Hacker Newsnew | past | comments | ask | show | jobs | submit | tobrien6's commentslogin

The Islamic clampdown on the Iranian people has gone on for too long. The younger generation is willing to die for their freedom. The internet is locked down, satellite comms jammed with Chinese tech. The streets smell of blood. Police and imported Arab thugs open fire into crowds of protestors. Hospitals are systematically raided to finish off the wounded.

It will be his greatest act as president if Trump sends real assistance, as the Iranian people are begging him. It will save countless lives. Either way, in the end, Persia will rise again, the lion will raise its head, the brutality of Islamic oppression will be cast off, and the world will come to know the true spirit of these people.


I hope you're right.


BYD is overhyped in the US for some reason. In the countries where they are available, they are considered decent budget vehicles. There's a reason why the Model Y is still the best seller in China.


https://autovista24.autovistagroup.com/news/what-were-best-s... says, for the first half of 2025:

  Best-selling BEVs in China
  January to June 2025

  Geely Geome Xingyuan     205,091
  BYD Seagull              174,912
  Tesla Model Y            171,491
and

  Tesla’s quarterly delivery boost meant its June results
  propelled the Model Y into third, from fifth in May
⇒ I doubt the Tesla model Y is the best seller in China.


> In the countries where they are available, they are considered decent budget vehicles.

Yes, and I want a decent budget vehicle here.


> There's a reason why the Model Y is still the best seller in China.

Yes, but it's the same reason the original bondi blue iMac was the best selling model of computer when it came out, while Apple themselves were still on something like 5% of desktop sales:

Lots of wood behind few arrows.

https://carnewschina.com/2024/07/14/best-selling-vehicle-bra...

(Caveat: first result, I don't know site reputation)


Same country that seems to disproportionately like Buicks? Yeah, “there’s a reason”, but it might not be what you think it is.


Some would say Tesla has been over hyped as well.


They can say, but in open markets Teslas sell better than cheaper BYD’s.


1. Are there any truly open markets for cars?

2. Tesla China has factories in China, was still #8 in Chinese market, with BYD at #1, if this source from 2024 is correct: https://carnewschina.com/2024/07/14/best-selling-vehicle-bra...


Australia, NZ.

I've test drove my mates atto3, I have model y. He's got lots of regrets, fixes. It's made from good materials, styled poorly and technology is like everyone else - poorly done. Price does reflect it, no free lunch.


X-Pengs are where it's at. They are taking over fast in Europe with better cars for less money.


The hype is because they are decent budget vehicles.


> There's a reason why the Model Y is still the best seller in China

...and that reason is that Tesla only has two models, the 3 and Y, that get any significant sales in China whereas BYD has several.

For example in the top 10 a couple months ago BYD had models that sold 70%, 69%, 59%, 58%, and 53% of what the Model Y sold.

When you look at cars sold per company rather than specific models Tesla is #10 in China with 1/6th of BYD's sales. Between BYD and Tesla there is Geely, Chery, Changan, Haval, a SAIC-GM-Wuling joint venture, a SAIC-Volkswagen joint venture, Toyota, and Xiaomi Auto.


I suspect that ML will be state-of-the-art at generating human-interpretable theories as well. Just a matter of time.


I'd be ok with this if paying for the game meant that I get access on all platforms the game is on.


This is as bipartisan as it gets. Senator Chuck Schumer, senate majority leader, has introduced a bill that legally requires all individuals with direct knowledge of non-human technology to disclose it. Both the senate and house are pursuing parallel tracks pushing for disclosure, both of which involve republicans and democrats. You're welcome to dismiss this all as a ruse, but you do so in ignorance.


Ah, the old “Believe Chuck Schumer, or you’re an idiot” argument.

I think I will dismiss this as pure, government-grade stupid, if not a “ruse” (which I don’t think most of these people have the intellectual ability to maintain against anyone but the most credulous partisan).


UAP sightings and encounters are a global phenomenon, documented by civilians and military/government individuals and declassified documents going back decades. If you begin even cursory research into the topic, you'll discover as much.


Ya, one of my favorite cases actually occurred next door to Botswana: https://en.wikipedia.org/wiki/Ariel_School_UFO_incident


I see how your response makes sense in terms of questions like "why does an electron exist", but what about the general notion of existence? It seems self-evident that there is something, even just experience. I interpreted the question as more "why something rather than nothing" which seems coherent.


I don't really follow. I interpret "there is" and "something" as quantifications, not predicates. What would be the predicate here?

The general notion of existence, at least in natural language, seems to be a shorthand for quantification or negation. I'm not sure what it would mean for "existence" to exist, short of adopting Platonism or some bizarre metaphysical system.

The "something rather than nothing" question, if stated coherently, would likely be beyond the scope of human knowledge. Although ultimately I don't think it can be stated in a way that makes sense.


Experience exists. Whether the contents of the experience are "real" or "not real", it is not coherent to claim that there is truly nothing at all -- what could even cause one to come to that conclusion if there's nothing at all? Platonism is one of the few (only?) games in town in terms of potential ability to furnish answers here (a Popperian scientific method has well-defined boundaries on the scope of explanatory power) -- and the more indications we get that physics can be derived from number theory and combinatorics, the more seriously I think it will be taken as a research topic. A more formally developed Platonism would also potentially be able to address the "something rather than nothing" question.

Why do you think "something rather than nothing" question doesn't make sense?


Well, I can't engage "Experience exists" because I don't treat "exists" as a predicate, as I stated earlier. What you seem to be asserting as straightforwardly true looks to me more like a malformed sentence. Yes, words and concepts arise in our language and are metaphysically constrained by reality. However, it seems to be quite a departure from this milquetoast linguistic fact to assume that natural language would, could, or even should map to the structure of reality—especially with respect to heavyweight "existence" claims—in any meaningful way.

I adopt ontological anti-realism and do not stray into "existence" claims because they simply raise too many methodological issues. The quotes here are important because I don't have an intuitive understanding of what people mean when they say "exists" in the first place.

You could rephrase "Experience exists" as "Humans experience" or "I experience," which is analogous to Cogito in Cogito, ergo sum but without the ergo sum. There's a subject and a predicate already, and "exists" adds nothing.

If you want to make substantive claims about what does or doesn't exist, your views are subject to relatively straightforward reductio ad absurdums. This is demonstrated pretty clearly in On What There Is.

> Why do you think "something rather than nothing" question doesn't make sense?

"Something" = quantifier, "nothing" = quantifier, existential "is" = quantifier. Again, I would insist on a predicate here.


Truth. Parent comment is deeply misinformed. BTC already has more than enough hash rate, and it is entirely independent of price.


That does not make sense.

The whole point of proof of work is that it is proof that someone has spent the value in electricity validating a block.

As the price of bitcoin rises, naturally more and more miners will beef up their rigs (that is, spend more money in infrastructure and electricity) to mine BTC, because the rewards and fees are worth more. Note if they didn't do this, anyone could easily have a "free money" arbitrage play, and basic economics says that discrepancy in prices (i.e. the value of a BTC on the market and the cost to mine one) must be arbitraged away.

Think of it this way, if hash rate is independent of price, why does it take the electricity output of Argentina to run the network? It certainly didn't take that much years ago. The reason being that as BTC becomes more valuable there is more reward and thus more competition for miners to mine. As that happens, the hash rate will rise as more power is added to the network.


It's backed by mathematics instead of a country. That doesn't make it better or worse, just different, and diversity is generally what people want in their portfolio.


It's not backed by mathematics any more than gold is backed by chemistry.


Uh, gold _is_ backed by chemistry. If gold were to change nearly any of its chemical properties, it would instantly lose its value. It is stable, generally non-reactive, non-toxic, lustrous, and of generally low abundance - all properties derived directly from its chemistry.

That is actually one of the strongest arguments, to my mind, for why some variant of a cryptocurrency will survive as a store of value. Gold kind of "organically" found its way into value almost precisely because it is backed by the physical laws of nature. Cryptocurrencies went up a level to be backed by the laws of mathematics.


Gold's value comes from being a Schelling point, not from any particular chemical properties - other than, like, not killing you or exploding.

Bitcoin is also a Schelling point.


Sure and cash is backed by a special linen-cotton blend of paper.


So, quite a lot then?

Making gold has always been the holy grail of chemistry. It's only valuable because chemistry says you can't make it easily.


Amazon's Opensearch (fork of Elasticsearch) natively supports vector-based approximate KNN (using https://github.com/nmslib/nmslib/) which is integrated with Opensearch's native filtering functionality. Elasticsearch also has similar functionality, but I don't know if their KNN code scales quite as well.


Opensearch only supports "pre-filtering" or "post-filtering," which leads to either high latency or incomplete results, as explained in the article.

This is why single-stage filtering was the most-requested feature for us.

From the Opensearch docs:

> You should not use approximate k-NN if you want to apply a filter on the index before the k-NN search, which greatly reduces the number of vectors to be searched.

> Because the graphs are constructed during indexing, it is not possible to apply a filter on an index and then use this search method. All filters are applied on the results produced by the approximate nearest neighbor search.

> If you use the knn query alongside filters or other clauses (e.g. bool, must, match), you might receive fewer than k results.

(https://opensearch.org/docs/search-plugins/knn/approximate-k...)

I know Elasticsearch is working on introducing vector search but it is not yet available. I don't know how they will support filtering.


The approximate kNN is quite nice for many use cases, and scales to billions of documents. However, you're correct that filtering happens on the results. This is only an issue in certain use cases where filtering is very narrow, as you can often just request much higher k than the number of results you really need without much slowdown.

If the filtering is very narrow, as you commented they also provide functionality to perform pre-filtering and then exact kNN on the results. This is of course higher latency, but still quite acceptable for many use cases (this is how I use it).

I believe there are use cases that Pinecone addresses better than Opensearch, but I want to let people know that there is a free, open-source solution which _may_ also work for their use case.

Elasticsearch does currently support vector search through script score using dense vector fields, however I suspect they are still working on improving it and I prefer the Opensearch implementation for the time being https://www.elastic.co/guide/en/elasticsearch/reference/curr...


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: