raducu
All knowledge work faces the same problem.
The grapes are sour.
Russia is absolutely in a hurry because the war is hurting it tremendously, it simply is unable to do so at the moment, but it hopes it can undermine Ukraine's external support, which is indeed going much better than its campaign on the ukrainian front.
I don't think that's the winning end game for Ukraine, but I also don't see there being a realistic winning end game.
One scenario would be that Russia f..ks of with that 20% taken and calls it a success (though strategically it is an enormous blunder), like they did in Finland and leaves Ukraine alone after that -- but is this realistic to happen?
Or Russia collapses internally and Ukraine does take back all the theritory.
Or the world steps in and secures a lasting peace for Ukraine.
Or Russia realizes Ukraine will outlast it because of external support and realizes how stupid and lose-lose this war is.
Or Ukraine collapses and Russia takes an even larger chunk and installs a puppet regime in Kiev.
But Russia seems unwilling to fuck off and call it quits because it's backed by China and has extended its influence in the USA and EU so it bets it can outlast Ukraine.
They can blanket band USING chinese models for all US/EU companies -- that's what they really care, not about chinese companies using chinese models.
However, let's talk realpolitik.
Ukraine innovates and Russia copies/addapts. Russia has ALWAYS been good at this game. Russia reminds me so much of the roman military in this sense, extremely good at copying and incorporating foreign military tactics/equipment.
Think of the initial long convoys ukraine destroyed, the bayraktars -- Russia addapted, the then the brilliant Eastern campaign where Ukraine managed to route the russians -- Russia adapted, that stopped working, then the indiscriminate massive artilery strikes Russia was doing -- ukraine started using better western artilery and himars, stopped the russians but the russians also addapted spread their logistics and so on; then the drones -- russia copied those even innovated with fiberoptic drones, then the Shaheds, ukraine adapted with fpv interceptors, then Russia started using jet powered interceptors, balistic missiles -- Ukraine initially stunned the world with Patriot interceptors but Russia kept firing until Ukraine ran out of interceptors, then the USV -- Ukraine had and still has great success with them, but look, Russia started using USVs also and Ukraine has no response to those.
It looks like the end will be decided by the country who just has more resources and willpower to continue.
I'm seriously scared for the next years -- the US is clearly turning hostile to Ukraine, Europe is turing far right with russian puppets everywhere.
They can pay the model provider, not amazon, but it would be less feasible because for now we have real competition among them.
History often rhimes but it doesn't always repeat itself.
I think there is genuinely a need for shopping agents and they could transform shopping in very good ways for the consumers.
Good thing for shoppers, bad for marketers, bad for people who relied on smarts, bad for margins, good for people who have capital and assets.
Then why not store the oil there directly instead of going all the trouble of pumping it in and out of the caverns? Is it so it doesn't rain on it?
Doing an angular -> react refactor with Opus 5 at work.
Amazed at how good it is but also amazed at how dumb it is after I told it I want a very basic and specific code review:do the file names match the contents? is one functionality/concern split in multiple files? do the variable names match what they do? does a method read nicely?
Anyway, I was very surprised at how bad it can be at this.
I suspect it's because the LLM simply does not focus well enough on a single method.
My gut modelling is that the critical files are quite small, some 4k LOC and the LLM tries to read them all in one go and my requirement is surprisingly difficult but the LLM does not realize it, loads all the input into context and simply cannot find a good signal per token/group of tokens.
I guess I can do it because I literally cannot fit 4k LOC in my head and I actually read only tens of lines but I can reeealy dig deep in those lines.
The LLM can remember lots, it has 1M context window, was trained with needle in the haystack problems, but it can severely underestimate how much computation is required with TASTE/universal computation type tasks.
Yes, LLMs are incredible when it comes to delivering well defined outcomes, it's evident they were RL trained very well. But they are incredibly bad at understanding nuance in a text.
It's incredible how agentic coding benchmaxing seems to correlate with writing good poetry but at the same time finding good names for files and variables and wiring beautiful code does not.
Same with getting basic details of who said what in an "adversarial" conversation between 2 humans -- LLMs are incredibly bad at it, with all their "almost discovering new maths" skills
Producing beautiful, coherent, maintainable code. RL trained the models to produce correct code but boy oh boy is it ugly if it's 10k+ LOC
it has to be some amazing router and while the models are open-weights, the knowhow to run them efficiently surely is not?
I presume you can't use deepseek?
There could be structural reason as to why it's not possible, not pumping water *
Some guys in a mediterranean country are drooling right now...
By energy you mean electricity,the nuclear, solar and hydro-- the kind China has installed new 2TW capacity over the last decade while US installed 0.2TW capacity in the same time?
I've been wrong before. However, when was the last time this business model made sense -- that facebook, SpaceX and others, all just pivot from their market niche to general purpose AI datacenter providers.
How on Earth does this make sense?
What happens in a few years when DeepSeek runs on the chinese chips like the Huawei Ascend at a fraction of the cost ?
These are all very high value added companies going into comodity AI hosting and they're all going to make a killing?
This is the kind of thinking that points to that reckoning moment being close.
This and people buying stock with borrowed money*
I mean let's look at bitcoin/crypto.
My experience as IT in modern banks was the opposite. The legal department were absolute assholes when it came to software features. And I'm talking absolutely 100% ok features, like paying your bills from the banking application.
The least fun, trigger happy, cover their buts people I've ever seen.
Like all they could ever say was NO. I guess they were heavily incentivized to just say NO to everything.
I give LLMs snippets of text messages exchanges with my wife and I can't believe how dumb the LLMs are of getting basic facts right let alone nuance.
I'm 100% not one those "LLMs are just stochastic parrots" people, but coding and coding-like activities are extraordinarily well fit for LLMs, but for things that there's less training data, LLMs probably do a lot worse
noob question here from someone who ony played a bit with iceberg and trino: what's the reason to do the analytics stil inside the postgres -- is it so that you don't eat up the IOPS/bandwidth of the main postgresql disks?
I did use pure nicotine and it was very bad for my health, probably due to high dosage, but still.
I've used heavy stimulants, benzos, opioids, dissociatives without an issue, but nicotine is in a class of its own in terms of how insidiously addictive it is.
But just from a health point of view: extreme arm and hand joint issues, forearm vascular issues that made my hands numb at night, palpitations/arrithmia like I was about to die when I used nicotine before sleep and I was drifting to sleep -- it really felt like I was about to die, like my heart was mangled up.
I've said it before: if you allow people to see how much others spent, they will try to climb up the "leaderboard".
It takes just ONE little praise for using tokens or one perk gained, and the GAME IS ON among the developers!
Only if you don't have psychological issues, then none of that really works.
I've done all the sleep hygiene and it does help a bit, but sleep polysomnogram shows my sleep architecture is toast + loads of microawakenings (and no sleep apneea).
Back when I was 20 I had a drinking problem. One time I drank so much that I passed out sitting at a table. Woke up with friends having stripped my clothes and washed them. I woke up at 9AM, feeling 100% sober, just anxious about my 20 missed calls from my mom. I got a bit drunk at about 33 and next day I thought I was dying.
That's how I learnt what hangovers were.
Again, at around 25, I helped my brother in-law move bee hives all night, including some 8 hours of driving.
Went straight to work and in the evening I had dinner with my wife at a restaurant.
Now I crash in bed at 9PM and if I'm lucky, I also sleep (but quite often I wake up at 2AM).
Getting old(er) sucks, and I'm only 42 and I miss so much how nice being in my 20 something body felt all the time.