ototot

Their old paper: https://dl.acm.org/doi/pdf/10.1145/3510457.3513031

It's basically dimensity 9400, and I think it's fairly comparable.

You can even install linux on Chromebooks, and ChromeOS has upstreamed / opensource many of their codes.

In other perspective, ChromeOS supports running Linux apps w/ GUI without much differences. You just open your terminal, type `apt install XXX`, then `XXX` should work out-of-the box.

I don't see any reasons that ChromeOS is less open then macOS

IMO, Linux kernel always lacks of auto-testing. Some sort of AI review may really help here.

Yes. I'm a ICPC World Final medalist.

Totally agree that IOI bronze is way easier than ICPC bronze. In terms of rank/ratio, ICPC medals are more like IOI gold.

I stated things like that because I thought it's a bit easier to let people know the difficulty difference. (Agree weirdly though)

OK. I think my opinion and definition on "easier" is indeed vague. For "easier", I'm only comparing the thinking difficulty.

Yes, medal is function of ranking but not difficulty.

Nonetheless, I would say that IOI more focus on thinking, which I to some degree is not that good at, while ICPC is more like a mix thinking and implementing. Therefore, my ability to implement stuff can improve my ICPC ranking but not IOI.

Because I'm a ICPC medalist (not this year though) but not a IOI medalist.

Another evidence is that you only have 5 hours to solve 3 problems in IOI, but you need to solve 10+ problems in ICPC. It's impossible to have all 10+ problems to at IOI level in ICPC.

I compare the difficulty by solving them myself.

For fractional scores, it depends on problems. In short, there are two types of problems in IOI. One is traditional problems that requires 100% correctness, and the other is continuous scoring.

The prior can still results in score between 0 and 100, but this is because there are subtasks in the problem. For example, a graph become a tree or even just a linear sequence. Nonetheless, you still need to ensure your algorithm is correct on all testcases in that subtask in order to get the score of that task.

Given that ICPC problems are in general easier than IOI problems. I wouldn't be surprise to see they can get Gold (even perfect scores) in ICPC.

Nonetheless, I'm still questioning what's the cost and how long it would take for us to be able to access these models.

Still great work, but it's less useful if the cost is actually higher than hiring someone with the same level.

Related: https://news.ycombinator.com/item?id=43528479, https://news.ycombinator.com/item?id=43527001

I also tried one-shot.

https://g.co/gemini/share/badd00a824d2

And OpenAI is announcing their ImageGen in 4o

https://news.ycombinator.com/item?id=43474112

The Gemini 1.5 tech report do reference some papers about supporting large context window.

and the price is 0.0 usd, lol

Maybe they just asked Gemini 2.5 to write the announcement.

It could be the code still stay there, but the direction has been changed.

Sublime is my default editor on Windows since 2012. I really enjoy its lightweight and cleanness.

Haven't read the post, but I think there must be something wrong if a type's safety depends on the Drop trait due to the existence of std::mem::forget.

Example of 'agentic': https://blog.google/technology/google-deepmind/google-gemini...

I believe this kind of issues happen in Google everywhere. Not sure who to blame for, but it’s definitely a severe crisis Google is facing. Google should spend time fixing it.

Is this a post posted by me? I have basically the same thought on the new control center. The new UI and edit flow truly disappoint me a lot.

The connection group is also a terrible idea. Why don't apple just let you do grouping yourself and make every button in it to be available.

https://news.ycombinator.com/item?id=35782630

https://news.ycombinator.com/item?id=28283030

similar: https://vuejs.org/?uwu

Given the range. Doesn't look like DNS

They layoff too many ppl.

why don't they use a real textbook?

Sorry what is TFA?

The largest forum (PTT) in Taiwan is also a BBS that you can access it with SSH.

https://github.com/ptt/bbs-sshd

Of course, nowadays you can also access PTT through websocket or a web interface.

This is not a bug. This is the model of C. You should blame C not GCC.

As I know, Rui was in the US (when he was at Google), so it is a fair statement.