Contemporary academia especially in the West has a massive surplus of staff.
Many people pursue academic careers solely for a comfortable lifestyle, doing minimal or even no research for long period of time. With extra lack of oversight that allows researchers to isolate themselves they create circles which cover each other.
Occasionally, folks outside of the circle come in and they start finding ton of fraud in the research with multiple big cases in past few years on top universities like Harvard for example.
"Many people pursue academic careers solely for a comfortable lifestyle, doing minimal or even no research for long period of time. With extra lack of oversight that allows researchers to isolate themselves they create circles which cover each other."
I want what you're smoking because that might be one of the biggest fabrications I've heard in a long time.
lol I have never worked as hard as when I worked for an R1 university. My big tech job is way more like a resort vacation, complete with snacks and drinks.
I found 2. One is an ongoing lawsuit and the other seems mostly like stupidity because people keep falling for the stupid AI grift. I can barely trust AI to produce basic boilerplate and they are trying to verify novel research with it?
Thanks to ireland all big US corporations saved hundreds of billions of dollars past few years so now they can get back to US with this massive cash for anything they want (ofc. nothing will get back to EU as long as they ignore tax heavens)
Not sure if this is true, he got "lucky" with technological advantage of warfare (Mongol bows) compared to other nations close to him as horse archers were literally "meta" to fight vs heavy/peasant infantry same case as Crassus fighting Parthians.
You mean you only need better bows to conquer the world? What about millions of warriors already trained in battles and already winning against the biggest and most competent military force in that world (China) after hundreds of years iterated tactics and strategies?
Not really true. Horse archers were a thing before him and after him. And many of the people he thought were either horse archers themselves, or allied with horse archers, or had fought horse archers for centuries.
Yes and no Mongol Bow was a thing that was just a lot better for this era compared to rest of Eurasia especially with combination with mongols tactics and rest of the regions didn't fight horse archers much for centuries but ofc. you can disagree.
I think its up for debate how much their bows were compared to other steppe people. Since we don't know exactly how the bow making changed across the centuries.
Even if we would accept that their bows were better then some other steppe peoples, he still thought many people that for sure had the same bows as he did.
In fact, most of his life, he thought people who had the same technology, including the invasion of the Qara Khitai.
Then the Jin had allies that likely had the same bow technology. However it seems most of those allies just switched sides.
Only the Khwarazmian Empire and Xia likely didn't have significant troops armed the same way.
If someone is interested in Byzantium fall and why this war was so bad for both empires, read some more about Justinian's Plague which killed ~35-50% of population and also halved economical output. It took about 200 years to get to the same place population wise for most of the empire.
Weirdly it didn't hit Persia as much outside of Mesopotamia, most historians estimate "only" ~20-30% of population died and shifted balance of power to Persian side, from almost renewed Roman Empire at 540 which most likely was getting back to ruling mediterranean world once again.
It should be noted, though, that by the time of Basil II in the 900s, the Eastern Roman Empire (aka Byzantium - a name neither they nor anyone else uses until the 19th century) had again become the most powerful military actor in the western world.
And then they got rekt in Manzikert, 1071 AD. 46 years after Basil II's death.
Let's not kid ourselves, the Byzantines had far more underlying problems throughout its history and only mostly survived by a.) the walls of Theodosius, and b.) bribing away potential invaders and getting others to fight their wars.
Generally battles like Manzikert or similar events were only as disastrous because they were inevitably followed by internal chaos and civil wars.
The free for all nature of the succession meant that whenever the empire was facing major difficulties (i.e. because the emperor was incompetent or due to factors he couldn’t really control) every general or noble within the eyesight of the throne thought that he should have it.
Turkish gains after Manzikert itself were limited and much of the imperial army survived. Of course it marched straight back to Constantinople for its commander to appoint a new emperor (since Romanos IV was captured by the Turks).
Then the Romans kept fighting each other for years while entire Anatolia was gradually lost.
Manzikert was a disaster but they certainly had enough resources to bounce back (i.e. it was certainly not worse than Cannea)
1204 also started as coup attempt by the Venetians to put the son of a former overthrown emperor on the thrown.
Exactly! Manzikert was barely a disaster in terms of numbers. Half of Romanos's army was not even on the field. They simply didn't come to his aid, most likely because they were led by a Doukas who wanted one of their own on the throne.
Byzantine history is full of such examples. Civil wars that last decades, never-ending internal strife, and the ever present threat from steppe tribes and caliphates. It's a wonder they survived as long as they did.
That's precisely what I meant by underlying problems. New emperor? Let's revolt. Emperor captured? Free for all king of the hill. Old emperor returns? Let's blind him, imprison him, and then free for all king of the hill.
Yes, Basil II had the unfortunate legacy of not leaving an heir at a very bad time. The Romans of the Eastern Empire had their problems but mostly just a continuation of the old Empire: dynastic jealousy and external enemies. As enfeebled as their last 300 years were, they were pretty unique in having lasted almost 2 millennium.
Late Mary Boyce attributes lower outbreaks of pandemics among Zoroastrian communities to the strict purity laws and practices. For example, it was unacceptable to pollute water, considered one of the sacred creations of Ahuramazda. To avoid polluting water, Zoroastrians would not wash their hands directly in rivers or bodies of water. There are hundreds of examples of these strict purity laws. [1]
The Plague of Justinian was also the first recorded major outbreak of bubonic plague. Black death seems to be the pivot-point at a lot of important moments in history.
cloud gaming is good if you live close to the servers and don't care about graphics, but playing with +60-100ms for every action feels very bad. It almost feels like playing on 15-20 fps PC and quality of streaming video is always a problem compared to native quality maybe AV1 will fix it.
7ms latency, 4k120fps with geforce now. 10ms on wifi. I'm not kidding.
It's ALMOST perfect. I play BF1 through it. Try it once (I believe they still have the "free for 1hr per session, infinite sessions"? That's what sold it to me).
I can play very intensive games (graphically) on my macbook on the couch. It's amazing, and I couldn't believe the 10ms on wifi. It's mind-blowing.
BUT I live near Amsterdam, where a server cluster is.
Also, about the graphics: I'm borrowing a 4080 every time. Everything is on max. If you're in a very (very) hard scene for compression, then yeah, you'll see (very little) artifacts. But I run it on 75mbit, and that's a LOT.
Depends on the game. I think I'm more sensitive to latency (less able to compensate) than most people. I couldn't enjoy playing Titanfall until I put my Samsung TV in game mode; I would just get hit and couldn't do anything about it playing League of Legends on my gaming laptop with a 4K monitor, but when I hooked up an external monitor, mirrored the screen, and ran a clock, I took photos showing my laptops' screen was behind by 30 ms. I started playing on an external monitor and started to win. I even found I had a hard time with some 1 player games such as Sword Art Online: Fatal Bullet if I didn't run in game mode.
On the other hand, I went through a phase where I did a lot of streaming from my PC to a NVIDIA Shield and an XBOX. Sometimes through wired Ethernet, something through an airMAX microwave link to my other house. Games like Persona 5 and Orcs Must Die 3 were just fine, but I could not play any Rhythm games, which I have a knack for, High-Fi Rush was no fun at all.
My kid plays fortnite using home streaming to an xbox, and says he doesn't notice the latency. I do the same on an Asus ROG Ally, and it's "good enough". I am not a competitive FPV player, but suffer from OCD and notice latency and it tweaks me hard.
I'm playing single player games via Parsec and the latency feels fine. Moonlight is tolerable but Steam streaming feels terrible for some strange reason. This is running two Wifi 6 devices so nothing is even wired. I often use a controller connected to my laptop, or even better use the wireless controller, connect that to the physical device then you bypass the controller latency and only the video has a lag, which is kind of a neat trick if you're close enough to the computer you're streaming from.
The only sorts of games I can't play are things like Binding of Isaac that are super dependent on reaction speeds, but even games like Elden Ring feel fine.
Sunshine on the server and moonlight on the client blows steam link out of the water in terms of latency. Even on my home network with everything on ethernet, steam link would stutter. I sometimes forget I am not directly connected to a computer while on the couch.
Did clicking on the Rent my PC tab really try to benchmark my GPU through my browser, or did I accidentally click another button on that page inadvertently that triggered that?
If the former, that's a terrible idea. If the latter, that button really needs a confirmation and explanation of what's about to happen.
I'm viewing on an Intel Mac and it hung my entire computer for like 15 seconds. I didn't even connect that it was related to viewing your site until I got the error at the end and everything unfroze.
It does. Sorry about the experience, we will try to improve it.
Having user confirm it is not a good option, because every click is a hassle.
What we could do is first run a very short version of a smaller benchmark, and if that takes too long, don't run the main one. Then the worst case you will have a 100ms lag at this point, which is way better than 5 seconds of reading.
It's a neat feature, I just think it'd be ideal to do a confirmation first. It wouldn't be a great experience if it happened on mobile, either.
Every click is a hassle, but principal of least astonishment applies here. Literally not a soul will be expecting that to happen when casually browsing your site.
What does utilization look like? I would be interested in running this on a spare machine but it's not clear how large the potential audience of renters may be in my area.
Right now the utilization is low (< 10%), but in the effort to prop the providers side the company is footing the bill and paying for availability approximately 50% of what the benchmark on the page tells you. This is a rather common strategy for bootstrapping any two-sided market.
If you wirte 100x faster code you could probably automate almost all of it away as it seems to be super trivial and already resolved problems.
I also use Claude for my coding a lot, but i’m not sure about real gains if it even give me noticeable speed improvement as i’m in a loop of “waiting” and fixing bugs that LLM made still it is super useful for writing big doc strings and smaller tests maybe if i focus on some basic tasks like classical backend or frontend it’ll be more useful
DEI was so good and righteous idea that it shifted most of the western world to the right more than dozens of islamist terrorist attacks in previous decades.
And ofc. all US corporations would follow this cultural change or they’ll risk antitrust/tax/law investigations from Trump and all EU government that’ll be elected in next few years.
Didn't the exact same companies introduced the exact same changes they now revert not out of firm belief but to demonstrate support? If someone adapts to the environment out of self-preservation then they were never supportive of any cause.
I wish people would stop humanizing big companies.
Inflation caused by the pandemic era policies upturned governments more that any acronym. The U.K. went the opposite of right, for instance. And Orban in Hungary might finally get the boot this year because of economic doldrums.
UK also is moving to the right reform + conservatives are pooling at around 45-60% atm. Labor is only in power cause reform and conservatives split votes
Only place that was in reverse of this trends was Poland in 2022 which shit to the center-right from “populist right” and possibly Hungary but we don’t know yet as Orban was losing pools for like 10 years but never lost election.
If economy is good people won’t care about politics when economy is bad folks start getting political and we see it in whole west France, Germany and Canada are without government and most folks in this countries are shifting more and more right.
The UK stayed basically the same and I think that was due to the fact that the previous government was already the centre-right wing party–and although they tried to move further right–the fact that they had done such a terrible job in government basically meant their goose was cooked no matter how many punitive policies they tried to enact or wars on wokeness they prosecuted.
The right-wing party of the traditional two (the tories) was absolutely slaughtered in the election and they've now elected as the party leader (although not without some controversy–apparently tactical voting gone wrong) a fairly standard right-wing populist in Badenoch.
There is also the increasingly substantial spectre of the far-right, with race riots taking place over summer and Musk's pet favourite Reform party gaining ground.
It seems very possible to me right now that at the next general election the UK will swing seriously to the right. If this does happen it will the fault of Labour as far as I'm concerned. Of course that's still a long way away and we can always live in hope.
Trump's administration proved ineffective last time and there's little reason to think it will be different this time. If he tries to sabotage companies for not abandoning DEI initiatives he'll be bogged down in the courts, it will take more than four years to resolve any of that and by then it will be moot anyway.
Insofar as companies are changing now it's because they no longer feel political pressure to put on performative displays to satisfy the sort of activist government regulators that were typical in the Obama and Biden governments. Companies like Apple which are true DEI believers and not merely bending to political pressure will keep their programs dispute Trump and Trump won't be able to much of anything about that.
>DEI was so good and righteous idea that it shifted most of the western world to the right more than dozens of islamist terrorist attacks in previous decades.
If by "shifting right" you mean people in the US voted for Trump because of DEI, the numbers don't really add up. Biden beat Trump by 7 million votes in 2020. Trump beat Harris by half that number in 2024. More than that, the Republicans ran the same anti-woke campaign in 2022 and yet the Dems still gained senate and governor seats then. It's not like DEI was any better in 2022 for that narrative to make sense.
There was no major cultural shift. Trump won because of the economy not DEI. Hell, I would wager half the democrats who worked in STEM fields thought DEI was counterproductive from the beginning and they still didn't vote for Trump because of it (e.g., pg). It's been debated around here since the days of that google engineer getting fired for his anti non-merit hiring criteria memo and long before that.
In AOC's district, the same people voted for AOC and Trump. People said "fuck this, I gotta put bread on the table. something's gotta change."
Can you share all the sources linking DEI policies to the right-wing shift? Because to me it looks more like a scapegoat post-factum where right-wingers tied themselves to the idea of DEI to fan the flames rather than it being a causal point. It's just another selling point for the hatred-fueled base of right-wingers, not too different from "immigrants are taking your jobs" or "women in the workforce are the cause of unaffordable housing/less well paid jobs", etc.
It's just a mirage to manipulate dissatisfied people, classic populist move.
GP has no obligation to do a literature search for you. This is a discussion site, not a scientific journal, so it is okay to have opinions based on other sources, such as anecdotes from personal life, first principles, or common sense/intuition.
Speaking of which, I see the same anecdotes that DEI policies pushed people to the right.
It's absolutely okay if it's framed that way in their original comment, which it wasn't and deserves push back unless they can sustain their argument. If it started with "in my opinion", "in my view" instead of a baseless factual statement.
If not framed in that way then I will ask for sourcing the bullshit :)
> Speaking of which, I see the same anecdotes that DEI policies pushed people to the right.
Yeah, that's why it's called "reactionaries", I'm sad you live among those people though.
It's okay to ask of course. My statement was that the ask does not have to be reciprocated.
Personally, unless there's specific claims about data proving something, I usually assume claims on discussion sites to be somewhat informed personal opinions, even if they are worded more strongly than this. I find this to be true more often than not. But I may be more skeptical than your average person regarding the scientific rigor in social sciences, as well as the ability of people to cherry pick specific papers that suit their claim, and therefore place less weight on literature citations versus experiences that I see in my daily life.
WebGPU don't have access to most accelerators in GPU so it would be super slow compared to CUDNN version but pure CUDA would be only x times times faster.
Many people pursue academic careers solely for a comfortable lifestyle, doing minimal or even no research for long period of time. With extra lack of oversight that allows researchers to isolate themselves they create circles which cover each other.
Occasionally, folks outside of the circle come in and they start finding ton of fraud in the research with multiple big cases in past few years on top universities like Harvard for example.