Eridrus

11 ms·

241.
by Eridrus 1y ago
I think the main takeaway from any startup stock advice is what this article starts with: you need to pick a good startup. The details all matter, but they all matter far less than that fact. People shouldn't lump all startups together
242.
by Eridrus 1y ago
I feel like there's some conceptual drift going on in Kagi's blog post wrt their proposed remedy. They argue that the search index is an essential facility, and per their link "The essential facilities doctrine attacks a form
243.
by Eridrus 1y ago
I dug into this a bit since these comments were just confusing, but apparently Prop 13 required tax increases to get a 2/3 majority of the legislature. So it seems like Prop 13 just fucked California.
244.
by Eridrus 1y ago
Did it have to be funded by a proposition? Could the CA legislature not have passed a law?
245.
by Eridrus 1y ago
If you don't try and reinvent everything, you can get economies of scale by doing what the rest of the world is already doing.
246.
by Eridrus 1y ago
Reviewers. Even small deviations from academic style get negative responses from reviewers.
247.
by Eridrus 2y ago
I don't think your complaint/experience actually lines up with the numbers here. In the Post-Series A numbers, the lowest numbers are in the ~0.5% range. This is at most 2 figures off what the founders could get together. In a wor
248.
by Eridrus 2y ago
The business model question applies to all of these companies, not just Google. A lack of workable business model is probably good for Google (bad for the rest of the world) since it means AI has not done anything economically useful and Go
249.
by Eridrus 2y ago
I am mostly imagining the "happy middle" scenario where you raise $1m and sell for $10m in cash, but modified to assume you raised $10m and spent $2m of it, and then got the $10m in cash from the acquirer, and still have $8m in th
250.
by Eridrus 2y ago
It's obviously better to raise the optimal amount and no more. But things are not always so clean, and the best time to raise is when your company is killing it, not when you're running out of cash and trying to make it to profita
251.
by Eridrus 2y ago
Sure, now you need to do this statically ahead of time. It's not unsolvable, but now you have a more complicated system. A better file format would not have this problem. The fix is also trivial (escape new lines into \n or similar) wo
252.
by Eridrus 2y ago
Doing sequential reading into a queue for workers to read is a lot more complicated than having a file format that supports parallel reading. And the fix to allow parallel reading is pretty trivial: escape new lines so that you can just kee
253.
by Eridrus 2y ago
High value searchers for Google are people looking for car insurance, which is not something anyone is doing with LLMs.
254.
by Eridrus 2y ago
Because sysadmins are taking separate actions to make the milk expensive.
255.
by Eridrus 2y ago
Scrapers still have an economic incentive to do what is easiest. Providing an alternative that is easier than fighting sysadmin blocks would likely cause them to take the easier route and make it less of a cat and mouse game for sysadmins.
256.
by Eridrus 2y ago
I'm mostly suggesting a mirror to some centralized location. GitHub is probably a good place to mirror to for code, but it could be elsewhere. I was actually thinking of a more general thing than just code, eg similar to CommonCrawl, b
257.
by Eridrus 2y ago
One of the reasons we have juries is to answer subjective questions like this, because many questions in law are actually subjective.
258.
by Eridrus 2y ago
I know a lot of FOSS people are hostile to AI in general, and this is an immediate problem, but I feel like a better solution for everyone would be for there to be some sort of central repo of this information that AI companies can pull fro
259.
by Eridrus 2y ago
You can probably think of some yourself. Or you can continue being obtuse, up to you.
260.
by Eridrus 2y ago
I do think people writ large do not read enough, there has been a lot of knowledge already written down and many engineers and researchers I work with would rather do than read. And the most detailed information is written, not in other for
261.
by Eridrus 2y ago
Because it creates bad incentives and workplaces. Nobody likes the conditions working in places that have to take counterintelligence seriously.
262.
by Eridrus 2y ago
Most people should be very uncomfortable giving a random gig economy worker access to their personal email accounts that act as fallback authentication for everything in their lives. The fact that our AI systems have this level of trustwort
263.
by Eridrus 2y ago
I applaud you for trying to tackle this, but after reading your docs a little I am skeptical of your approach. Your example of having a rule saying that the user's email address is not put in a search query seems to have two problems:
264.
by Eridrus 2y ago
Eval sets are not an appropriate tool for evaluating progress on security problems since the bar here is 100% correctness in the face of sustained targeted adversarial effort. This work largely resembles the Politician's syllogism; it&
265.
by Eridrus 2y ago
Given the fact that nobody actually knows how to solve this problem to a reliability level that is actually acceptable, I don't know how the conclusion here isn't that Agents are fundamentally flawed unless they don't need to
266.
by Eridrus 2y ago
Just wanted to +1 this comment and say Vanta made SOC2 way more intimidating than it was. What made it easy was talking to a startup that wanted soc2 and had it themselves who recommended an auditor who helped us untangle what was actually
267.
by Eridrus 2y ago
It's a huge mental leap from "I like the output of nepo babies" to "the blandness in the works is due to the psychology of wealthy people". I would suggest reading more about the history of development and ask yours
268.
by Eridrus 2y ago
Yeah, I strongly disagree with this too. I've spent a lot of time reviewing code and doing code audits for security (far more than the average engineer) and reading code still takes longer than writing it, particularly when it is dense
269.
by Eridrus 2y ago
Congratulations to both teams :) Getting to the moon at all is a huge accomplishment. The article actually mentioned IM-1 (though not by name) and got me looking.
270.
by Eridrus 2y ago
The reason to let rich people keep their money is twofold: ethical economic activity is positive sum and rich people have shown an ability to generate these positive spillovers, and letting them keep it gives them an incentive to engage in

More ›