[Shootout-list] Directions of various benchmarks

[Bengt Kleberg]

skaller wrote:
...deleted
> Yes I know, but if you apply this 'external library rule' it all
> boils down to some arbitrary distinctions, and you end up,

i belive that we have to live with arbitrary distinctions. what we 
should do is to document the reasons for choosing this particular 
distinction over the alternatives. sometimes this will make it possible 
to make the ''best'' distinction at he time. sometimes it will show 
where we went wrong in our reasoning, and enable creating a new and 
better distinction.

the worst kind is the unstated, unargued arbitrary distinction.


> instead of comparing PCRE with PCRE with PCRE with PCRE,
> comparing PCRE with PCRE with FAIL and FAIL. Again not
> really interesting is it?

if i understand you correctly we will have some langugaes that have pcre 
builtin. they always register as pcre. then we have langugaes without 
pcre builtin. they register as fail.

if the normal way to use these languages is with pcre, why is there no 
builtin pcre? if pcre is not the normal way, why is there no 
alternative? (perhaps they are not supposed to be used for the task at 
hand? perhaps should they fail then?)


bengt