5 ms·

Completely and utterly deserved. Arxiv, like Wikipedia and arguably sci-hub, is an absolute boon to humanity. What I find interesting is that its endorsement sy

Completely and utterly deserved. Arxiv, like Wikipedia and arguably sci-hub, is an absolute boon to humanity. What I find interesting is that its endorsement system does largely work at making it very easy to submit papers whilst reducing the amount of spam. Sure, there are some papers that are not completely brilliant -- or indeed, actually any good at all -- but they have the same "right" to be read as their huge CERN brethren. In some fields of physics, paid journal publishing is really a niche in comparison to stuff on the ArXiv. In others, and other areas of science, not quite so much yet. I also very much like the fact that it is open, mostly LaTeX based (machine-readable maths!) and free.

quickthrower2 5y ago
Wikipedia is one of my refuges from the where the rest of the web seems to be going. It’s one of the few that don’t cause me to hit reader view (mobile safaris defacto ad blocker) immediately.
- janto 5y ago
  Unfortunately Wikipedia has also been heading in a disappointing direction. I now expect everything on there to be filtered through a specific political perspective. Deleting information is much easier than adding, so pesky facts are unlikely to persist.
  - Kye 5y ago
    It does seem that many pages get claimed by one editor or another who sees it as their domain, but I haven't seen any study on the politics of these wiki fiefdoms. Most I see are conservative, sometimes extreme.
  - danielheath 5y ago
    Somewhat true of the main pages; if you want to know whether something controversial is being left out, check the talk page.
  - prionassembly 5y ago
    The Wikipedia process is still pretty good last time I checked? There was a notability challenge in en.wikipedia on Olavo de Carvalho, a Steve Bannon type who wrote some philosophy books (like -- on Aristotle and Epicurus) as a younger man (and keeps getting his "philosopher" self-nomination challenged because he didn't go and get a college major in philosophy). I was for keeping the page, but on balance I think the process, arguments and result were fair -- even if almost certainly carrying some underlying political motivation. We have to focus on the process. It's the only way to general axiology.
    - janto 5y ago
      I don't have faith in the process anymore. Wiki-lawyering to remove content, is too effective. To what end? Saving disk space? I'm no longer motivated to fight deletionism in my domain. I don't want to spend time that is not appreciated. I suspect that as soon as a page is tagged for deletion by someone, others that enjoy being the arbiters of what is and is not WP, flock there. If they lack interest in the topic, they will vote to remove with many technical reasons. Statements that others do find it relevant or interesting, are simply ignored.
  - alithejumbo 5y ago
    I second this opinion about Wikipedia.
sam_lowry_ 5y ago
Yep. the next Einstein award should go to Alexandra Elbakyan.
- robot0x13 5y ago
  and Aaron Swartz
Vetch 5y ago
> and arguably sci-hub Inarguably, not arguably. In my mind there is a clear separation between pre and post sci-hub. The stark increase in experienced friction when accessing scientific knowledge since sci-hub's pausing has highlighted its profound value. There's also libgen, the closest thing to a bastion of all written human knowledge. It'd be fun to read a short story about an AI trained on libgen data that makes significant contributions to fusion, longevity and cancer but whose creators are ineligible for any prizes and are jailed for life for multiple counts of copyright violation.
- nefitty 5y ago
  On one of huggingface's explanation blurb of a training set, they half-jokingly speculate that GPT was trained on libgen. Apparently Google won't reveal what book data it used...
  - harias 5y ago
    Google books probably contains as much information as Libgen. It's simply doesn't allow public access.
photochemsyn 5y ago
Wikipedia has some major issues for real research and there are good reasons why universities don't allow students to include wikipedia links as citations in their papers. The information there simply isn't curated properly. A main gripe is that more often than not, the supporting links for claims made in wikipedia articles are broken or of poor quality. Another is the tendency of ideologues to remove anything they disagree with in their particular domain. It's useful for finding trivial information (flag of Botswana, say) but otherwise I usually block wikipedia from search results.
- duskwuff 5y ago
  > there are good reasons why universities don't allow students to include wikipedia links as citations in their papers That reason is because Wikipedia is a tertiary source. It's too far removed from a source of information to be appropriate to cite. It has nothing to do with the reliability of the site; it's basically never appropriate to cite any kind of encyclopedia. If there's a specific piece of information you found on Wikipedia, you chase down Wikipedia's source for that information and cite that. If you're using Wikipedia as a general reference, you don't need to cite that at all.

quickthrower2 5y ago

Wikipedia is one of my refuges from the where the rest of the web seems to be going. It’s one of the few that don’t cause me to hit reader view (mobile safaris defacto ad blocker) immediately.

janto 5y ago
Unfortunately Wikipedia has also been heading in a disappointing direction. I now expect everything on there to be filtered through a specific political perspective. Deleting information is much easier than adding, so pesky facts are unlikely to persist.
- Kye 5y ago
  It does seem that many pages get claimed by one editor or another who sees it as their domain, but I haven't seen any study on the politics of these wiki fiefdoms. Most I see are conservative, sometimes extreme.
- danielheath 5y ago
  Somewhat true of the main pages; if you want to know whether something controversial is being left out, check the talk page.
- prionassembly 5y ago
  The Wikipedia process is still pretty good last time I checked? There was a notability challenge in en.wikipedia on Olavo de Carvalho, a Steve Bannon type who wrote some philosophy books (like -- on Aristotle and Epicurus) as a younger man (and keeps getting his "philosopher" self-nomination challenged because he didn't go and get a college major in philosophy). I was for keeping the page, but on balance I think the process, arguments and result were fair -- even if almost certainly carrying some underlying political motivation. We have to focus on the process. It's the only way to general axiology.
  - janto 5y ago
    I don't have faith in the process anymore. Wiki-lawyering to remove content, is too effective. To what end? Saving disk space? I'm no longer motivated to fight deletionism in my domain. I don't want to spend time that is not appreciated. I suspect that as soon as a page is tagged for deletion by someone, others that enjoy being the arbiters of what is and is not WP, flock there. If they lack interest in the topic, they will vote to remove with many technical reasons. Statements that others do find it relevant or interesting, are simply ignored.
- alithejumbo 5y ago
  I second this opinion about Wikipedia.

sam_lowry_ 5y ago

Yep. the next Einstein award should go to Alexandra Elbakyan.

robot0x13 5y ago
and Aaron Swartz

Vetch 5y ago

> and arguably sci-hub Inarguably, not arguably. In my mind there is a clear separation between pre and post sci-hub. The stark increase in experienced friction when accessing scientific knowledge since sci-hub's pausing has highlighted its profound value. There's also libgen, the closest thing to a bastion of all written human knowledge. It'd be fun to read a short story about an AI trained on libgen data that makes significant contributions to fusion, longevity and cancer but whose creators are ineligible for any prizes and are jailed for life for multiple counts of copyright violation.

nefitty 5y ago
On one of huggingface's explanation blurb of a training set, they half-jokingly speculate that GPT was trained on libgen. Apparently Google won't reveal what book data it used...
- harias 5y ago
  Google books probably contains as much information as Libgen. It's simply doesn't allow public access.

photochemsyn 5y ago

Wikipedia has some major issues for real research and there are good reasons why universities don't allow students to include wikipedia links as citations in their papers. The information there simply isn't curated properly. A main gripe is that more often than not, the supporting links for claims made in wikipedia articles are broken or of poor quality. Another is the tendency of ideologues to remove anything they disagree with in their particular domain. It's useful for finding trivial information (flag of Botswana, say) but otherwise I usually block wikipedia from search results.

duskwuff 5y ago
> there are good reasons why universities don't allow students to include wikipedia links as citations in their papers That reason is because Wikipedia is a tertiary source. It's too far removed from a source of information to be appropriate to cite. It has nothing to do with the reliability of the site; it's basically never appropriate to cite any kind of encyclopedia. If there's a specific piece of information you found on Wikipedia, you chase down Wikipedia's source for that information and cite that. If you're using Wikipedia as a general reference, you don't need to cite that at all.