Newspaper Sites?

milst1

When I visit an article on the http://www.nytimes.com website, the little Zotero icon in the URL field is a rolled up newspaper, and Zotero saves the citation categorized as a newspaper article.

But other newspaper sites (e.g., http://www.sandiegouniontribune.com/) have the generic Zotero web site logo in the field and I have to manually convert the cite to newspaper.

Is there a way to fix this?

Apologies for not doing a very thorough search on this topic first. Please send me to previous discussion if it exists. Many thanks.

adamsmith

Zotero only imports correctly where either the website has enough metadata embedded in the header (which is almost exclusively the case for academic sites) or where someone wrote a translator specifically for that site.

For the US, that's only the case for the major papers, I believe among daily papers currently NYT, FT, WSJ, and WP and maybe the LA Times (that one breaks quite a bit because they keep changing their site). There is no translator for the SD Trib, nor for most other papers with mostly regional appeal.

milst1

Thank you for your response.

So is it very difficult to write a translator for some of these regionally important newspapers? I do business research and some of the regional papers are very important for research on national industries (e.g., Orlando Sentinel for the theme park industry, San Jose Mercury News for tech, etc.).

I would be willing to give it a whack, maybe over winter break. :-)

adamsmith

Really depends: if you know some javascript or other types of programming and ideally have a basic idea what an xpath and a regex is, then it's quite easy. Otherwise it's quite hard.
Documentation:
https://www.zotero.org/support/dev/translators/coding
and a simplified version that will likely work for most newspapers:
https://www.zotero.org/support/dev/translators/framework

milst1

Admittedly not in my wheelhouse and I haven't written a line of code since I created a website in 1994 using html, if that even qualifies.

But having to manually adjust the citation every time I come across this problem might be costing me as much time as it would take to understand what an xpath and a regex are. Thank you very much for the link to the documentation.

adamsmith

If you do want to do it yourself (which is very much encouraged) several of us are happy to try to help (within reason, of course--you have to do the heavy lifting), preferably on the development listserv at
https://groups.google.com/forum/#!forum/zotero-dev

If you have a research budget, it's also something that you could contract out. E.g. I do that type of work for 200-300US$ a piece for simple newspapers sites and you may be able to find cheaper options on freelancing platforms like Upwork.

milst1

Thanks for the information. I don't have a research budget.

Does one have "rights" to a translator that one develops, or does it get incorporated into Zotero and become a public good?

I'm just curious. If I learned how to do this, I would want others to be able to benefit.

Looking for my codeacademy password...

adamsmith

you can obviously keep a translator to yourself (no one would be able to prevent that), but if you make it available, that would need to be under the same free&open license (AGPL) that Zotero is published under, so you wouldn't keep any exclusive rights. You'd be credited in the translator itself. Not very visible, but it's certainly a piece of pride.

zuphilip

All translators from Zotero are online available at https://github.com/zotero/translators/ (and they are also locally in your profile folder).