RE: https://infosec.exchange/@david_chisnall/117540788341445069
I can't remember who I learned this from (maybe @raganwald?) but a good trick for reading computer science papers if you work in industry is
1. Find a bunch of papers on the problem you need to solve.
2. Notice what baseline they're all comparing their new shiny techniques to.
3. Implement that instead.
Update: it was Jay Kreps, probably via Avery Pennarun's blog: https://x.com/jaykreps/status/219977241839411200?a=W6GRXkYLlAZ3X-AlM_Y&s=19 Thanks to @DRMacIver for tracking it down.
@pozorvlak @raganwald @DRMacIver I've often found this approach useful, but usually I prefer the "related work" section of a dissertation, which is usually a very good review of the relevant field at the time of writing, even if the actual original contributions in the dissertation are not of any use to me. By their nature, at the time the dissertation is written, they're unproven, while the related work is the stuff that's notable enough for people to mention it as related work, after all.
