- you are failing to get the point
it's an (maybe the most) extreme example of something which is "gen AI" but not problematic and as such a naive "rule" saying "no gen AI at all" is a pretty bad competition rule design
- they ban any for of gen AI no matter the context
so as an extrema example if you artists used that line smoothing algorithm you game isn't qualified anymore
- > If AI use produces obviously inferior work, how did it win in the first place?
they uses some AI placeholders during development as it can majorly speed up/unblock the dev loop while not really having any ethical issues (as you still hire artists to produce all the final assets) and in some corner case they forgot to replace the place holder
also some of the tooling they might have used might technically count as gen AI, e.g. way before LLM became big I had dabbled a bit in gen AI and there where some decent line work smoothing algorithms and similar with non of the ethical questions. Tools which help removing some dump annoying overhead for artists but don't replace "creative work". But which anyway are technical gen AI...
I think this mainly shows that a blank ban on "gen AI" instead of one of idk. "gen AI used in ways which replaces Artists" is kinda tone deaf/unproductive.
- it's even worse then that
there is a whole basked of technologies which you can label as "gen AI" but which have non of the problems why people hate "gen AI"
as a very dump example, some pretty decent "line smoothing" algorithm are technically gen AI but have non of the ethical issues
- > Factually incorrect.
no please read what I wrote
_local elected leaders_
they are the leaders each member state democratically elected in their own way
and that makes a lot of sense the EU isn't a country after all so using the already democratically elected leaders makes a lot of sense
> They have no authority to draft policies on their own.
yes neither did I claim so, the EU is by far not perfect
> Which evidently there is a lot of!
yes, but that is mainly a reflection of corruption in local Politics
- yes, and here is a fun fact, most of the push for mass surveillance comes from the European Council, the thing is that literally are "just" the locally elected leaders...
not some vague far away "the EU (personalized)" thing
which also mean you can locally enact pressure on them
furthermore the EU supreme court(s) might have more often hindered mass surveillance laws in member states then the council pushing for them...
and if we speak as of "now", not just the UK, but also the US and probably many other states have far more mass surveillance then the EU has "in general".
so year the whole "EU is at fault of everything" sentiment makes little sense. I guess in some cases it's an excuse for people having given up on politics. But given how often EU decisions are severely presented out of context I guess some degree of anti-EU propaganda is in there, too.
- this is simply not true
it was the EU which had stopped many similar unhinged attempts from the UK when the UK was still a member
similar it had been the EU which had shut down various other surveillance nonsense of the EU
you are basically pretending the EU is a person with one uniform opinion and goals
but it's like the opposite of it, like in a lot of way
it's a union of states, each having a vastly different goals and culture and non of them having a "single uniform opinion" either but (in most cases) a more complex political field then the US (on a federal level)
Furthermore the most influential organ of the EU when it comes to making changes is literally a composition of the elected leaders of the member states. So for most big controversial decisions the driving and directing force isn't "the EU" but but the various elected leaders of the member states. For EU citizens blaming "the EU" instead of blaming your own elected leaders is common, but pretty counter productive, as it's basically pretending you have no power to change things.
Furthermore in the EU you have an additional parliament which (in general) needs to ratify laws and two high courts which can (and in context of mass surveillance repeatedly have) shut down misguided "laws", including in many cases local attempts at mass surveillance laws.
So while some parts of the EU have consistently pushed for mass surveillance in recent years other parts also have consistently moved against it.
In general while the EU needs a lot more transparency and some more democratic processes in some aspects a lot (not all) of the "stories told to make the EU look dump/bad" have a lot of important context stripped from that (like e.g. that a lot of the current push for surveillance comes from the locally elected leaders not the EU parliament or some other abstract "the EU" thing, it's your own countries leader/lead party(1) which does or at least tolerates that shit).
- for quite a while I through many of those dump "internal network scanning automatized pentests" where pretty pointless
but after having seen IRL people accidentally overlooking very basic things I now (since a few years) think using them is essential, even through they often suck(1).
(1): Like due to false positives, wrong severity classifications, wrong reasoning for why something is a problem and in generally not doing anything application specific, etc.
I mean who would be so dump to accidentally expose some RCE prone internal testing helper only used for local integration tests on their local network (turns out anyone who uses docker/docker-compose with a port mapping which doesn't explicitly define the interface, i.e. anyone following 99% of docker tutorials...). Or there is no way you forget to set content security policies I mean it's a ticket on the initial project setup or already done in the project template (but then a careless git conflict resolution removed them). etc.
- and assuming you have a practical way to
- verify the attestation
- make sure it means the code they have published is the attested code
- make sure the published code does what it should
- and catch any divergence to this *fast enough* to not cause much damage
....
it's without question better then doing nothing
but it's fundamentally not a perfect solution
but it's very unclear if there even is a perfect solution, I would guess due to the characteristics of phone numbers there isn't a perfect solution
- This article https://signal.org/blog/building-faster-oram/ has some details but is more focused on improving their solution other blogs from the are "we want to build this soon" kind of blogs. It seems that most articles about this topic either have too little content to be of interest or are technology previews/"we maybe will do that" articles about things Signal wants to implement, where it's unclear if they did do that or something similar.
To cut it short they use Intel SGX to create a "trusted environment" (trusted by the app/user) in which the run the contact discovery.
In that trusted environment you then run algorithms similar to other messengers (i.e. you still need to rate limit them as it's possible to iterate _all_ phone numbers which exist).
If working as intended, this is better then what alternatives provide as it doesn't just protect phone numbers from 3rd parties but also from the data center operator and to some degree even signal itself.
But it's not perfect. You can use side channel attacks against Intel SGX and Signal most likely can sneak in ways for them to access things by changing the code, sure people might find this but it's still viable.
In the end what matters is driving up the cost of attacks to a point where they aren't worth in all cases (as in either not worth in general or in there being easier attack vectors e.g. against your phone which also gives them what they want, either way it should be suited for systematic mass surveillance of everyone or even just sub groups like politicians, journalists and similar).
- the problem only affect a subset of HDMI 2.1 features, not HDMI 2.0
but the steam machine isn't really super powerful (fast enough for a lot of games, faster then what a lot of steam customers have, sure. But still no that fast.)
So most of the HDMI 2.1 features it can't use aren't that relevant. Like sure you don't get >60fps@4K but you already need a good amount of FSR to get to 60fps@4k.
- This is a fight you are going to lose.
A better approach would be to put your energy into making sure the used methods are _reasonable_.
We don't require every FSK16 game sail to register the buyers name, age, contact info on physical checkouts etc. In most countries a law requiring that would be seen as excessive and in some places unconstitutional.
Instead it's fine to visually look at a id, and if it "obvious" they are adult (e.g. very old person) we don't even require that. And thats fine. Because we don't need a perfect prevention we just need something which helps parents parent "a bit" and helps "a bit" in cases where parents don't parent.
If everyone fight "all age check solutions" the chance that they get fully ignored and some horrible shit gets passed into law is very high.
If everyone fighting also provides a alternative and strict guidelines about what is and isn't acceptable in their opinion there is a chance for reasonable solutions being implemented instead.
(Like e.g. put a age gate header into http responses, like "min-audience-age: region=US, age=123; region=EU, age=456", say OS must have a API where you pass that in and they say yes/no for that account, do NOT require any crypto, signing etc. This is not fraud prevention but parenting helper. The OS then can store `18+|age` internally and have some integrations with country specific age verification services (it must only store 18+|birthday and only birthday iff <18, I guess for US 21). But there is no need to prevent anyone from changing this value with e.g. windows regestry changes, except if it's a child account. So require any widely _sold_ OS to have a parent controls/child account functionality.
But really any solution which effectively requires mass surveillance, exclude hobby OS or similar, require some clever signing scheme involving device attestation etc. is VERY excessive and unneeded.
- yes I don't mean "special interest"
and I'm aware that people with ADHD don't really have any way to direct it
and that it can easily lead to them neglecting everything from them self, over work to social relationships
so it will help more then it hurts in university
but it still can matter before, even if it's just a parent mistaking a hyper focus on some science topic with a special interest in it and then exposing you to more science related stuff earlier one in life
- What I don't understand (but also wouldn't be surprised about if it is misrepresented by the article) is:
- why would you get a single, for ADHD, non-social-related anxiety, non-sever autism or depression (especially in the later case you probably shouldn't be in a single)
- I mean sure social anxiety, sever autism can be good reasons for a single.
through in general the whole US dorms thing is strange to me (in the EU there are dorms, but optional (in general). And 50%+ of studentsfind housing outside of it (but depends on location). This allows for a lot more individualized living choices.)
- there is so much wrong with the first few paragraphs of this article
1. some of the things they list as "disabilities" are sicknesses which _can_ be disabling but not per see disabilities
2. all of the things listed aren't one/off but have not just huge gradients, but huge variations. You might be afflicted in a way which "disables" you from living a normal live or job but still might be able to handle university due to how it differs.
3. non of the things list is per-see/directly reducing your ability to have deep understanding in a specialized field. ADHD sometimes comes with hyper focus, which if it manifest in the right way can help you in university. It's also might make more "traditionally structured jobs" hardly possible for you and bad luck with how professors handle their courses is more likely to screw you over. Anxiety is often enough more topic specific, e.g. social anxiety. This means it can be disabling for many normal jobs but not affect you in universities which don't require you physical presence, but if they do you basically wait out the course and then learn after being back home. In rare cases it can also help with crunch learning before an exam. Etc. etc.
Actually if we go a step future all of the named health issues can make it more likely for you to end up in high standard universities. Hyper focus on specific topics from ADHD might have started your journey into science even as a child. Anxiety might have lead to you studying more. Since might have been an escape from a painful reality which later lead to you developing depression.
If we consider how high standard universities can cause a lot of stress which can cause an out brake of anxiety or depression in some people it just is another data point why we would expect higher number of health issues (if you lump a bunch of very different issues together like they do).
Later they then also throw in autism in the list of mental issues, even through autism always had been higher represented in academia due to how it sometimes comes with "special interests" and make socializing as a child harder, i.e. it can lead to a child very early and very long term focusing on scientific topics out of their fully own interest. (But it doesn't have to, it can also thoroughly destroy you live to a point "learning to cope with it" isn't possible anymore and you are basically crippled as long as you don't luck out massively with your job and environment.)
Honestly the whole article has a undertone of people with "autism, ADHD, anxiety, depression" shouldn't be "elite" university and any accommodations for them should be cut.
Now to be fair accommodations have to be reasonable and you have to learn to cope with your issues. Idk. how they are handled in the US, but from what I have seen in the EU that is normally the case. E.g. with dyslexia and subtle nerve damage making hand writing harder I could have gotten a slight time extensions for any non-multiple choice exams. I didn't bother because it didn't matter all (but one) exams where done in a way where if you know the topic well you can finish in 60-70% of the time and if you don't even 3x time would not help you much (and the extension was like flat 15min). That is except if my nerve damage or dyslexia where worse then I really would have needed the time, not for solving questions but for writing down answers. There was one exam which tested more if you had crammed in all knowledge then testing understanding, in that exam due to dyslexia and my hands not being able to write quite as fast as normal I actually last some points, not because I didn't know but because I wasn't able to write fast enough.
The point here is if done well people which don't need accommodations shouldn't have a huge benefits even if they get them, but people needing it not getting it can mean punishing them for thing unrelated to actual skills. Live will do so enough after university, no need to force it into universities which should focus on excellence of knowledge and understanding.
- S3 isn't JSON
it's storing a [utf8-string => bytes] mapping with some very minimal metadata. But that can be whatever you want. JSON, CBOR, XML, actual document formats etc.
And it's default encoding for listing, management operations and similar is XML....
> but I feel like we missed an opportunity here for a standardized interface.
except S3 _is_ the de-facto standard interface which most object storage system speaks
but I agree it's kinda a pain
and commonly done partial (both feature wise and partial wrong). E.g. S3 store utf8 strings, not utf8 file paths (like e.g. minio does), that being wrong seems fine but can lead to a lot of problems (not just being incompatible for some applications but also having unexpected perf. characteristics for others) making it only partial S3 compatible. Similar some implementations random features like bulk delete or support `If-Match`/`If-Non-Match` headers can also make them S3 incompatible for some use cases.
So yeah, a new external standard which makes it clear what you should expect to be supported to be standard compatible would be nice.
- ceph depends a lot on your use case
minio was also suited for some smaller use cases (e.g. running a partial S3 compatible storage for integration tests). Ceph isn't really good for it.
But if you ran large minio clusters in production ceph might be a very good alternative.
- the thought that this might be done one recommendation of ChatGPT has me rolling
think about it, with how much bad advice is out there in certain topics it's guaranteed that ChatGPT will promote common bad advice in many cases
this means that for some use cases (early QA, design 3D design tweaks before the final graphic is available etc.) they are fully useless
it's both viable and strongly preferable to track placeholders in some consistent way unrelated to their looks (e.g. have a bool property associated with each placeholder). Or else you might overlook some rarely seen corner cases textures when doing the final cleanup
so no, placeholder don't need to be obvious at all, and like mentioned them looking out of place can be an issues for some usages. Having something resembling the final design is better _iff_ it's cheap to do.
so no they aren't failing, they are succeeding, if you have proper tooling and don't rely on a crutch like "I will surely notice them because they look bad"