Talk:Suggested pre-submit checklist

Ideas

I have a similar list on my userspace User:Charles R Greathouse IV/New sequence guidelines which is based on the issues I often see with new submissions. Feel free to take anything that looks useful to you.

Charles R Greathouse IV 00:13, 2 February 2011 (UTC)

There are some suggestions in there which even I should heed. (There's a couple of sequences I would have never submitted if I had given it a little more thought). Alonso del Arte 01:03, 2 February 2011 (UTC)

I have removed my page, replacing it with a redirect here. Charles R Greathouse IV 04:31, 1 July 2012 (UTC)

Overview

This page has a lot of really good advice. Unfortunately it seems to be drifting away from its original purpose: a checklist for use before submitting a sequence. I was thinking about adding an overview at the beginning that would be a simple checklist, while retaining all the existing advice on special cases below. My first attempt:

□ You have provided your name.

□ The terms of the sequence as you have it are absolutely correct.

□ The sequence is not already in the OEIS in any shape or form.

□ You can provide at least four terms of the sequence.

□ You can give a short description or definition of the sequence.

□ The short description or definition uses only standard terminology (not TeX/LaTeX, not Mathematica, not Maple, etc.).

□ You have references to books and journals that are directly relevant to the sequence.

□ You have checked that all external links work at the time you submit the sequence.

□ You have given one or two examples (unless your sequence is very simple).

□ Any programs given will run on a fresh start of Mathematica, PARI/GP, Maple, etc. (to make sure you have not overlooked some library or definition dependence).

□ The cross-references to other sequences are directly and obviously relevant to the sequence under consideration.

□ You have either keyword:nonn or keyword:sign, and have added other keywords that apply (e.g. keyword:base for base-dependent sequences).

□ There should be spaces or other breaks every 75 to 80 characters at the most.

□ There are no accented Roman letters or non-Roman letters in the submission (only printable ASCII characters).

What do you think?

Charles R Greathouse IV 18:41, 29 June 2011 (UTC)

I think it's good, and even better if we can make it so there are checklist boxes that would print out on paper. Alonso del Arte 18:48, 29 June 2011 (UTC)

I did a few small edits above, e.g. bold text and

, not Maple, etc.

(e.g. keyword:base for base-dependent sequences)

(only printable ASCII characters)

Concerning: "Any programs given will run on a fresh start of Mathematica, PARI/gp, Maple, etc." What about the version? Newest? — Daniel Forgues 05:44, 1 July 2012 (UTC)

We don't have a way of expressing that particular metadata at the moment. It's unfortunate. Charles R Greathouse IV 06:38, 1 July 2012 (UTC)

I added checkboxes. — Daniel Forgues 07:22, 1 July 2012 (UTC)

In most cases, we can use program comments. For example, A199264 has a program that will run in any version of Mathematica and one that requires 7.0 or later (for ArrayPad). The latter has a Mathematica comment to that effect. Alonso del Arte 16:16, 1 July 2012 (UTC)

Updates

I think this is a very useful page, and should be linked to on the SUBMIT page. However, there are a few points that might be arguable:

• There is ambiguity (intended?) about "Roman letters" and "printable (ASCII/ANSI/latin1) characters". I think there is most probably no more any device connected to internet which considers "ä, ö, ü" (standard letters in the German alphabet) and "é, à, ..." (standard in French) as non-printable. This may be considered a biased ("central European") point of view, but I think it is a fact. Without going as far as to say that "OEIS is UTF-8 encoded", I think that "moderate use" of such characters should be allowed specifically in the REFERENCES section, beyond names in AUTHOR and other signatures.
• The "line break or space" after 76 characters seems an incorrect instruction. In contrast to this, I would rather advice authors NOT to put explicit line breaks (nor (more than one) "padding spaces") in COMMENTS, especially. OTOH, in "difficult situations", authors should indeed make sure to have ONE SINGLE space wherever it makes sense, in particular around "=" signs, after "," in lists of values, etc., in order to "allow" (but not enforce) line breaks wherever possible.

I think the other good points (especially the first few) in the list would have more impact if those points which are arguable would be put aside or be somehow distinguished as "lower priority". (I'm thinking of my activity as Editor, where I (almost) cannot think of instances where accented Roman characters would have posed a problem, but still 90% of the submissions violate basic principles (relevance, (cross)references, ...) which come first in the list.) — M. F. Hasler 23:03, 4 November 2012 (UTC)

I agree with almost everything you've said and I have removed the last two points from the checklist. But in regards to the linebreaks, the problem, in my opinion, is more the bad example set by all the generating functions already in the OEIS with dozens of coefficients running on and on without any whitespace whatsoever to assist reading. New contributors see that and emulate it. Alonso del Arte 23:37, 4 November 2012 (UTC)

Examples

That's an interesting page, thanks. Minor nit, I prefer more than only one or two examples. Often the first term is special, e.g., implicitly uses prime(0) = 1, and therefore should be covered in the examples. That would leave only one ordinary example, not good enough for me, I don't grok number theory without numbers. For the typical case numbers k such that… it is clearer, about four or five examples might do, two or three "good" k, one or two "bad" k, and the first really interesting k.

Full disclosure, I haven't the faintest what “in the case of linear recurrences, be sure to add the signature” means, but as long as linear recurrence is a red link it is not my fault :-p –Frank Ellermann (talk) 20:36, 7 March 2020 (EST)