[impendia]

> But publishers of subscription journals insist that such views are misguided — born of a failure to appreciate the value they add to the papers they publish, and to the research community as a whole.

Okay, let's hear it:

(crickets)

Commercial publishers offer one and only one thing: the journals' names. I want to publish in Journal fur die reine und angewandte Mathematik and Inventiones mathematicae and lots of other journals with names every mathematician has heard of (and some of which are even in English!) I want the warm glow of having my ego thus flattered, and tenure committees are looking for the same thing too.

But nobody will actually go and read the journal. Why bother? If you want to actually read the paper, that's what the arXiv and my personal webpage are for.

[toufka]

And this used to matter when you subscribed to a few journals and they arrived at your home or workplace. Those that did were your sources of even access to the information.

Now we have this thing called 'google' - where if I want to find something out, I can find it. I can find it regardless of it being published on my personal blog, or on Nature's website.

[mjn]

For middle-tier journals I agree. But people actually do check the latest issues of the top-tier journals. Journalists in particular check those almost exclusively. If your article is coming out in the newest Science, there are a lot of people who will see it when the table of contents comes out.

[impendia]

Okay, fair enough.

But not really in math. By the time anything appears in the Annals of Mathematics, which is the top journal, people will have already heard about it.