3 ms·
You install Telegram, you contact the bot https://t.me/scihubbot https://t.me/scihubbot, you send him a DOI and it will answer with the PDF (if it has it). Som
by giomasce 9y ago
You install Telegram, you contact the bot https://t.me/scihubbot https://t.me/scihubbot, you send him a DOI and it will answer with the PDF (if it has it).
Sometimes it says (in Russian) there is an error fetching the paper: I understand that in such a case it does not have the paper, but the paper is available on GenLib, so you can go there. But it is not frequent.
- dredmorbius 9y agoThat sounds as if it's adding another layer (and possible failure mode) between you and Sci-Hub. How does the bot itself access the content? What keeps it from being attacked as Sci-Hub has, by publishers?
- giomasce 9y agoSciHub is unfortunately rather opaque on its internals (and that is definitely something that I would like to see changed), but it seems that the Telegram bot is blessed by the official Twitter account. In practice, I have often found it more reliable than the website, and it does not have problems with finding the right TLD. Also, as I said, it retains the history and it shared is with other devices, provided that you consider that a plus.
- dredmorbius 9y agoSciHub has operated at a set of IPs for most or all of its existence, certainly the past several years so far as I'm aware. So long as you resolve it by hostname at those IPs, by whatever means, you'll be able to reach it there. If the IPs themselves are accessible. If Telegram works for you, great. Meantime, if some school or research organisation that cannot afford to pay Western prices for academic content cares to set up their own networking infrastructure to provide access similar to the methods I've suggested, also great.