he’s still got a big heart on the inside

it was about letting the students collectively set their own punishments and self-govern their behavior to shut down the possibility of authoritarian abuse. anarchists have been exploring this idea for centuries

no it’s the way to put authoritarians out of power

it’s basically anarchism actually

get told they are supposed to be the better than the others

better than those not at Princeton, not better than others at Princeton nor to the point of sabotage. in fact anecdotally they get told that even though they were probably top of their high school class they would be average or even low-performing, and will have Princeton’s full support in managing the transition, especially the mental preparation for “mediocrity”

a culture of immense pressure and a zero sum game where collaboration between students is unthinkable

making sure your friends are working honestly doesn’t have to be competition. i don’t like the implication that collaboration has to be conspiring to cheat against some authoritarian figure, instead of making sure all of your friends succeed.

yeah…

well to be fair, unruffled hasn’t answered mine either

but rimu’s a bit of a surprise to me, probably since i’ve never really “known” (seen) them much before other than “python dev, of piefed, making a third popular activitypub threadiverse client with lots of ambitious features”

still, i think some of the things directed that way in this thread are quite unfair

i’m interested in seeing data of only bans (preferably just sitebans, but if that’s not possible all their bans would be fine) by admins. commag moderation style has a lot more leeway

yeah that’s exactly why i don’t think the attitude score helps

that’s how it is by default for some reason. i haven’t added to me dictionary either and that’s how it shows in libreoffice calc. it seems like it just doesn’t squiggly any “word” with numbers in it.

flippant antagonism is the worst thing against discourse on the fediverse and we can do our part

i disagree

and note that I opposed the defederation, yet i still disagree

i don’t see how that has bearing on what i said lol. i argued about not creating many new communities and not closing inactive communities. the last new community i heard of were the two or three replacements for feddit, and just the 41st top community (41st since I estimate the photon pager has 40 items per page by default, and i clicked the top of page 2) has not have posts in over a month, let alone mod actions.

i’d say it’s more accusations of being a troll overall—original post is https://lemmy.dbzer0.com/post/67198913

as devil’s advocate i don’t think account karma should be a factor. downvotes would only be an extension of how exclusionary the instance is. dbzer0’s moderation is quite democratic, after all

the decision to graph per-community (i’ll call those "commag"s) doesn’t make sense to me. it seems to me like a bad approximation for lifetime total users that doesn’t control for either instance attitudes on commag creation or troll account registration (these usually don’t create communities). i’m unacquainted with pawb but i wouldn’t think they’re very ban-happy, and the fact that these graphs show them as twice as ban-happy per-community as, and a bit more ban happy than ml should tell you this isn’t very good methodology.

it seems more like dbzer0 and pawb don’t create many communities (but don’t close stale ones either, unlike .world), which makes sense as the ones that do exist are quite focused and targetted to the instance userbase.

Another researcher, Davi Ottenheimer, pointed out that the security section (Section 3, pages 47-53) of Anthropic’s 244-page documentation “contains no count of zero-days at all. With no CVE list, no CVSS distribution, no severity bucket, no disclosure timeline, no vendor-confirmed-novel table, no false-positive rate.”

excerpts from the summary of the post linked in “Devanash ultimately concluded”, a lot of which Register repeats (which I think is a good thing since the copyediting makes the language a lot more accessible and wide-reaching and of course it was credited):

The bugs are real. 17-year-old FreeBSD RCE, 23-year-old Linux kernel heap overflow, 27-year-old OpenBSD TCP flaw. LLMs catch these because they can reason about the gap between what code does and what the developer intended. Fuzzers and static analysis literally cannot do this.

The coverage is wrong on almost every detail. The “181 Firefox exploits” ran with the browser sandbox ( yes, the thing that stops browser exploits) off. The FreeBSD exploit transcript shows substantial human guidance, not autonomy. The “thousands of severe vulnerabilities” extrapolates from 198 manually reviewed reports. The Linux kernel bug was found by Opus 4.6, the public model, not Mythos.

The moat is thinner than anyone reported. AISLE tested eight models including a 3.6B model at $0.11/M tokens. All eight found the FreeBSD bug. Mythos’s actual lead is in multi-step exploit development, not detection. That’s a narrower and more replicable advantage than what’s being sold.

how is loops going?

they do?

so did they?

hourly CN air quality map, unit is µg/m³: https://www.air-level.com/

reuters says “The WHO considers PM2.5 concentrations above 50 micrograms per cubic metre “severe” air pollution.” but I couldn’t actually find that in WHO global air quality guidelines or anywhere else. china considers >500 µg/m³ “severe” (严重) and everything on the map would be below that

well you no longer have to do the off-label calculation