[seqfan] Re: Links and references in the OEIS, my policy

[M. F. Hasler]

Indeed the problem is that the reference section does not allow links. So
if one wants to add a link to a reference, one has to move it to the links
section.

But nobody would ever delete a serious reference in the links section if
ever the link would break (which only happens when the link was badly
chosen).

I was speaking about the "bad" links people add, usually for self
promotion. Links that are never visited and remain broken for 20 years
without ever somebody nothing it because nobody would ever look at that web
page.

In popular sequences like primes, triangular numbers etc, there are too
many such links which finally make the "links" section useless, or a pain
to use (the interested reader has to search himself for useful links to
serious references in an ocean of useless links). It's easy to find 200 web
pages dealing with primes or triangular numbers or Pythagorean triples etc,
by just typing this keyword into Google.
 If OEIS wants to provide something useful, it must not try to compete with
Google in providing a maximum number of links to web pages which deal with
a given subject. Google is better at this.
The added value OEIS *could* provide would consist of carefully choosing
the best among these. If we don't operate a strict selection, the links
section is worth less than a single link to google.con/search?q=keyword

I proved this by taking at random a typical sequence in which 80% of the
*non-serious* links were broken for decades and nobody ever noticed it.
(I never talked about links to serious references. These are not a problem
in any way! and actually they almost never break, and in the rare cases
this happens, a better link to these references is readily found. If ever
this weren't the case, the reference could be moved back to the references
section. Nobody ever deletes a serious reference because of a broken link.)

--
Maximilian

On Sun, Dec 8, 2019, 13:32 jean-paul allouche <jean-paul.allouche at imj-prg.fr>
wrote:

> Why not? in particular if the reference is a paper, a book, etc.
> and the link... a link. In any case, better a duplicated entry than
> an --ultimately dead-- link.
> best
> jp
>
>
> Le 08/12/2019 à 18:21, michel.marcus at free.fr a écrit :
> > But then we will have duplicated entries in refs and links ?
> >
> > ----- Mail original -----
> >
> > De: "Chris Thompson" <cet1 at cam.ac.uk>
> > À: "Sequence Fanatics Discussion list" <seqfan at list.seqfan.eu>
> > Envoyé: Dimanche 8 Décembre 2019 16:09:07
> > Objet: [seqfan] Re: Links and references in the OEIS, my policy
> >
> > On Nov 30 2019, Neil Sloane wrote:
> >
> >> Dear Sequence Fans, There have been discussions among the editors that I
> >> don't agree with, andI would like to make some comments:
> > [ ..snip.. ]
> >> 3. Replacing a reference by a link is not always a good thing. When the
> >> link breaks, as it will sooner or later, the citation gets [broken link]
> >> added. Whereas the original reference could still be useful if you have
> >> access to the right library.
> >>
> >> 3.1 That is why we keep references and links separate.
> > I do very much agree with this, and consider it a mistake that entries
> > began to be moved from "References" to "Links" just because an online
> > version became available. Of course, it is useful to provide such a link
> > if one exists. If the link subsequently breaks and cannot be mended, drop
> > the link but keep the reference - which would be the natural action if
> > it had remained in the "References" section.
> >
>
>
> --
> Seqfan Mailing list - http://list.seqfan.eu/
>