These visualizations are cool, but I feel only a little misleading. Any 2D relation can be plotted in 3D, and setting z=0 shows the original line in 2D. The whole notion of fuzzy is really just saying "let's make a heat map of how far z is from 0". That said, I can see how for some people, thinking about the roots of a function in 3D this way _could_ help unlock deeper understanding of where your roots are at than just staring at 3D graph.
Also, the visualisations are to some extent an artifact of how you arrange the equation: rearranging terms between the left and right side won't change the line but it will change these visualisations. The first example illustrates it nicely: the 'hole' only exists because of a redundant division by (x^2 + y^2) on either side of the equation.
> The predictor is like a barista who starts making your usual order the moment you walk in. If you are a regular, this is fantastic: the coffee is ready when you reach the counter. If you order something random every day, the barista keeps pouring drinks into the sink.
I laughed out loud reading this. Interesting writeup. I wonder what kinds of tricks like this exist for computation graph compilers like JAX.
We purchased a Chinese-made projector from Amazon, which was surprisingly inexpensive (~40 USD). Upon connecting it to the internet, it placed a constantly running feed of ads on the corner of the screen, even while movies were playing. There was no way to disable it either. Even though it's not a stick, it's a similar principle.
I remember reading an analysis on one of those projectors; the author found a residential proxy running on their device. I would recommend keeping these things off the internet.
Dunno if this is the same issue, but someone found malware in their projector. I'm not sure about the accuracy since the report is blatantly AI generated: https://github.com/jrm360seclab/aodin-vo1d-malware
To be fair, everything is Chinese made. I would be even the Apple TV and NVIDIA Shield are made in China and if a state actor is determined to get a malicious payload in....
To play devil's advocate, when someone says "Chinese made" they're usually well aware of your point, and are more using it as a common way to describe product mills spitting out countless devices with dubious quality or configuration.
Of course theres good products made in China, and plenty of entirely Chinese brands killing it doing their thing.
It's not racism at all to be weary of (any) political system, its overreach and the incentives of the people living in it, be it China or America or Russia.
Its based on the most common heuristic people have developed in regards to the phenomenon. What do you think about 'alphabet soup company' instead, referring to the tendency for names to be a mix of random letters? Otherwise, you can try and create a better term for 'unaccountable third parties using US platforms to dodge liability for their product made out of the cheapest components and software possible' and see if that catches on.
Yes it is also the US companies that are a problem but these are two separate problems and need different terms.
This isn't about state actors though. There's a world of difference between a name brand (possibly even a Chinese one) versus what I would term "chineseum". It's nothing to do with China per se and everything to do with purchasing from the extreme low end of the market. It just so happens that the vast majority of that segment is manufactured in China at present.
I think normally when people say Chinese made in this way, what they're really communicating is that there's no (meaningful) brand. All they know about it is that it is from China.
I dare not ask why you would do such a thing, instead, I will simply ask if you now think the reason was good, and I will hint at you that if the reason was "convenience", then you should answer "No".
The streaming sticks the article is discussing basically need network access to function. They might support streaming from local media sources and file shares too, but that is also done over WiFi. Unless you have a properly firewall controlled home (very few people do, I'm pretty nerdy and most devices on my network can just NAT to the outside these days) then just giving it a WiFi connection gives it access to the wider network from your location.
You'll probably find the projectors are pretty much the same hardware and OS as the sticks except with the projection device added where the stick just has an HDMI output. It might have HDMI-in too so it can just be used as a screen for another device, but there are definitely some units out there that are network-play-only.
You aren't wrong about giving cheap crap like this access to your network (and via that the public network) is risky, but that convenience you (and I) would say no to is exactly what they are bought for.
At least in capitalism you have the choice to look for a malware-free alternative. 100% USSR, had it survived to the IoT era, would penalize you for not having a state-mandated surveillance device on at all times.
I agree fully with your assessment of USSR but basically any nation state with the power does such things.
Show me a COTS smartphone where the end-user can burn the OTP fuses for his personal public key, so they can have it boot their own custom signed firmware, and control exactly what runs in TrustZone's SW Secure World?
You can flash yourself GrapheneOS with your own keys for the bootloader. Then again "I can't make sure all manufacturers aren't in collusion" when FBI sues Apple and others (and fails) over suspects' phone access is quite different from "every device sold in the country must have government malware", as it is in China.
I can't find a device in the USA that doesn't come with government malware. Is this another instance of the USA accusing China of everything the USA is doing (like with the credit scores)?
For instance, if you use an Apple phone and the government wants to see the pictures you took, they can just get a copy of them from Apple using iCloud.
Yes, using a warrant from a court. If the government wants to break into my house they can also do that with a warrant. Not exactly the same as having a key to my house though, and very far from a malware.
Also to be more specific, In apple ecosystem I can opt for encryption where they can't
You're free to opt out of icloud/gcloud and the government will have a tough time extracting the pictures from your phone, as multiple lawsuits have shown.
Get the sarcasm, but of course there is scam outside of capitalism. Its just that the capitalistic model almost turns it from an inconvenient bug into a mainline feature.
Not saying there is an absolute perfect alternative, anyone who says that is usually shoveling smoke, but there are flaws with this economic model to be addressed.
Not true at all. I'm from Poland which was occupied by communist for a long time, and I can guarantee you - the amount of scam we had under that rule was orders of magnitude larger than what we have now.
Yeah I did forget about that. When you flatten the pay structure across the board, it makes bribes and scams so much more desirable. But also, communist structure in practices is sort of the total opposite of capitalism at a distance.
It was said that Karl Marx was completely right about Capitalism and completely wrong about Communism. And that is fairly accurate, both have big flaws.
Most times, the opposite of one bad idea is another bad idea.
I think it's really far fetched to say capitalism is bad idea. It's great system, it has some problems, but the upside is so big and alternatives are so bad that it's really unfair to call it bad system.
I think that encouraging corporations to destroy our environment (e.g. climate change) as fast as possible to maximise profits is a very good reason to call it a bad system. Yes, some goods and services become much more efficient, but now we're all going to have to pay the price for it.
Yh, the end of civilisation is a good thing after all, so enabling greedy fuckers to accelerate all life on Earth ever more rapidly towards destruction has to be good ...
Most American ex-pats don't really understand that the thing that makes ex-pat life so attractive is that, for most of people's lives, being American in a foreign country has traditionally conferred a wide range of benefits (this is most clearly exemplified by the way Americans living in a foreign country refer to themselves as "ex-pats" not "immigrants"). The ex-pat solution assumes American exceptionalism as its foundation.
Historically Americans have benefited from income asymmetry and a fairly wide-spread desire by foreign nations not to cause too much legal trouble for US nationals abroad.
I have quite a few friends that do live, quite happily, abroad. But the common pattern for them is a.) fluency in the native language b.) historical association with the country c.) fairly large cash reserves so they can ignore any economic problems these countries are facing.
That sounds more like FIRE / geographic arbitrage than expat in the traditional sense.
What people often mean by a "true expat" is something completely different: someone sent abroad by their employer, usually with a generous expatriate package (home-country salary, local allowances, housing, private schooling for children, etc.)
More broadly, though, expatriate simply means someone living outside their native country. Ex means "out of", and patria means "native country" or "homeland". It's that simple.
So the word itself isn't limited to wealthy retirees or corporate transfers. All immigrants are expatriates, although not every expatriate is necessarily an immigrant.
Most places have their own politics. What differs is how often they come up. As a foreigner you're usually spared the involvement in those discussions because people think you're not interested and don't want your outsider opinion anyway.
People also embed themselves in different communities when they move anywhere, even to a different city or state in the same country. It's a clean reset.
It doesn't always last forever. I know several people who tried to move somewhere, including internationally, when politics got heated in 2016. Most of them came back eventually with a realization that politics is everywhere, it's just a matter of how much you're embedded into the places it's discussed.
Politics is everywhere, but the state of US politics today is exceptionally bad by Western world standards. I hate the government in my country, but their corruption and incompetence is nothing compared to the Trump administration.
In that sense most EU countries are a positive upgrade.
I (American) worked in Spain (Cáceres, Extremadura) ~2015-2017 in tech. It was a wonderful experience. Extremely talented, hard-working, and friendly co-workers. Great health-care and education systems. I think since then rising housing prices partially due to migration have become an issue, but it's a really, really nice place.
To add context Extremadura is a member of the "poor" Spain. To US people could be useful to think on a sort of New Mexico.
Pros: Great food, interesting cultural past, only one language to deal with and not complicated accents to grasp (more important that most people think), gorgeous wild areas, uncomplicated people, maybe a little on the introverted side at first, but solid gold after a while.
Cons: Risk of poverty sadly high, bigger than many US states (but with better government support and healthcare). Harsh continental climate very hot and very cold. Not for everybody (but US has plenty of places with similar or worse weather). The trains and communication roads are also under-average for the country and many people don't really speak English.
In many of the non highly touristic places you can live well if you can adapt to the cons. Housing prices are lower, life expenses cheaper and buying a house should be affordable with a decent job (Don't try this in Barcelona or Madrid). Portugal is close, and is even cheaper, to the point of some people living there and working in Spain. To support the same standard of living in Barcelona, Valencia or Madrid you need to plan in advance, to stomach the stress that unavoidably come with big cities, and earn much more.
In Spain if you can speak English well you will be automatically seen as a great researcher.
More pro: Spain may be one of the best places in Europe to raise kids (culturally, especially), though a good chunk of the country has emptied out of its young people - so it can vary by region. It's a shame that their birth rate has gotten so low. Crime overall is very low, especially if you're not tied directly to it.
More cons: You will eventually have to shift to a very different customer service climate and hours. The development levels can vary quite a bit, especially "modern" infrastructure like internet outside of the major cities (maybe that's gotten better in the last 5 years). Bureaucracy of some institutions (government/finance) can be extremely frustrating.
Regarding internet infrastructure, the fiber optic network seems to be quite robust these days. There are still places where it isn't available, but compared to e.g. the US it's another world entirely. Even small cities typically offer 1 or sometimes 10 Gbps hookups in many areas.
Regarding bureaucracy I'm afraid that there has been no such progress. Extremely frustrating that even when I do everything right, I get told to come back another day for my trámite because they just...haven't done it! Sorry!
Also American, and I'm interested why you praise the education system. I have a child who will be entering the Andalusian system at some point and although it seems better than my Oklahoman system, that is, uh...damning with faint praise.
Some of my non-Spanish European colleagues also have commented that the education system is kind of "good not great" especially compared with other Western/Central European countries. However, I understand the Spanish system to be somewhat federated; perhaps the difference between Extremadura and Andalusia would explain the difference in opinion.
gotta be ready for the crazy heat in the summers, they don't call it extremadura for nothing. unironically best lodging is rooms that were dungeons in the castle.
Recently, I made an Arduino UNO that I showed to have better switching characteristics than a commercial board. It was a great project to help me understand how seemingly inconsequential routing practices can lead to issues down the line.
You should be able to improve it further with a 4 layer board, as you can have the signal lines closer to their reference planes, and can more easily avoid signals crossing each other and breaking reference planes.
Liquid crystal elastomers will most likely never be used in humans because, in order to drive the phase transition (mematic mesogens going from isotopic to anisotropic phase) necessary for macro scale work, the LCE has to be heated well beyond 100C. Even in non-thermal contexts, you need kilovolts to influence a doped bulk LCE. I just don't see it happening.
to be fair, the approach is usually covered in snowpack for most of the year, so impact is minimal by foot traffic. However, most of the protection is fixed, which could have lasting effects if something were to rip out.
For other mountains with dry summits in the summers, I would agree: the effects of erosion are frightening
Today I scheduled a dentist appointment over the phone with an LLM. At the end of the call, I prompted it with various math problems, all of which it answered before politely reminding me that it would prefer to help me with "all things dental."
It did get me thinking the extent to which I could bypass the original prompt and use someone else's tokens for free.
And this is another easily solved problem by someone who knows what they are doing…
Voice -> speech to text engine -> LLM creates JSON that the orchestrator understands -> JSON -> regular code as the orchestration -> text based response -> text to speech
Notice that I am not using the LLM to produce output to the user and if the orchestrator (again regular old code) doesn’t get valid input, its going to error. Sure you can jailbreak my LLM interpretation. But my orchestrator is going to have the same role based permission as if I were using the same API as a backend for a website. Because I probably am
Source: creating call centers with Amazon Connect is one of my specialties
Sure - but does this have the context of the original question that the user asked? If not it seems that it isn’t really conversational and more of a “compiler”.
How would something like “I want an appointment either on Monday afternoon after 4pm or one on Tuesday before 11am” work?
Unless all the parameters given by the user fit within the constraints of the json format then the LLM would need the context of the request and the results to answer properly, would it not?
This is a constrained space. I would do the naive implementation at first and then talk to the humans (like you) and then my JSON definition would include a timespan type field.
My orchestrator would then say “I have these times available [list of times]. What time would you like?” and then return a specific LLM prompt to parse the information I need once the user responds. But I would send that exact text to the user. Yes I’m purposefully constraining the implementation where the LLM is never used for output and never directly controls the backend
There is also the concept of “semantic alignment” where you ask the LLM to generically answer the question - “does the users answer make sense with regard to the question” as a first level filter that only returns true or false. This is again a constrained function that you pass in the question and answer to the LLM and if you get something besides true or false your code errors.
The purpose of an LLM or even before that an old school intent based system (see my link) isn’t perfection it’s “deflection”. The more that you can handle through automation the less you have to bring a human in. An American based call center when a person is an agent costs from $3–$7 a call fully allocated. An automated call can costs tenths of a penny.
Of course that doesn’t include the cost of the accepting a call in the first place over a 1-800 number and in my case the price that AWS charges per minute for Amazon Connect
> This is again a constrained function that you pass in the question and answer to the LLM and if you get something besides true or false your code errors.
Code erroring is fine for code, but what is the user experience here? Some sort of “computer says no” generic response, or something more contextual?
I’m trying to picture what the user says and hears as a response to an off-the-beaten-path question. Is it just “I don’t understand, here’s how to phrase it?”.
If there is an issue, they are transferred to a human operator. “I’m having trouble understanding you, let me transfer you to someone who can help”. On the CSRs screen, they will see the conversation that has taken so far.
There is also sentiment analyst built into the prompt so it can detect a negative sentiment and automatically short circuit the process and transfer to a human.
NLP doesn’t have world knowledge and with one prompt, I can support almost any language. Of course the speech to text engine is specific to the language
haha for sure some one has made a little aggregator for this and saving tokens. I bet you gotta dig for a while though before you find a company exposing Opust 4.6 to customers and not flash 2.5 lite
reply