.12 and 2.13 lived for so long that people learned its
programming interface, and believe it is the reference.  Sorry, but it
was junk.  If an Autoconf is right, it is definitely after 2.13.
Before, it was a mere miracle that it worked properly.  It was
incredibly badly quoted.

| > Go for [blind push & pop].  You *never* know if you're right or not.
| > People
| > don't seem to understand how bad static vs dynamic is.  Study this:
| 
| Given the usage constraints, I'm not sure I understand how
| it applies.  On the other hand, somebody may get ambitious
| and figure out how to do more complex stuff, ...

Sorry, I don't understand you :(