Counting ArXiv Delays

35 points by fi-le 3 days ago | 11 comments

jszymborski
Just wanted to say that the plots are so nice. They perfectly balance visual noise and information density, _and_ they're aesthetically pleasing.
jazzpush2
Yes, the daly time recently is absolutely ridiculous. Been waiting on a preprint that's already been accepted in a top journal for seemingly no reason.
bagrow
I'd like to see a cumulative version of that power law plot. How many papers had a delay of at least X days.
I have been bit by several inexplicable, long delays recently. In my case, I suspected the reason was choosing stat.ml as primary and cs.lg as cross-list, which, after my time in purgatory, was flipped (without asking me, of course). My choice was completely defensible.
It's almost like peer review all over again!
doc_ick
With the rise of ai, ai-based papers, and ais writing terrible papers; why are you surprised or frustrated?
Wouldn’t you want some kind of barrier against slop?
- fi-le
  If you read the article, you'll find that, surprisingly, the delay I was personally observing likely has nothing to do with AI-assisted submissions.
  doc_ick
  “And here we found the culprit. The paper I was submitting belongs under cs.CY, Computers and Society, as I now know after reviewing all the categories more carefully.” I’d think the indirect cause is ai, given its recent bull/glass shop approach.
  fi-le
  Within the cs.CY category, I only found a slight increase in delays since 2016, that seems to be well within error bars, but it could also be that my sample size was too small.
- gdiamos
  No, I want arxiv to host the paper, not to review the paper.
  I wouldn't want my google drive to start telling me my paper was too sloppy. I just want a link.
  Jaxan
  You can also just put a paper on your website (or google drive). If arXiv isn’t working for you, why submit there?
  Personally, i see no problem with delays, research takes longer than a few days. Reviews take a few weeks or months even.
- tokai
  The growths of scientific papers published has been understood to be exponential since the early 50s.[0] So it really shouldn't be a surprise. Even without AI.
  [0] https://garfield.library.upenn.edu/price/pricequantitativeme...
  doc_ick
  Hmm, I am surprised to hear it’s understood to be exponential, but thanks for the reference. I will read up on it.