djoldman

قبل 12 يومًاdiscuss

> This year, the economist decided that both the midterm and the final exams for his course would be of the take-home, closed-book type (there is a certain tradition of this at Ivy League schools).

These news articles are just tiresome at this point. Obviously folks cheated previously, obviously it's easier now, obviously the answer has been to not have take homes all along.

djoldman

قبل 25 يومًاdiscuss

Follow the Nordic model has never been anything but a fantasy.

All the Scandinavian countries put together have slightly more people than the NYC metro area and are extremely homogeneous in terms of ethnicity, religiosity, etc. as compared to the USA.

In short, it's never been a good policy testbed for the much much larger and more diverse USA.

djoldman

قبل 27 يومًاdiscuss

> So you can imagine how astonished I was last month when an American politician said that it was impossible to earn a billion dollars.

> She wasn't saying, of course, that it's impossible to become a billionaire.... What she meant was that it's impossible to get that rich without doing something bad — without cheating in some way.

> But now you at least understand, from having done the math yourselves, that you don't have to cheat to become a billionaire. You've seen for yourselves that there are only two numbers in the calculation, the growth rate and how long it continues. If it's impossible to make a billion dollars without cheating, which of those two numbers is impossible?

AoC quote:

> There’s a certain level of wealth and accumulation that is unearned. You can’t earn a billion dollars. You just can’t earn that. You can get market power, you can break rules, you can abuse labor laws, you can pay people less than what they’re worth, but you can’t earn that.

Come now @pg.

$2 million * 9.45 months * 93% growth rate = earning a billion dollars, ok. Does that really address what AoC was saying? She wasn't saying that the math doesn't math.

djoldman

الشهر الماضيdiscuss

What's rarely addressed in these articles is the question: if the product/service is so bad relative to the cost, where's the competition? Specifically in this article about fire trucks, they say that margins have tripled... ok, why isn't anyone jumping on that?

djoldman

قبل شهرينdiscuss

> Society gains massively from future workers/tax payers

It seems to me that, all other things equal, future workers/tax payers will lead to economic increases proportional to their costs.

A reasonable forward looking plan / budget scales with the population size. Therefore there would be no need for these special one off exceptions and nudges.

All these little bandaids add up to complexity that necessitates more bandaids.

djoldman

قبل شهرينdiscuss

> Under the US Cloud Act, American companies are required to hand over all information they store to the government if requested to do so, even if it is stored abroad.

Hrm. It's my understanding that a US company is required to give almost no data to any government without a warrant.

djoldman

قبل شهرينdiscuss

Non-paywall:

https://www.msn.com/en-us/money/other/us-is-starting-to-see-...

djoldman

قبل شهرينdiscuss

> Replace ‘CTF’ with ‘high school’ or ‘university’ and you’ve described the total slow motion collapse of education; the only saving grace is that most of it requires in person presence.

So something like, "Frontier AI has broken the 'high school' or 'university' format"?

The hype surrounding AI is just pervasively exhausting: you've got the folks talking about an entire new age for humanity where we're shortly going to take over the entire universe. And you've got the folks talking about how our entire society is crumbling.

Education is one place folks seem to throw up their hands and say nothing can be done.

The fix is simple: students are to be evaluated on their performance in person. That's it.

Any other "collapse of education" isn't due to AI, it's something else.

djoldman

قبل شهرينdiscuss

I would love for the standard to be to ALWAYS report the required amount of memory to load and run a model in bytes of RAM alongside any other metrics. I'd love to see time to first token, token throughput, token latency as well but I'd settle for memory size as described above.

Essentially, many people want to know what the minimum amount of memory is to run a particular model.

Parameter count obscures important details: what are the sizes of the parameters? A parameter isn't rigorously defined. This also gets folks into trouble because a 4B param model with FP16 params is very different from a 4B param model with INT4 params. The former obviously should be a LOT better than the second.

This would also help with MOE models: if memory is my constraint, it doesn't matter if the (much larger RAM required) MOE version is faster or has better evals.

I'm waiting for someone in anger to ship the 1 parameter model where the parameter according to pytorch is a single parameter of size 4GB.

djoldman

قبل شهرينdiscuss

Not to support or attack the rationale behind the css or html standards but these have exact real world SI unit meanings:

   CSS  |                    | Exact Size | Exact Size    
   Unit | Name               | (Inches)   | (Millimeters) 
  --------------------------------------------------------
   cm   | Centimeter         | 50/127     | 10            
   mm   | Millimeter         | 5/127      | 1             
   Q    | Quarter-millimeter | 5/508      | 1/4           
   in   | Inch               | 1          | 127/5         
   pc   | Pica               | 1/6        | 127/30        
   pt   | Point              | 1/72       | 127/360       
   px   | Pixel              | 1/96       | 127/480

https://en.wikipedia.org/wiki/Inch

https://en.wikipedia.org/wiki/Millimetre

https://www.w3.org/TR/css-values-3/#absolute-lengths

djoldman

قبل شهرينdiscuss

Could you share your thoughts about neuralink? Is there enough signal for this to really work?

djoldman

قبل شهرينdiscuss

> Have you never met a bad doctor? A shoddy lawyer? A barista with a PhD?

I presume the implication is that bad doctors and shoddy lawyers exist and just because they have advanced degrees doesn't make them good at what they do. This seems reasonable.

BUT, I find it fascinating that people who aren't doctors or medical experts think they can spot a "bad" doctor or people who aren't lawyers or experts in law think they can spot a "shoddy" lawyer.

A good doctor/lawyer makes good decisions and executes beneficial actions given the facts surrounding a situation. It's pretty hard to judge whether those decisions and actions are good or bad if one isn't an expert.

That's a huge motivating factor for professional licenses.

djoldman

قبل شهرينdiscuss

I came across a comedy clip where the employees are fighting over how many billion tokens they were using and assumed it was a joke.

djoldman

قبل شهرينdiscuss

Unfortunately, it's probably not worth your time to read 99% of arxiv papers, LLM generated or otherwise.

Ever pick a random one and really dive in?

djoldman

قبل شهرينdiscuss

Just a reminder from January: "California completely drought-free for 1st time in 25 years after winter storms"

https://abc7.com/post/california-has-zero-areas-dryness-firs...

djoldman

قبل شهرينdiscuss

Is anyone actually at a company that is purposely trying to use a ton of tokens? It gets expensive really fast.

djoldman

قبل شهرينdiscuss

> Runo extracts by meaning, not DOM position. Site redesigns and HTML changes won't ever break your pipeline.

Bold claim.

djoldman

قبل شهرينdiscuss

TIL:

> eCall was made mandatory in all new cars approved for manufacture within the European Union as of April 2018.

https://en.wikipedia.org/wiki/ECall

djoldman

قبل شهرينdiscuss

France.

djoldman

قبل شهرينdiscuss

> The thing is, the cost of discouraging the wrong kid -- the one who ends up curing cancer or otherwise innovating in an extremely useful area -- is unbounded.

> The cost of encouraging the ones who fail can be heavy, but at least it's finite.

Assuming that every kid has a non-zero chance of being the "right kid," then discouraging only one child results in infinite cost and so every child should be encouraged to try to cure cancer...

Submissions

John Deere owners will get the right to repair equipment under FTC settlement

Farmer, marketer at odds over sales of white nectarines

Profiling.sampling – Statistical Profiler

No one can force me to have a secure website [pdf]

When Should AI Step Aside?: Teaching Agents When Humans Want to Intervene

Primesieve: Fast Prime Number Generator

US, Iran say they have agreed to a two-week ceasefire

Stack Overflow (Beta)

100 Prisoners Problem

Reward, Risk, and Regulation: American Attitudes Toward Artificial Intelligence [pdf]

Software Development Job Postings on Indeed in the United States

The Robotic Tortoise and the Robotic Hare

Special Commitment Center

US SEC preparing to scrap quarterly reporting requirement

Anthropic sues Trump admin. seeking to undo "supply chain risk" designation

Sioyek is a PDF viewer with a focus on textbooks and research papers

Project Maven

Apache Otava

Project Pele

Visible Spectra of the Elements

comments