use data from MathSciNet to update publications list

[haraldschilly]

create library-publications-mathscinet.html page generated by the bib file.

• [x] file added as mathscinet.bib and integrating it into build process
• [x] fixing errors -- see https://github.com/sagemath/publications/pull/76
• [ ] fixing special character translation table, e.g. for Mart\\'\\inez
• [ ] website integration: https://github.com/sagemath/website/issues/112

[mmasdeu]

I have compiled all the publications in MathSciNet citing either "Sage Mathematics" or SageMath. Since MathSciNet is not friendly when giving back these data, the process is far from automatic...

all_sage_citations.zip

[williamstein]

the process is far from automatic... = hour(s) of pain. Thanks @mmasdeu !

[haraldschilly]

ok, wow, 727 publications are a lot. Any idea how we can reconcile this with what we already have? We can't just replace it, because there are potentially some publications in our list that aren't in this one and with some we e.g. link to the actual PDF file.

[mmasdeu]

I think that the list as it is now is not very useful in 2017. It is also very biased (because it requires extra steps from the authors). I would propose that it is kept as a legacy subpage as it is now, but the main page mirrors what can be found at MathSciNet at any given time (+- some months, until this can be properly automated). Any thoughts?

[haraldschilly]

So, just to understand you correctly, you mean to remove much of the content that's on the current publications page? e.g. the first thesis "Martin Albrecht. Algebraic Attacks on the Courtois Toy Cipher" is not in the MathSciNet publications bibfile. It also to the thesis PDF itself being hosted on sagemath.org.

It's 2006, so I can understand you saying that this isn't useful, though. I just want to point this out as a detail to think about.

What we can definitely do is to start a new page "library-publications-mathscinet.html" and integrate this in it. Later, we can think about switching over, merging, etc...

[haraldschilly]

Or, "H. C. Rhaly Jr. and B. E. Rhoades. The weighted mean operator on ℓ2ℓ2 with weight sequence ..." from 2014 would also be gone, since it's not in the MathSciNet list either.

[mmasdeu]

Sorry, my comment was (deliberately?) not very clear! I was suggesting to put the existing list as a subpage so no information is lost, but the more automated list in the main page. Of course there may be reasons to do the opposite, but I don't think that for now these should be merged.

And yes, preprints and ArXiv etc won't appear as of now...an alternative is to use the (IMO much more chaotic than MathSciNet) Google Scholar...

Marc

marc.masdeu AT gmail DOT com

On Wed, Jul 19, 2017 at 10:54 AM, Harald Schilly notifications@github.com wrote:

Or, "H. C. Rhaly Jr. and B. E. Rhoades. The weighted mean operator on ℓ2ℓ2 with weight sequence ..." from 2014 would also be gone, since it's not in the MathSciNet list either.

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/sagemath/publications/issues/75#issuecomment-316413516, or mute the thread https://github.com/notifications/unsubscribe-auth/ABkAIi-bRq86ErXTt_j6HJuOgvn6S-_tks5sPhiwgaJpZM4ObDL4 .

[haraldschilly]

ok, I've started to integrate this file. See https://github.com/sagemath/publications/pull/76 ... it currently fails with some encoding problems, though.

[haraldschilly]

Several MRREVIEWER entries do contain wrongly encoded utf8, or some other issue. I've removed all of them. Let's see if this works.

[haraldschilly]

Ok. https://github.com/sagemath/publications/pull/76 merged. There were also problems with some names. I don't know what's going on, but I fixed it in a way that it at least compiles. See: https://github.com/sagemath/publications/commit/b0d1d5a2ac574be1250b538ab7e3feddb7e48a88 for what I changed. Better ideas and closer review of how it is processed is welcome!

The follow up of this for the website itself is https://github.com/sagemath/website/issues/112

[haraldschilly]

ok, next step is to fix the special character translation table. i.e. we have to figure out how to display Mart\\'\\inez and some other names.

replace_special function here in https://github.com/sagemath/publications/blob/master/pubparse.py#L932

[dimpase]

@haraldschilly - What is the status of this? I see that e.g. the 1st paper in mathscinet.bib does not seem to show up on the website. Is mathscinet.bib not used in building the website?

[haraldschilly]

@dimpase I don't know, I haven't looked into this since my last comment. Just by looking at my last comment, there seems to be work left to do, before this can be used. Hence, to your second question, the answer is no.