r00tanon

My first impressions and thoughts on the presentation speak to the need for human mathematicians to accept, absorb, and understand new AI-generated proofs—and to the fact that it takes time to do so, as well as the increasing burden of sorting through essentially unsolicited material, which robs that same group of the time to work. Software developers are seeing this same issue with AI-generated contributions of software changes in the open-source arena. What they are seeing is a very high number of trivial submissions, which Dr. Tao points out is a propensity of AI generation. The second problem is that submissions are often not good, forcing reviewers to read hundreds if not thousands of lines of intricate yet incorrect code—and they waste their productive time doing so, making them less amenable to future consideration of such contributions. This is a real problem. As to the issue of human mathematicians needing to accept work as useful before it can enter wider consideration—now that is an interesting problem if humans are indeed unable to understand work AI creates that is useful and correct, but not understood to be so (perhaps due to the above issues), and yet we have to consider a point in time when models become more capable. Let me put it another way: how long would you spend trying to get your dog to understand calculus? AI has inexhaustible patience (unlike humans), and it will keep trying to explain its ideas and theorems to us (or our dogs) far past the point humans would be willing to listen—though our dogs, at least, are amenable to treats and used to us talking a lot while handing them cookies.

Josh Collinsworth paints an accurate picture here, and he’s not wrong. People will be hurt by the displacement and disruption that new technology brings, not just AI. It has been happening since humans first picked up stones and sharpened sticks. No one says you have to be happy about it, or even optimistic, even if you happen to be part of a privileged class.

I was going to mention Yegge's recent blog posts mirroring this phenomena.

There's also this article on hbr.org https://hbr.org/2026/02/ai-doesnt-reduce-work-it-intensifies...

This is a real thing, and it looks like classic addiction.

It's a tough game for Perplexity as long as they use Google's search, and with Google now providing "AI Summaries," and with the whole ad ecosystem (invented by Google) designed to turn us into predictable consuming robots, what's a pre-IPO company going to do? Fight the good fight and stick with a subscription model, or knuckle under and start sending us Google ads?

Sucks. I will say this though: as soon as I see ads in my Perplexity search results, I'm cancelling my paid subscription and simply going back to Google. Without any other differentiating feature set, why would I pay Perplexity?

RIP, Perplexity; it was good while it lasted.

How about looking at an ast-based method for making changes across code base? https://www.reddit.com/r/Python/comments/17tvm06/astgrep_and...

Bullshit.

Reading these comments after falling asleep to SNL sketch re-runs. They all sound oddly sarcastic and ironic.

Good article. I, too, am fascinated by calculators and how they work, especially the HP scientific RPN calculators. So much so, that I ended up developing this:

https://play.google.com/store/apps/details?id=com.limpidfox....

Reading down further I see this is a follow-up to "40 Algorithms Every Programmer Should Know". So, is this new book the same 40 algorithms from the prior book with 10 added?

If so, perhaps the real book could be called "The One Book Writing Algorithm You Will Ever Need."

The android app wants full access to the file system? With the "trust us, we'll only really access the folder you tell us to..."?

Am I reading that right?

It's like Game of Thrones - without the intelligent intrigue.

Wow. I see almost nothing controversial in the McKinsey article. In a nutshell it addresses hard metrics (DORA) gathered via the repo tooling (commits, MRs, etc) which is enhanced with some outcome metrics, and they combine with soft metrics (SPACE) which are inevitably fluffy on the numbers combined with some other measurements to fill out arguable gaps. But overall, it didn't read like a manifesto.

As far as the criticism from the "ex-finance developers" - yeah, some of whom have done a lot of other noteworthy things outside of "fintech". I don't think it was overly harsh. In fact, maybe just focusing in on certain things that inevitably leave a bad taste - such as assigning numbers to human being's activities or otherwise trying to distill human activity down to hard numbers can lead to abuse - no kidding - almost a tautological argument.

That said, happy people with a shared vision and supportive management delivering a product/idea that is contributing to the team's concept of "the greater good" will always be the sweet spot for productivity.

I would lean more towards the SPACE metrics as the best way to measure, but allow that hard metrics such as DORA can be helpful as alarm bells.

Pair programming can be a productive, even fun, activity in limited situations. Now that we have the option of our pairing partner being human or AI, it sort of highlights a lot of the situations where pair programming is NOT productive, or even fun.

I have had a nearly identical experience. Thanks for sharing.

We've reached the point just past the elbow on the Moore's Law graph where things are skyrocketing. Any projections we humans make about the future advancement of AI are hopelessly inaccurate.

For example, running a GPT-4 model on your laptop will likely be possible in the next five years*.

*It may even happen within six months. At that time, much of the concern surrounding private conversations with an LLM will evaporate, and attempts to regulate its use will likely be hopelessly ineffective.

Yep. It was fun while it lasted, but I expect none of the manufacturers will support hacking or mods on these devices in the future because there's too few people doing that compared to the masses they want to sell to, and they have zero time or money to train technical support staff. They want to sell you an appliance that can't be repaired, has planned obsolescence and locks you into the platform by (effectively) keeping your data hostage.

So anyone who has already unlocked their Pixel 2 is good to go right? Getting the latest Pixel unlocked requires an internet connection, but once done it's done.

So the complaint is that in the far future the vendor might not support that on a six year old phone (by that time)? But, it is supported now, so what's the beef?

My old Pixel 2 only works now plugged into power and on WIFI because the battery is dead. I can get super-pissed off at Google for not supporting battery replacement on the Pixel 2, but it's a 6 year old phone. How long should I expect them to support it knowing at the time about the sealed battery. I can't get that pissed seeing as I bought it knowing the facts at the time :)

ELI5 - everything about a phone only works when the carrier allows it to connect to the network.

So, the way this is done is to, well, connect to the network and allow the phone to be unlocked - which according to the article is what happened?

So the real complaint is that the Pixel can't be loaded with a customized Android OS (or Linux, etc.) without being connected to the internet first and this is bad because the vendor might put bloatware, spyware, etc. on the phone, which, once you install your OS will be gone anyway?

OK, well then other than the underlying hardware needing to still be recognized on the carrier's network, which pretty much means you still have to be connected. The point is you aren't going to be stealthy using the carrier network on a phone, so at the EOD you have a pretty expensive device you can really only use on WIFI using your own VPN without the Vendor's software.

So what's the advantage over a small tablet based on OTS SOC hardware that you have full control over? IOW why buy the phone in the first place?

Like others responding here, I was lucky to see the LCM just before the pandemic killed it. Truly sad to see it go.

Product description sounds like Luke Skywalker's description of Princess Leia's rescue reward to Han Solo. "More then you can imagine."

To which Han replies, "I can imagine quite a lot."

Unfortunately just diagnosed type 2. Can't find Ozempic anywhere due to this crazy drug shortage and off label prescribing.

Main reason would be they have a lot of legacy code written for Oracle.

For new projects, the choices are wide open.

The fact that we exist and live is baseline extraordinary.

What we do after that is so-called gravy.

But seriously. Measuring ourselves against the Einsteins is a path leading nowhere.

What you do adds to the universe.

Interesting that summon and auto park will "not be available and will be restored once vision gets parity with the sensors..."

Wow. Wondering when that will happen, as I still get phantom braking in odd situations with my 2022 model 3. Case in point on a freeway straight away with no other vehicles and nothing but grass fields and gray sky. Car suddenly brakes as it tops a gentle hill.

I can only go by my experience in that a passion for making computers do things was the seed. From there it drove a desire to learn more. I don't think most people can sustain the desire without that. It really should be about feeding the innate talents. Could I make a living playing music? No. It's not a one size fits all thing.

I often think what Richard Feynman says "What do you care what other people think?"

In general I've found people don't really spend a lot of time dwelling on other people's actions. They have their own problems.

We are all our own worst critics. In most cases it's best to let things go until you have more information or feedback. Otherwise take note of your actions and try to let go of guessing what other people think.

If your intentions are good but you think you might have screwed up, try to give yourself the benefit of the doubt and return the favor in the reverse.

Severless.com framework has lots of plugins that allow local emulation of AWS services like S3, dynamodb, rds, etc. that your application will be using. It also helps with running debugging remote lambdas and deploying them. There's also SAM.

If you are going serverless, one of these toolkits makes it a lot easier.

I admit I was being somewhat facesciuos. But clearly If you took credit for the work your computer did running your program, you would probably not get an argument from your computer until the conditions I mentioned were met

Well, presumably, if it is your computer, running your program, using your equipment, you've got the copyright until your computer gains consciousness and a lawyer capable of filing a lawsuit.

Good news. We've found lambdas orchestrated with step functions and properly split out as to function can do a lot of data processing even with the limits you stated before needing to switch over to AWS batch or ECS. Nice to have that tool in the belt. I'll check it out.