inertiatic

It's clear it means that there's a large overlap in titles and they are available in better quality on the PC platform?

We now have a tool that can be useful in some narrow domains in some narrow cases.

I get being reserved about where this goes, but saying something like this is quite insane at this point.

We are essentially offloading our moral responsibility to a black box that cannot be held accountable.

We already did it with companies, buddy!

Here you go https://github.com/apache/lucene/pull/656 - no need to do anything from the user side to trigger it as far as I know.

Lucene and ES implement a shortcut for filters that are restrictive enough. Since it's already optimized for figuring out if something falls into your filter set, you first determine the size of that. You traverse the HNSW normally, then if you have traversed more nodes than your filter set's cardinality, you just switch to brute forcing your filter set distance comparisons. So worst case scenario is you do 2x your filter set size vector distance operations. Quite neat.

You know, humans pick up hobbies like cycling or running which they do consistently for years, listen to specific music genres or even electronic music which is mostly just a beat, hang up a painting they like in the living room and look at it for years and years, go out to their favorite place to eat consistently or cook the same passed down family recipe, and in so many other aspects avoid sudden changes, and you're surprised that for video games we enjoy the same formula repeatedly?

An engineer starts at $200/hr

Starts?!

I'm still looking for a systematic approach to make a hybrid search (combined full-text with embedding vectors).

Start off with ES or Vespa, probably. ES is not hard at all to get started with, IMO.

Try RRF - see how far that gets you for your use case. If it's not where you want to be, time to get thinking about what you're trying to do. Maybe a score multiplication gets you where you want to be - you can do it in Vespa I think, but you have to hack around the inability to express exactly that in ES.

I don't know how this is something related to AI - you could polish and embellish your resume before LLMs too, I'm fairly sure. I guess this gets the clicks.

Not being to remember small details about certain projects is also perfectly fine for people who have worked for more than a couple of years. Unless you can discover a pattern of lying like the author supposedly did then I would just be perfectly fine moving on to another topic.

4.8b words on English Wikipedia. Knowledge cutoff of 6 months. A valid use case is to search across Wikipedia and ground your answers. Trivially proves that RAG is still needed.

I use Firefox as my main browser and it's not a viable alternative to Chrome if you have the very common usage pattern of keeping tens of tabs open.

Interacting with this model is just supplying your data over to an adversary with unknown intents

Skynet?

People wait for things all the time and if it can take away the hassle or driving your kids somewhere, maintaining multiple vehicles for multiple family members etc. I'd bet that it does take off. Public transport doesn't solve groceries or other types of shopping though even in highly connected cities, especially for people with health problems or just aging ones.

Lucene which is what ES builds upon is a library with bindings in languages other than Java, and it's Apache licensed.

Didn't we have multiple open-weight frontier model announcements?

I find it really challenging to talk with people who have completely separated their work from their emotional being. For me, work is part of life and it should be a meaningful one.

I like my work too but when I hear things like this it's so hard not to cringe because of how much of a place of privilege this comes from. Most jobs are things no one would want to do if they didn't need to survive, and that's fundamentally built into our society and economic system. Let them eat cake though.

Solved to the point that squeezing a bit more recall out of vector retrieval isn't the problem anymore.

I think this is a bit of a strawman. I don't think recall is the main point these systems are trying to sell us on, it's more about robustness and ease of use compared to building something inhouse or using a lower level library to build a system on top of it just for this small part of your overall project/product (be it RAG, search, whatever).

I guess Lucene-based solutions, while very mature overall in terms of engineering, lagged behind this functionality (out of caution, trying to build what's going to be long term useful) and are also perceived a bit too cumbersome. So these stores do make sense, I think. The core functionality is nothing too complex (at least HNSW), but hiding it behind a stable black box with just a few inputs and levers, has value for people that are likely to use these stores.

I do, and it's very very rough around the edges to be honest. Lots of things broken, things are even breaking between releases suddenly in unexpected places. Or at least, I'm used to working with more robust data stores. If my work was more high stakes, I'd have already advocated for moving our vector search to something more robust. Thankfully it's not and I can just maintain what we're making with not too much stress, and enjoy seeing this OS project grow from a user perspective (haven't seen a data store go through this very initial phase in my career yet).

Support from the team is great however, and congrats to them for this round!

The fact that the stats that drive this could be weighted in any arbitrary way makes the overall rankings pretty worthless in my eyes. And looking at the top 10 vs bottom 10 kind of validates these thoughts. I would much rather live in most of the bottom ten than in the top ten, if we're talking long term.

Brand new laptop so haven't got a chance to really get a good sense of an average. I think depending on the workload I got 9-12 hours thus far.

I'm on a 6850U Ryzen Windows machine and it's really smooth. Previously I too was using an 11th gen XPS 13 but on Linux, where battery was approximately 2 hours long. This machine feels both smooth, cool and for my usages there's no slowdowns on battery either (usually running few browser windows, a VM that's not taxed too hard and an IDE).

If you're making $200/hour, you probably don't share a lot of the problems other people have around buying goods of all sorts. Furthermore, you're likely to go shopping with friends/significant other/kids anyway and touching something when you're there is free - assuming you could even find clients to pay you your $200 an hour on a Saturday morning for example.

Overall, I don't think it's a bad idea to develop an intuition for most things you consume, even if you're well off.

an average call center employee can cost $30 for a complex customer question

How? I've worked in a call center where big companies outsource their customer support in Eastern Europe, and we were paid 3-4 euro an hour. Even if a "complex customer question" takes an hour (which doesn't happen often at all), what sort of overhead can there be?

So if you have a question that someone with HN-level IQ cannot resolve themselves then customer service cannot profitably help you.

How is it a question of intellect? Customer service might have access to more information not in the public, might have access to other instances of the same problem where they had to find out how to interpret the rules and disambiguate, and most importantly, they usually have the power to take action (untick a box somewhere, offer a refund, whatever).

And how would that be profitable long term? Just by virtue of the fact that in a market with actual competition, leaving your customer dissatisfied is not profitable.

No, I didn't, I'll reiterate what I said - that most of the CEOs start as founders and remain with their companies until they're successful (without being replaced to get there) - and beyond, so most successful companies actually have "inexperienced" CEOs.

And how many successful CEOs of large companies were successful without demonstrating they could manage a large org before getting promoted.

I don't know, most of them? Most companies have their founders as their CEOs until they become very successful, and these are their first gigs as CEOs of a big company, right?

I've had almost the exact same list of problems on my Dell XPS 13 running Linux.

you’re all confirming that contributing to corporate capitalism is incompatible with personal freedom

Isn't it obvious? You're literally giving away more than a third of your life working these jobs to do things that you don't personally want or benefit from. You think the more restrictive thing that is being really imposed on you is being unable to tweet "Fuck you" instead of spending like half of your waking time away from your family and loved ones?

I'm personally not a fan of capitalism conceptually, not just the corporate kind, but it always strikes me as weird how other leftists think that's the unthinkable part, that someone would fire you for speaking out of line while still acknowledging that the other party is part of an organization that kills thousands and suppresses nations for oil for example.

It certainly reads like an attack, and that's to someone who agrees with the sentiment. And as an observation, in professional contexts you can generally be poetic and figurative when talking about good things ("Gary is such a godsent and a genius for showing us the way to restructure our department"), but it's in bad taste to do it to express negative things ("I worked with Gary and he's an absolute mouth-breather and defecated all over our infrastructure").

You're saying on the documentation:

nor send your code to another third-party service.

But aren't you actually doing that? Sending things to OpenAI to use as context?

Not really. You need to drill to make any oil. So if you drilled and spent 10 parts to make it happen (flying people there, energy to drill etc.) and extracted 11 parts, you're selling 1 of those and then using the next 10 to mine the next 11 to get another 1 to sell. It would still make sense economically as long as you were turning a profit overall (so factoring the total cost into what you're actually selling), although it would be a huge waste in absolute terms.

I do doubt that would be a realistic estimate though.