3 ms·
Oh, yes. The long and pompous drivel that scientific papers seem to be padded with these days makes me wonder if they're all busy gaming "paper length" metrics
by bell-cot 4y ago
Oh, yes. The long and pompous drivel that scientific papers seem to be padded with these days makes me wonder if they're all busy gaming "paper length" metrics, playing to an imagined audience of clueless could-be-donor billionaires, or what.
- PaulHoule 4y agoMy A.I. looks at all the cs abstracts posted to arXiv every day, I skim maybe 50 of the abstracts and download maybe 15 PDFs. I am a little shocked at how many papers are fewer than four double-spaced pages, scarcely more than a tweet if you're trying to explain something that is fundamentally new to people who aren't intimately involved in that exact corner of that exact subfield. (If my metric for "quality" was "it might frontpage if posted to HN", the superficial characteristics of the paper would be "about 25 pages with a splashy image on the first page" but unfortunately you can't tell that from the abstract, but maybe you can get the number of pages and the content first page without downloading the whole PDF.)