This post is by Lizzie.
I am working with some colleagues on how statistical methods may affect citation counts. For this, we needed to find some published papers. So this colleague started downloading some. And their university quickly showed up with the following:
Yesterday we received three separate systematic downloading warnings from publishers Taylor & Francis, Wiley and UChicago associating the activity with [… your] office desktop computer’s address. As allowed in our licenses with those publishers, they have already blocked access from that IP address and have asked us to investigate.
Unfortunately, all of those publishers specifically prohibit systematic downloading for any purpose, including legitimate bibliometric or citation analysis.
Isn’t that great? I review for all of these companies for free, in rare cases I pay them to publish my papers, and then they use all that money to do this? Oh, and the university library signed a contract so now they pay someone to send these emails… that’s just great. I know we all know this is a depressing cabal, but this one surprised me.
In other news, this photo is from my (other) colleague’s office, where I am visiting for a couple days.
I like how this post ended, in a way that’s not even comparable with how lib subscription fee ended
Did the letter from the University mention the legal basis of that restriction?
No, but based on the text they sent I assume it’s buried in the licenses the university agreed to (and we’re all party to if we want to use the library perhaps).
I hope the university is being efficient about this, so that the same employee who sends these emails is also the person who fakes the university’s U.S. News statistics and is also the person who’s in charge of doing nothing about reports of research misconduct and is also the person who runs a fake journal on the side. Each of those jobs is important, but individually they take up so much time on their own, so it would make sense to hire one person to do them all.
I need an office like your colleague has. And a daily driver like that boat. And a long lunch break for fishing.
Jd:
This is actually part of the standard benefits package for deans at the University of Nevada, Reno. So if you publish enough papers in the International Supply Chain Technology journal and you play your cards right, this job could be yours!
hot dog, I can probably draw marginally better than figures 2 and 3. 9wt fly rods, flashy streamers, and salt water vibes are in my future!
I’ve done several text mining studies of journal articles. Most of the big academic publishers allow bulk downloads for text mining purposes, as long as the university has a subscription. But you usually need to access the right URL or go through the publisher’s API. The first Google search result for “taylor and francis text mining” says they allow it, but they want you to contact them first. Your colleague’s library should have a point person for this kind of thing. I’d be happy to give you all a couple hours of my time if you want help sorting this out.
It was a librarian who contacted my colleague and there was no offer of helping them to get permission. Just a “immediately refrain from using Zotero, or any other tool or script, for this activity. Please reply to this email acknowledging that you have received it.” I fear this college may only have a point person to enforce crazy contracts they signed with publishers ….
This is perhaps obvious, but did your colleague talk to your librarian? That’s what librarians are for.
Also: there’s plenty to complain about regarding academic publishing, but what you’re writing about here doesn’t seem valid or sensible. Surely you can see the aim of the block against mass downloading, which professional society / non-profit journals also have?
It was a librarian (see my above reply) and, yes, I can see why they may not want mass downloading for some reasons, but why they would block IPs from places that subscribe to their journals and “specifically prohibit systematic downloading for any purpose, including legitimate bibliometric or citation analysis” doesn’t make sense. I’d hazard a guess that professional society / non-profit journals don’t spend so much time and effort on this (also, those publishers all include professional society journals now).
Good points! I doubt that the publisher is competent enough to check that the IP address isn’t from a subscribing institution. Hopefully your librarian will advocate / fight for you. I’m fortunate to have university librarians who are very good; we’re currently not subscribing to Elsevier (https://library.uoregon.edu/news/2022/uo-libraries-elsevier-contract-negotiations)
It having an aim doesn’t make it a good idea. I bet most of those papers the colleague was downloading were already on Libgen or have fulltext publicly accessible in Google Scholar. Horse, barn… (Would an economist call this post an example of ‘deadweight loss’?)
In the 90’s to 2000’s you could be excused for not getting it, but as things have changed dramatically in the scientific publishing / university education / enormous graft at the expense of the taxpayer world, if you’re publishing papers at all past 2010 or so you’re a part of the problem, like people who manufacture lead alloys and sell them to arms-makers who sell to enemy armys that are shooting up your own villages.
The only reasonable position at the moment is to be marching on washington and striking / refusing to publish anything in any journal / demanding that the US Govt through the library of congress fund the creation of a massive free online archive that anyone can contribute their papers to. It would be reasonable to have some rate limit I guess, to prevent pure spamming, maybe one PDF a day per person or something. Storage in triplicate of a 1MB paper costs about $.00004 plus electricity etc, so that’s a few pennies per year per person if you submit a PDF a day.
These days, said publications should be shared via ActivityPub published to the various keywords in the article, searchable and subscribable by authors, institutions, keywords, and such through ActivityPub federation.
Sure the federation stuff is pretty darn new, but the need for a public freely available archive has been obvious for well over 15 years
At this point it’s just Stockholm syndrome keeping the whole thing going. The graft publishers extract is part of the cost of grants which are sent to grant recipients, with 50-60% overheads sent to the institutions, the whole thing is a massive scam. Academic publishers receive something like $20B a year in payments.
The NSA massive utah datacenter only cost $1.5B to build and is believed to store “exabytes” of data (10^18 bytes). For 10M researchers in the US sending 10 MB of publications on average once a month, that’d store the output of all US researchers publications for 833 years. Of course it’d make more sense to also store datasets etc so maybe we could make it store all the output of PDFs and datasets for 80 years… with continual replacement of ever denser storage, that one facility (or one like it, or perhaps two geographically distributed) would store the entire product of all US research indefinitely.
Given these restrictions, I don’t know how we’ll ever get an LLM trained on good methodology, to help with critiquing or synthesizing articles, or providing (especially juniorish) researchers with substantive decision support as they design studies and plan analyses.