I agree, and I'll explicitly add "history" as a part of "science" in the research sense you meant. I'm having a heck of a time trying to access historical records of medieval England from here in Silicon Valley because, although the information I need (land deeds, court records, etc.) was uncovered & translated by Victorian historians, and much of it exists in both text and image form and is ALREADY ONLINE in Google Books, the various English universities & their presses, English historical associations, libraries, etc., all jealously guard the information instead of releasing it, even to the point of preventing Google from showing it.

Example: Cambridge Univ Press takes books that were Victorian databases of medieval records, and in the name of "protecting our precious history", they just photocopy the Victorian pages, reprint them on newer paper, and put them under new copyright. If you want to look something up in the 10-volume database, then either blindly buy the $300 set of books and hope to find some items of interest scattered therein or go to the nearest library that has the full set, which turns out to be on the other side of the planet at Leeds Uni in Yorkshire, which will allow you to see these ancient texts printed in 2013 if you pay for library use and make an appointment at least 2 days in advance....

Meanwhile, the full text of the books is on Google Books at Google's expense and available in any browser, but Google is forced by the "protectors of history" at Cambridge to cloak many of the pages of 1000-yr-old data, because history is too precious to allow the unworthy to see their photocopied Victorian texts without first going on an old-fashioned quest, bribing the boatman, answering the troll's three questions, etc.

My hope is that at some point there will be a cultural change among "historical preservation" organizations where they decide that the greatest thing they can do to promote their field is to find every original document in every collection, carefully photograph it in hi-rez using whatever optical frequencies bring out the most faded detail, and contribute it to a free online host. Next step is then to create and freely post transcriptions, translations, and indexes, so that ANYONE can use the data for research, not just those who have enough gold in their purse, time for the quest, and can "answer me these questions three".

I’m with you, but the reason these documents aren’t free and online is money, of course. These archives, libraries and institutions are chronically underfunded. Scanning is expensive. They use copyright as revenue generation to plug the gap. I’m certain the academics that work there would love to publish all of history online for free, as long as you can tell them where their salary is going to come from.

I agree with everything you said, so I'm just adding some thoughts.

I'm trying to do historical research at my own expense and, like many others, will continue to do so as long as I can get access to raw data. Google will provide much of the data for free, and I (and MANY others) will do the analysis for free. It's another "open source" vs "commercial" situation. No one is required to give me expensive stuff for free, BUT if OTHERS are willing to pay the costs to scan, host, and provide it to everyone for free, but the "protectors of our precious heritage" won't allow it, it's no longer about the love of history but more like a repetition of history: we've got the power, so pay us for our, uh, "services".

As for scanning costs, though, Google is already scanning and hosting many of these materials for free and are being actively prevented from doing more (and even showing what they've already scanned), so I suspect we aren't far from having all we need for a 21st Century equivalent of a Carnegie Foundation. If Andrew Carnegie could build public libraries around the world, the robber barons of our century could fund an online Library of Alexandria to serve as a central repository. They could then offer to museums, private collectors, etc., to do the scanning and hosting and maybe even some ad sponsorship or donation mechanism to provide a small income stream back to the owners of the originals.

So basically the universities should buy the books, store them, repair them, scan them, edit the PDFs, make books, pay for all of that and then give you the PDFs for free.

Not sure why they should.

Should scientists and everyone else get papers for free paid by tax payers? Yes.

Universities should support research, they shouldn't only be capitalist machines. But even while chasing profits they could at least sell the damn books online and give you a table of contents.

Agree! It's really ridiculous and quite sad. I am no longer at a University so basically everything I read is stolen via Scihub.

Hopefully, more and more researcher will continue to preprint their work. Something we're trying to help with at https://www.authorea.com (disclosure: I work there)

It's a great site, which documents a horrible situation.

It'd be cool if it also showed which papers have preprints online. And which (I presume virtually all) are available via SciHub. Even links, if possible :)

And a funny thing is that, even though I work in a university, I still find myself using Scihub sometimes. Either we do not have the access (old papers are a problem) or I'm not present at work.

> Imagine how much better science reporting would be if every popular science article was expected to link directly to the full papers they were referencing.

I agree, but for the most part newspapers aren't even getting the "referencing" part right. That is, they just lazily write "a new study" instead of providing an actual, useful reference to the primary source. I don't think this would magically change with open access: Even for non-open works there would usually already be the possibility to link to a public abstract, but most science reporting simply does not care enough.

I think Computer Science is slowly headed in this direction.

I am usually able to find an Arxiv listing for most of the papers that I read.