For people that like functional style and using recursion for everything, TCO is a must. Otherwise there’s no way around imperative loops if you want decent performance and not having to worry about the stack limit.
Perhaps calling it an “optimization” is misleading. Certainly it makes code faster, but more importantly it’s syntax sugar to translate recursion into loops.
You don't need full fledged TCO for that; see Clojure's recur for an example. Zig recently added something similar but strongly typed with match/continue. These all map exactly to a closed set of mutually recursive functions with a single entry point, which is quite sufficient (and then some) to fully replace iterative loops while still desugaring to the same exact code.
Indeed there are more explicit versions of such mechanisms, which I prefer, otherwise there’s always a bit of paranoia about recursion without assurance that the compiler will handle it properly.
I did like it, but for me it was fixated on 3-5 comments from the last 1-2 months that got a few more upvotes. It didn’t really work as an overview for the year. Still, a pretty cool thingy :)
Yeah, same here. For the comments it took into account it made pretty great roasts, but would have been better if it was actually comprehensive over the course of the year.
Thanks. I now run a two-step process: first pass reads through all posts and comments to extract patterns, second pass uses those to generate the content. Should be much more representative of your full year now :)
My impression was the same as the poster: it still over-indexes on a couple of recent posts.
Of course, it's possible that we've both been repeating ourselves all year long! I mean, I know I do that, I just think I've ridden more hobby horses than it picked up. :-)
It's fun, though. Thanks for sharing - a couple of my "roasts" gave me a genuine chuckle.
LLMs certainly struggle with tasks that require knowledge that is not provided to them (at significant enough volume/variance to retain it). But this is to be expected of any intelligent agent, it is certainly true of humans. It is not a good argument to support the claim that they are Chinese Rooms (unthinking imitators). Indeed, the whole point of the Chinese Room thought experiment was to consider if that distinction even mattered.
When it comes to of being able to do novel tasks on known knowledge, they seem to be quite good. One also needs to consider that problem-solving patterns are also a kind of (meta-)knowledge that needs to be taught, either through imitation/memorisation (Supervised Learning) or through practice (Reinforcement Learning). They can be logically derived from other techniques to an extent, just like new knowledge can be derived from known knowledge in general, and again LLMs seem to be pretty decent at this, but only to a point. Regardless, all of this is definitely true of humans too.
In most cases, LLMs has the knowledge(data). They just can't generalize them like human do. They can only reflect explicit things that are already there.
I don't think that's true. Consider that the "reasoning" behaviour trained with Reinforcement Learning in the last generation of "thinking" LLMs is trained on quite narrow datasets of olympiad math / programming problems and various science exams, since exact unambiguous answers are needed to have a good reward signal, and you want to exercise it on problems that require non-trivial logical derivation or calculation. Then this reasoning behaviour gets generalised very effectively to a myriad of contexts the user asks about that have nothing to do with that training data. That's just one recent example.
Generally, I use LLMs routinely on queries definitely no-one has written about. Are there similar texts out there that the LLM can put together and get the answer by analogy? Sure, to a degree, but at what point are we gonna start calling that intelligent? If that's not generalisation I'm not sure what is.
To what degree can you claim as a human that you are not just imitating knowledge patterns or problem-solving patterns, abstract or concrete, that you (or your ancestors) have seen before? Either via general observation or through intentional trial-and-error. It may be a conscious or unconscious process, many such patterns get backed into what we call intuition.
Are LLMs as good as humans at this? No, of course, sometimes they get close. But that's a question of degree, it's no argument to claim that they are somehow qualitatively lesser.
I was quite underwhelmed by the demo, I'll try to be as constructive as I can.
The core functionality I expect for such a service is for it to automatically detect who I am, I've seen other marketing services do this, there are ways to map IPs to companies and other techniques. Of course, it's rather creepy and not super helpful to the user, but it may have its (shock) value for making certain kinds of products stand out.
And the personalisation itself... Anyone can make a call to an AI + Search service and generate a new version of the HTML with some slightly modified text, which was not all that different, appealing or accurate in the tests I made. I would suggest upgrading to a higher quality model, proper AI can do much better than this if given the right context.
I suppose it's nice that you are making this easy, if you built your site with a visual website builder this wouldn't be completely trivial to replicate. But still, not a very defensible business for now. I suppose that with good marketing and a serious roadmap to beef this up it could be a viable idea.
The number one request we have is to integrate deanonymisation, so you’re right on the money there. That’s coming in the next couple of weeks or so…
Wrt the changes being text-based for now, we do actually have image and complex element and layout generation working, but have kept it as an experiment for pre-rendered pages until we are confident we can get it right in most cases. (Some early beta users used Kenobi to send out in some cases thousands of customised landing pages with imagery)
We‘re also starting our on demand product with text only precisely because we want to hear what people think we should be working on next as we are a super small team of three!
I do appreciate you taking the time to respond. I wish you luck.
I suppose I was frustrated due to a mismatch of expectations and the fact that I do things like this every day with AI, it feels rather trivial to me.
But I can see how it may appeal to a wider market. I remember coming across a couple websites that were doing this automatically pre-AI, simply detecting who I was and displaying it in some basic ways. And yeah it was a bit weird, but it sure stuck in my mind for a long time, and it was a data-broker type company anyways, it triggered the thought "well if it can do this with me, it must have good data about everyone".
And I can see the more general case for B2B to surface the right use-cases for the user and such. I've interacted with many people with a business profile that would certainly click a "just show me what I care about" button, hell many technical people would love it too just to remove all the whishy-washy hype language and just see what the thing does.
> many people with a business profile that would certainly click a "just show me what I care about" button
You've encapsulated better than I did the kind of visitor segment we're building for right now.
> I do things like this every day with AI, it feels rather trivial to me.
I also agree with this sentiment. It's how I started ("surely it can't be hard to jiggle a page around now with LLMs") and it mostly worked! But the edge cases and heuristics are also proving to be a big chunk of the effort "iceberg" :D
Remember Conglomerates? It just keeps changing name.
Free competitive markets are not an emergent natural phenomenon, they are a technology of civilised societies, and without governments constantly keeping markets free, we keep reverting back to to robber barons and eventually petty warlord kings, that's the natural low-energy state of humans if you let it go unchecked.
In practice, just the act supporting the lesser of the two evils has brought so much more evil.
If you want to do good, fine, but make sure you are smart about it and actually achieve that aim. The US has shown that its not good at this, regardless of intentions, they should just refrain from action until they get their shit together.
When US initiates aggressive unilateral military action on other sovereign countries, US bad yes, of course.
Old-school UN-led "police action" as in Korea is one thing, at least there's a somewhat universal institution making judgements on which countries need to be "saved" under a consistent legal framework, but that's such a slippery slope too.
The US does not have the authority to make such decisions and definitely does not have a good track record of them. It's just vigilantism at a large scale, at best. Even when being charitable about intent, the US did do some things in legitimate good faith, at least partially, the results are always catastrophic. There's been no instance of actually positive outcomes for the local population, it has always destroyed the country for decades to come and set the stage for significantly worse regimes.
No, that's not at all what I said, read it back please.
My point is that history has shown that such action is extremely counterproductive if you actually care about doing good for people under such regimes, particularly when the decision is made impulsively by a single country with a biased perspective and no consistent system or criteria to make sure it's a smart thing to do.
Anyone that supports such action is using inconsistent moralistic arguments to justify blatant power grabs. It may be well intentioned, but you are just making yourself feel good by fighting the bad guys, while doing even more harm to innocent people and making it all worse in the long-run. Very American indeed.
And frankly, right now, the US is not exactly in a position to be a judge of what is moral in the first place.
History has also shown dictators engaging is mass genocide like in Russia, Germany, China, Cambodia, RDC, etc. I am not sure that the idea that leaving dictators alone does less harm to their people is a lesson from history.
Read my comment back as well! I am saying this is neither a moral nor legal matter. So I am not using a moralistic argument.
While I like the sentiment, we have to be somewhat pragmatic. The sanctions on Russia have had a deep impact on the EU economy, mainly the energy crisis and other connected systemic consequences. Germany and much of central and eastern EU became highly dependent on Russian natural gas over the last 20 years, and higher energy prices in general have been quite harmful to the already precarious industrial and agricultural sectors (high-tech farming as in NL, while quite profitable, is very energy intensive and sensitive to tightening margins).
Most of EU (and UK) is on (or near) recession right now, except for some southern EU countries which are doing surprisingly well, although relative to a long period of hardship after the 2008 crisis. It's not an acute recession, but there's no clear way out of this stagnation on the horizon, and the people are really starting to feel the squeeze.
Of course, the root cause of this is much deeper, the Russia situation was just the spark. EU industry has been complacent for decades, believing that while less competitive on costs and scale we still had the technological edge, which ironically led to severe underinvestment in R&D. And giving up on nuclear is backfiring badly too.
I do think the (shrinking) majority still believes that the (limited) actions against Russia were worthwhile, since they are not threatening sovereignty in general, they are threatening EU's territorial integrity at our doorstep. It is unacceptable, and while it is a heavy price, not retaliating would have much more catastrophic consequences.
But cutting off trade with US over Venezuela? Forget about it, EU's dependency on US is orders of magnitude higher than it was with Russia, it would be absolutely deadly to the EU economy.
Perhaps calling it an “optimization” is misleading. Certainly it makes code faster, but more importantly it’s syntax sugar to translate recursion into loops.
reply