Received: by 10.180.82.193 with SMTP id k1mr1495344wiy.0.1350559364450;
        Thu, 18 Oct 2012 04:22:44 -0700 (PDT)
Path: q10ni65173218wif.0!nntp.google.com!feeder1.cambriumusenet.nl!feed.tweaknews.nl!194.134.4.91.MISMATCH!news2.euro.net!newsfeed.freenet.ag!feeder.erje.net!eternal-september.org!feeder.eternal-september.org!mx04.eternal-september.org!.POSTED!not-for-mail
From: "LudovicoVan" <ju...@diegidio.name>
Newsgroups: comp.lang.prolog
Subject: Re: Multi-argument vs. JIT indexing (Was: Destructive assignments (Was: Win from call/n))
Date: Thu, 18 Oct 2012 12:20:23 +0100
Organization: A noiseless patient Spider
Lines: 1
Message-ID: <k5ooq3$lrn$1@dont-email.me>
References: <k4ehls$urk$1@news.albasani.net> <2012Oct12.123146@mips.complang.tuwien.ac.at> <k59h75$uvf$1@dont-email.me> <k59nt3$naf$1@news.albasani.net> <92edbe3f-33e4-42c0-a9f3-6ecf4d18892f@googlegroups.com> <2012Oct15.202232@mips.complang.tuwien.ac.at> <k5jch2$rbv$1@news.albasani.net> <k5jh3f$6fa$1@news.albasani.net> <k5k24m$ea0$1@news.albasani.net> <507e9e3a$0$921$afc38c87@read01.usenet4all.se> <2012Oct17.143309@mips.complang.tuwien.ac.at> <507ecc7a$0$923$afc38c87@read01.usenet4all.se> <2012Oct17.194630@mips.complang.tuwien.ac.at>
Mime-Version: 1.0
Injection-Date: Thu, 18 Oct 2012 11:22:44 +0000 (UTC)
Injection-Info: mx04.eternal-september.org; posting-host="9090c192861416d488220b95965efcb9";
	logging-data="22391"; mail-complaints-to="ab...@eternal-september.org";	posting-account="U2FsdGVkX1+piSNHdPns1a2JQe326vCUZKNEq6t/ndY="
X-MimeOLE: Produced By Microsoft MimeOLE V14.0.8089.726
In-Reply-To: <2012Oct17.194630@mips.complang.tuwien.ac.at>
X-Newsreader: Microsoft Windows Live Mail 14.0.8089.726
Importance: Normal
Cancel-Lock: sha1:64G5dMs5qGALGs2BJEzphMgAvjA=
X-Priority: 3
X-MSMail-Priority: Normal
Content-Type: text/plain;
	format=flowed;
	charset="iso-8859-1";
	reply-type=original
Content-Transfer-Encoding: 7bit

"Ulrich Neumerkel" <ulr...@mips.complang.tuwien.ac.at> wrote in message 
news:2012Oct17.194630@mips.complang.tuwien.ac.at...
> Jan Wielemaker <j...@invalid.invalid> writes:
>>On 2012-10-17, Ulrich Neumerkel <ulr...@mips.complang.tuwien.ac.at> wrote:
>>> Jan Wielemaker <j...@invalid.invalid> writes:
<snip>

>>>>This is also more or less what SWI-Prolog does.  The only difference
>>>>is that it doesn't do anything at compiletime.  Indexing opportunities
>>>>are considered at the first call, creating the best single-argument
>>>>index for that call.  ...
>>>
>>> In SWI, indexing is better on the first argument (that is similar
>>> to Eclipse) than on the other arguments.  In the first argument,
>>> determinism is maintained even if the first clause contains
>>> a variable.  On the other arguments this is not the case
>>
>>Yes.  Basically, there are linked lists ....
>
> What I hope is that we find a good terminology for all of this
> without referring to implementation details. Currently,
> multi-argument indexing is a very vague term.

I'd venture that it would be nice if a standard distinguished between -say- 
explicit and implicit indexing.  The former would be in control of the 
programmer, explicitly stated in code, the default being the usual 
first-argument indexing, and would guarantee to the programmer his or her 
own coding choices.  The latter would be implementation-specific compile- 
and run-time automatic indexes, which would improve determinism and 
performances under the constraint that they do not alter the expectations 
set by explicit indexing.

-LV