develooper Front page | perl.perl6.users | Postings from September 2018

Re: Appropriate last words

Thread Previous | Thread Next
From:
Shlomi Fish
Date:
September 11, 2018 19:13
Subject:
Re: Appropriate last words
Message ID:
20180911215207.7ca3d29e@telaviv1.shlomifish.org
Hi all,

On Mon, 3 Sep 2018 11:09:35 -0700
Larry Wall <larry@wall.org> wrote:

> On Mon, Sep 03, 2018 at 11:45:58AM -0500, Stephen Wilcoxon wrote:
> : Why the change in die handling between Perl 5 and 6?  Suppressing line
> : numbers with newline was very handy.  Alternatively, adding some sort of
> : directive would be more straight-forward (at least for Perl 5 users moving
> : to Perl 6).
> 
> Well, we did take Perl 5 users into account quite a lot, since many of
> us came from that direction, and we hope more will come in the future.
> 
> On the other hand, language design is all tradeoffs, and every time you
> add a low-powered feature (or keep one, in this case), you raise the
> barrier to entry for folks coming from other cultures, and you waste
> mindspace remembering things that don't buy you much.  For similar
> reasons, <> changed to lines() and all the magic rules about which
> variables are local to a loop went away. Likewise, a lot of the arcane
> knowledge of how a given function behaves in scalar or list context went
> away, mostly by splitting them into distinct operators that are easier
> to read and document.  Of course, the tradeoff is that that are then
> more operators.
> 
> So, sure, you could argue that we've just substituted one kind of arcane
> knowledge for another, but at least you can justify something like
> 
>     exit note “Phooey”;
> 
> by mere function composition without appealing to the authority of
> a particular paragraph in the manual.  And of course you could also
> compose your own function if the situation ever rises more than once in
> your program.
> 
>     my &eep = &exit ∘ &note;  eep “Phooey”;
> 
> In the specific case of this feature, one could also argue that putting
> that much semantic weight on the final character of the string is
> violating some kind of end-weight or one-pass principle, forcing a kind
> of mental time-travel for the reader, if not for the compiler.
> 
> Anyway, don't be a language designer if you want to make everyone happy
> all the time.  :-)
> 
> Well, actually, you can want it, just don't expect it... :-)
> 

I agree and I wrote about it here back in 2011 -
https://shlomif-tech.livejournal.com/57811.html ("Your Programming Language
Must Suck"). I also concentrate links to the more general theme of why one can
never please everyone here:
http://shlomifishswiki.branchable.com/Never_Try_to_Please_Everyone/ .

Regards,

	Shlomi
> Larry



-- 
-----------------------------------------------------------------
Shlomi Fish       http://www.shlomifish.org/
http://www.shlomifish.org/humour/ways_to_do_it.html

All Chuck Norris has to do is *look* at Perl code and it interprets itself
out of fear and respect. — DrForr
    — http://www.shlomifish.org/humour/bits/facts/Chuck-Norris/

Please reply to list if it's a mailing list post - http://shlom.in/reply .

Thread Previous | Thread Next


nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About