W3C home > Mailing lists > Public > public-hydra@w3.org > November 2015

Re: Infinite TPF fragments

From: John Walker <john.walker@semaku.com>
Date: Tue, 3 Nov 2015 15:56:44 +0100 (CET)
To: Ruben Verborgh <ruben.verborgh@ugent.be>
Cc: public-hydra@w3.org
Message-ID: <1440060789.329286.1446562604816.JavaMail.open-xchange@oxweb05.eigbox.net>
Hi Ruben,


> On November 3, 2015 at 2:37 PM Ruben Verborgh <ruben.verborgh@ugent.be> wrote:
>
>
> Hi John,
>
> > In a recent post [1] Ruben suggested "The number of pages (and next links)
> > per fragment MUST be finite."
>
> That suggestion no longer holds, but yes, I suggested this :-)
>
> > Would there be any use of a mechanism to indicate the fragment is infinite
> > to prevent a client 'dumbly' going into an infinite loop.
>
> It would not be an infinite loop in the sense of "being stuck", because new
> data is always pulled.
> But it would indeed never terminate.

Indeed, that is a better way to put it.

>
> > An obvious example of such a fragment would be the sequence of prime
> > numbers, or any other infinite series (Fibonacci, factorial).
> > Not sure of the practical use of this, but perhaps worth considering.
>
> In combination with streaming results, this would not really be an issue.
> E.g., for practical purposes, http://fragments.dbpedia.org/2015/en can be
> considered "infinite",
> but you can still query it:
>
> http://client.linkeddatafragments.org/#datasources=http%3A%2F%2Ffragments.dbpedia.org%2F2015%2Fen&query=CONSTRUCT%20WHERE%20%7B%0A%20%20%3Fs%20%3Fp%20%3Fo.%0A%7D

Sweet :-)

Although fan starts whirring rather alarmingly in my laptop after not too
long...

>
> Best,
>
> Ruben
>
>

John
Received on Tuesday, 3 November 2015 14:57:29 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 3 November 2015 14:57:29 UTC