Machine-readable exports

[arbox]

Hi @chrisseaton,
I highly appreciate your effort to collect Ruby related research sources!

Since I copy them to my bibliography manager (Zotero for now and probably ever) I thought, that it will be useful to provide a download section for an e.g. BibTeX file with all citations.

Or we could turn it around and generate the whole site using e.g. jekyll-scholar from machine-readable sources.

I would like to help you with it since I update my BibTeX files anyway.

[chrisseaton]

Yes we could have a BibTeX file, and we could possibly generate the page from that.

Can you pitch how you think it would specifically work or do a proof-of-concept?

[arbox]

I'll prepare a PR for jekyll-scholar.

[chrisseaton]

I'm keen to preserve most of the look-and-feel of the current site by the way - such as the badges, which I think are useful, and the simple reference formatting that hopefully won't put people off. I don't know how much of that is achievable in an automated system but let's find out.

[arbox]

Yep, I target an exact copy of the current site but generated from a BibTeX file. Sections and tags/badges can be easily preserved via notes in the bib source.

[chrisseaton]

@arbox if you're still working on this you should probably rebase in order to avoid a tricky merge.

[arbox]

@chrisseaton ,
will do!

[chrisseaton]

@arbox did anything come of this idea?

[arbox]

Not that much. For now I have all publications in a Zotero library which can be exported as BibLaTeX (we could give it to users as a Download). The site generation from this file is still in progress.

[chrisseaton]

Are we still going to do this or shall I close the issue?

[arbox]

We are going to do this.

My initial thought on generating the whole page from the BibTeX source doesn't sound as fruitful as expected since we'll need to convert all PRs into a BibTeX source or require people to provide a valid BibTeX snippet.

Please look at this except of the whole bibliography: https://www.zotero.org/groups/2274671/rubybib/items

The site structure is reflected now in form of tags (both categories and interpreters). Tags can be exported as keywords in BibTeX.

We could either link to this Zotero group and let people export some or all items in a format of preference, or we could do a BibTeX export and upload it directly to the RubyBib site.

I commit to support this Zotero group library (and Zotero seems to stay for long) and add new sources from the PRs.

What do you think?

[chrisseaton]

we'll need to convert all PRs into a BibTeX source or require people to provide a valid BibTeX snippet

I think that's ok - people opening PRs will normally be paper authors who will know BibTeX. Or we can write it for them if they open an issue.

We could either link to this Zotero group

I don't know anything about Zotero but I don't mind linking to it if it has a simple BibTeX export link? We could do that right now if the bibliography were fully populated there, and we'd update it ourselves after updates. We'd need to let ideally anyone do that?

[arbox]

@chrisseaton,

while compiling bibtex sources I've checked index.md for links correctness using the awesome_bot gem: bundle exec awesome_bot index.md --allow-dupe.

The results are below:

Issues :-(
> Links 
  01. [L026] 302 https://dl.acm.org/citation.cfm?id=3046681  → https://dl.acm.org/citation.cfm?id=3046681&cookieSet=1 
  02. [L034] 301 http://stefan-marr.de/downloads/dls16-marr-et-al-cross-language-compiler-benchmarking-are-we-fast-yet.pdf  → https://stefan-marr.de/downloads/dls16-marr-et-al-cross-language-compiler-benchmarking-are-we-fast-yet.pdf 
  03. [L046] 404 http://kuramitsulab.github.io/paper/rujit.pdf/  
  04. [L064] 302 https://github.com/HPI-SWA-Lab/TargetSpecific-ICOOOLPS/raw/gh-pages/call_target_specific_method_arguments.pdf  → https://github.com/raw/HPI-SWA-Lab/TargetSpecific-ICOOOLPS/gh-pages/call_target_specific_method_arguments.pdf 
  05. [L082] 301 http://www.matthiasspringer.de/downloads/BP2012H1_intra-language_collaboration.pdf  → http://matthiasspringer.de/downloads/BP2012H1_intra-language_collaboration.pdf 
  06. [L085] 302 http://id.nii.ac.jp/1001/00083504/  → https://ipsj.ixsq.nii.ac.jp/ej/?action=repository_uri&item_id=83510 
  07. [L088] 302 http://id.nii.ac.jp/1001/00073661/  → https://ipsj.ixsq.nii.ac.jp/ej/?action=repository_uri&item_id=73661 
  08. [L094] 302 https://dl.acm.org/doi/10.1145/1869631.1869634  → https://dl.acm.org/doi/10.1145/1869631.1869634?cookieSet=1 
  09. [L104] 302 https://dl.acm.org/citation.cfm?doid=1449764.1449796  → https://dl.acm.org/citation.cfm?doid=1449764.1449796&cookieSet=1 
  10. [L127] 302 http://www.cs.rochester.edu/u/scott/papers/2014_PLDI_DPR.pdf  → https://www.cs.rochester.edu/u/scott/papers/2014_PLDI_DPR.pdf 
  11. [L130] 301 http://researcher.watson.ibm.com/researcher/files/jp-ODAIRA/PPoPP2014_RubyGILHTM.pdf  → https://researcher.watson.ibm.com/researcher/files/jp-ODAIRA/PPoPP2014_RubyGILHTM.pdf 
  12. [L154] 301 http://www.cs.umd.edu/~jfoster/papers/pldi16.pdf  → http://www.cs.tufts.edu/~jfoster/papers/pldi16.pdf 
  13. [L156] 301 http://www.cs.umd.edu/~jfoster/papers/oops13.pdf  → http://www.cs.tufts.edu/~jfoster/papers/oops13.pdf 
  14. [L158] 301 http://www.cs.umd.edu/~jfoster/papers/popl11.pdf  → http://www.cs.tufts.edu/~jfoster/papers/popl11.pdf 
  15. [L160] 301 http://www.cs.dartmouth.edu/reports/TR2011-686.pdf  → https://www.cs.dartmouth.edu/reports/TR2011-686.pdf 
  16. [L162] 302 http://projekter.aau.dk/projekter/files/61071016/1181807983.pdf  → https://projekter.aau.dk/projekter/files/61071016/1181807983.pdf 
  17. [L184] 404 http://lafo.ssw.uni-linz.ac.at/truffle/debugging/dyla14-debugging-artifact-0557a4f756d4.tar.gz  
  18. [L192] 301 http://phenomenal-gem.com  → https://github.com/phenomenal/rphenomenal/blob/master/public/The_Phenomenal_Gem_Poncelet_Vigneron_2012.pdf 
  19. [L200]  http://dspace.mit.edu/openaccess-disseminate/1721.1/102281 Net::ReadTimeout 
  20. [L214] 301 http://www.biomedcentral.com/content/pdf/1758-2946-5-35.pdf  → https://www.biomedcentral.com/content/pdf/1758-2946-5-35.pdf 
  21. [L216] 301 http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2951089/pdf/btq475.pdf  → https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2951089/pdf/btq475.pdf 
  22. [L236] 301 http://link.springer.com/chapter/10.1007/11787044_13  → https://link.springer.com/chapter/10.1007/11787044_13 
  23. [L240] 302 http://jla.sagepub.com/content/12/1/56.full.pdf  → https://journals.sagepub.com/doi/pdf/10.1016/j.jala.2006.07.013 
  24. [L244] 301 http://ieeexplore.ieee.org/xpl/articleDetails.jsp?tp=&arnumber=6476557  → https://ieeexplore.ieee.org/xpl/articleDetails.jsp?tp=&arnumber=6476557 
  25. [L256] 301 http://dx.doi.org/10.6084/m9.figshare.97244  → https://dx.doi.org/10.6084/m9.figshare.97244 
  26. [L260] 403 https://www.ipa.go.jp/osc/english/ruby/  

Wrote results to ab-results-index.md.json

I've corrected 403 and 404 errors (#38), the redirects should be handled separately (I would do it, but it's not strictly required). The awesome_bot is bundled, probably we should add it to CI checks.

[chrisseaton]

The redirects are often HTTPS. I'll look at these.

[arbox]

Yes, mostly. In some rare cases there was a migration.

[chrisseaton]

Links fixed in c7e080c.