[E-Lang] empirical and statistical comparison of languages

[zooko@zooko.com]

 MarcS wrote:

> > The Java code was written in 1997/1998 -- the Collection API wasn't added
> til
> > the end of 1998.
> 
> Yeah, but there was a hashTable in the API on the first day it shipped.

Uh...  <Zooko looks in docs.>  Oh yeah -- java.util.Hashtable.

I'm embarassed that I had forgotten about this.  I definitely never implemented
my own hashtables in Java back then, and I don't really remember using
java.util.Hashtable for anything specific.  It's possible that I did use it
rarely, but it clearly wasn't the fundamental data structure to me back then the
way it is now.  (Mojo Nation uses Python "dicts" for all inter-broker messages
as well as for most internal data structures.)


> The author did note that the Java programmers might be less skilled than
> counterparts in other languages since it was early in Java's evolution. That
> could have had something to do with it.

Uh-huh.  ;-/


> Personally, I think I still would have used the Java hashtables if faced
> with this problem, but I would have had sympathy for people who didn't.

I can't really say the same thing.  I probably *did* face problems that would
have been more easily and efficiently solved with a hashtable than with the
iterative data structures, but the hashtable wasn't the tool that was "close at
hand" so that I realized how it could be applied.  I think you are right that
supporting hashtables with syntax helps.


Regards,

Zooko