We may have reached Peak Bogost: Ian Bogost's GTA6 review in The Atlantic (already a series of words I'm unhappy to have typed) includes a comparison --- favorable --- of his abs to that of the video game protagonist.
@DKThomp Very difficult to answer in tweet length, but:
NYC: Labs are very big SaaS companies except with massive capex. SF: You do not understand what they are building. NYC: We understand SaaS reasonably well, thank you.
This is probably public somewhere && also my confidence is low, but in Windows-1252 there's no repr for an uppercase sharp s (ß), so "ß".upcase => "SS"; a one-byte overflow (of 'S').
The race is on: whoever is first to publish an RCE POC for the Ruby 3.x/4.x 'ß' case-conversion bug gets to come up with the best vulnerability name of the year.
https://t.co/xuMPB4HWyo
@4Dgifts I think discussions in insular groups of subject matter experts are less interesting and valuable than ones happening among all the general software practitioners, who ultimately make all the decisions that matter.
Yes, it's true, Java4Shell is probably not nearly as bad as Log4Shell. So what? That tells you practically nothing. Serverside vulnerabilities are almost never as bad as Log4Shell was.
Because this question is polarized alongside LLMs people are weirdly invested in PHP4Shell and Java4Shell not being dangerous; there's a take (with growing currency) that anything requiring real exploit development is low-risk.
This brought to you by the guy on HN who said the FAA should be running this entire system under emulation so they can reliably reproduce these kinds of failures, because that's how a complex continental distributed system definitely works.