[seqfan] Re: Looking for a recent draft

[M. F. Hasler]

On Sun, Jan 15, 2023, 06:33 Michael Branicky wrote:

> are you thinking of A089640?
>

I'm hijacking this thread (which anyway isn't for [seqfan]s but for
[oeis-editors]...).

In that sequence i was one of these instances where in the program section
of sequence Axxx we have a copy of a program for sequence Ayyy.
Such a "stale" copy of code which doesn't belong there is always very bad
for many reasons:

# it simply doesn't belong there

# it  unnecessarily fills up space and clutters the record of sequence Axxx

# there may be a much better version of the program "in the sequence" where
it belongs, but readers won't go there but copy-paste the obsolete program
for Ayyy together with the code for Axxx. If readers were obliged to go
there, they might find some important other information there.

# the different stale copies of Ayyy might not even work if a language
changes as was the case from python 2 to python 3

# even worse and maybe most importantly, one might discover that the code
for Ayyy yielded wrong results for larger terms. Then all depending
sequences that used a stale copy instead of the corrected version will
"silently fail", i.e., produce wrong values without anyone noticing it.

# all of this is even worse when the programmer doesn't call the stale copy
of the function Ayyy with it's proper name, but for example g(.) or L(.) or
t1(.) or similar, which is unfortunately very often the case, even for very
elementary and important sequences like (1, 1,2, 1,2,3, ...) and/or the
same with 0 and similar.
 In that case, when a bug is found, it's nearly impossible to find all the
places where this bug leads to incorrect values in other sequences.

# there is no justification to practice this bad habit since it takes only
a click on the name of the function Ayyy to get the latest version of that
program from where it ought to be.

So, to come to the point, shouldn't we discourage this bad and potentially
subversively harmful practice explicitly in the style guide or a similar
place?

-Maximilian

On Sat, Jan 14, 2023 at 4:43 AM Michel Marcus <michel.marcus183 at gmail.com>
> wrote:
>
> > Hello
> >
> > I am looking for a recent draft, where there was a discussion about a
> > formula (by Robert Price ?) .
> > The formula was something like a(n) = Sum_{k=0..n} C(k,
> n-floor((3n+2)/2)).
> > But it appeared to give a(n)=0 for all n ?
> >
> > But I can't find that sequence.
> > If you were the author of the draft, can you tell me ?
> > I'd like to know how it ended.
> >
> > Thanks.
> > Michel
> >
> > --
> > Seqfan Mailing list - http://list.seqfan.eu/
> >
>
> --
> Seqfan Mailing list - http://list.seqfan.eu/
>