Cannot save anything from LexisNexis on Zotero 3.0 standalone

gideonse

This is very odd. I cannot save anything from LexisNexis using Zotero 3.0 standalone. I am using Chrome (though the problem is the same in Firefox), and there are no buttons in the URL, and when I download a PDF of the article and try to retrieve the metadata, Zotero says there isn't. Does anyone know what's going on?

adamsmith

not odd. There is no working LexisNexis translator and there hasn't been for years.
The problem is that Lexis doesn't provide any usable metadata to go by. I looked into writing a translator and I could make it work for my university's version of LexisNexis and the sources I use, but it was much too unstable for anything else.

fbennett

What adamsmith said. None of the legal vendors (Westlaw, Lexis, FastCase, Bloomberg) supply structured metadata in their pages. You can read more about the sad story here (or just in case that site is down, from Google cache here).

adamsmith

the same applies for the non-legal newspaper and news archives on lexis nexis academic.

gideonse

I was seeing references to this situation, but none that were very recent. How amazingly lame of LexixNexis, et al, especially since they offer exporting to EndNote and RefWorks. I'm trying to save to RefWorks and then export to Zotero, but RefWorks is so slow and unstable! I wonder if their refusal to work with Zotero is connected to it being open source, which must freak the poo out of an old media, old economy company like LexisNexis.

I wonder why some libraries are recommending the download-to-PDF and then retrieve-metadata-from-PDF method when that doesn't work. I did get some meta data from a PDF I downloaded via Explorer, but it was the *wrong* data.

adamsmith

oh, the Endnote export is new, that's great.
As a warning - it won't be very good - the data they supply is minimal.

That said:
1. Right now you can already import the Endnote output into Zotero. Save the RIS file that it will offer you and open it with Zotero.
Someone with more experience with Standalone than me should comment on the current workflow with this - with Zotero for FF it imports directly without any saving etc. in between.

2. If I saw this correctly, we may actually be able to write a translator for that. I'm not sure if the URL-structure across different LexisNexis versions is stable enough, but I'll give it a try at some point. That will take time, though, the site has everything to hate you can imagine.

sdspieg

Still no progress on this one? And is there really no way to somehow parse the exportable full-text files in such a way that at least the key fields (publication source, date of publication, title, byline, actual text etc) get entered in the right Zotero-fields? Here's an example that - it seems to me - shouldn't be TOO hard to parse, as it IS always the same structure:

1 of 367 DOCUMENTS




The Prague Post

April 3, 2013

The Mali dilemma

BYLINE: Prague Post staff

SECTION: NEWS

LENGTH: 1361 words

adamsmith

the problem are the frames on that page - they turn this into a bit of a nightmare.

sdspieg

But I mean the downloadable exports... I also notice that we can now export in some bibliographical formats too. When I select RIS, FF asks me automatically whether I want to import them into Zotero, which is nice. But I'm typically also interested in having the the full-text as well, and here some parser that could just put the relevant texts into the right Zotero record would of course be very useful. Any ideas on that?

aurimas

the problem are the frames on that page - they turn this into a bit of a nightmare.

We should be able to handle this properly. The only problem I foresee is that we currently won't get the Zotero pop-up showing progress.

I'd give this a shot, but as you mention, the RIS export is very minimal and I'm not keen enough on law citations to be able to properly parse the metadata out of text (though I can see how that would be a fun thing to do for LexiNexis </sarcasm>)

sdspieg

Well, the thing is that LexisNexis also have one of the better full-text newspaper collections, so it's not ONLY for lawyers :) As to the progress button - if that's the only drawback, I certainly wouldn't lose sleep over it!

aurimas

As to the progress button - if that's the only drawback, I certainly wouldn't lose sleep over it!

Granted that this does not affect Firefox (and maybe we'll come up with something for the other browsers), but the progress pop-up is somewhat important. I can see users being confused and trying to click the import icon several times resulting in multiple copies of the citation. (btw, the previous report where this topic came up is here)

sdspieg

I see what you mean. And yes, it is nice to see how far along the import is (e.g. which pdf files and which have not yet been downloaded). But still - if I sent one of you a file with a few hundreds of articles from LexisNexis, would it be hard to write sthg that can parse those and at least tag things like title, byline, date of pub, text, etc - all of this in such a way that we can then import that file into Zotero? If anybody would be willing to try - if you pm me, I will send you the dropbox link.

fbennett

I wrote a translator for law review articles on Lexis TLS several years ago, which worked reasonably well, so it can be done. I wouldn't want to end up on the wrong end of a lawsuit over their terms of service, though. What would really help would be a comfort letter from Lexis saying that they wouldn't take exception to the preparation of a Zotero translator covering specific resources.

(Sorry to be the bearer of a wet blanket on this, but I've just finished rereading some of the cases initiated by the Wexis twins, and they show a remarkable degree of creativity.)

sdspieg

Right. I'm really going to have to take a closer look at those things. But all I want to do is to use some textmining tools on these articles. So instead of just reading 100s of articles, I want to quickly get a feel for the main topics, the trends, etc. And then use that as the basis for a 'human' analysis. But I guess that's still close to e-discovery that I presume these guys are worried about (as they see big bucks there).

Frank, do you happen to know of a non-(or minimally-)legalese literature on this? Or could we get anybody from LexisNexis into our discussion on this here - to tell us what can and what cannot be done, and to explain their reasoning?

fbennett

I can contact someone at the Japan office, and ask for onward contacts. I'll write by private mail if I hear anything back, and post to this thread if a clear answer emerges.

sdspieg

Thanks much! And fingers crossed.

But so Frank, do you think that something like I propose here: to download a set of newspaper articles as LexisNexis allows, to parse those in such a way that we can enter them in Zotero with the key fields in the proper place, to store them (temporarily) in the Zotero (sqlite) database on our hard drive, to then textmine them with a bunch of textmining tools (like papermachines), and to then analyze those results and use some visuals in published articles - might be illegal? How do we infringe on their rights (or on those of the original copyright holders) then? We're not just giving those articles away to third parties, we're not even storing them permanently (although even on that score - I still have filing cabinets full here at home with academic articles I used to copy in the 90s 'for fair use') - we're just analyzing them.

fbennett

Exactly so. The problem isn't with copyright, but with the terms of service, to which fair use and the first-sale doctrine do not apply.

Your use is perfectly orthodox, it's just a matter of not getting incidentally squashed. Discretion is the better part of valour and all that.

I've posted that note (copy to you). We'll see that emerges.

sdspieg

Ok. To be continued.