Or shard it - divide your objects up based on some criteria (hash the name of the object, use the first N digits of the hash to assign to a shard), and distribute them across multiple redis instances. Yes, you then need to maintain some client code to pick the right redis instance to fetch from, but you can now pick the most $/memory efficient instance types to run redis, and you don't have to worry about introducing disk read latency and the edge cases that brings with it.
Edit: looks like redis has some built-in support for data sharding when used as a cluster (https://redis.io/docs/latest/commands/cluster-shards/) - I haven't used that, so not sure how easy it is to apply, and exactly what you'd have to change.
To add another anecdote to this thread, a few years ago, I picked up an old Consew 226R compound walking foot machine to do canvas work for a boat. That machine was a beast once I tuned it. Fast, powerful, smooth, quiet, and with a servo motor, had awesome slow and high speeds, would punch through anything. Its only real downside was the size, and it only did straight stitches. For canvas work, that was everything I needed.
I later picked up a Sailrite LSZ to do some zigzag work on sails. Compared to the Consew, the sailrite feels really clunky, even with the Sailrite servo motor upgrade. It's relatively rough and noisy compared to the Consew, and has less power. Main upsides are the size, and the zigzag stitch for sail work where you need some stretch.
I've ultimately kept the Sailrite because it's still pretty solid, and can be stored on the boat where I need it the most. If I was doing upholstery or canvas work in a permanent shop, I would've kept the Consew, hands down, it was a much more capable and pleasant machine.
The walking foot feature is very useful for heavy materials. I tried sewing light-weight materials on both of these machines and it was awful (bunching, tearing on the foot, can't use really fine needles, etc).
If you understand what it is doing, then you don't. But the layman will just see a computer that talks in language they understand, and will infer intent and sentience are behind that, because that's the only analog they have for a thing that can talk back to them with words that appear to make sense at the complexity level that ChatGPT is achieving.
Most humans do not have sufficient background to understand what they're really being presented with, they will take it at face value.
I think this is partly explained by most of the marketing and news essentially saying "ChatGPT is an AI" instead of "ChatGPT is an LLM."
If you asked me what AI is, I'd say it means getting a computer to emulate human intelligence; if you asked me what an LLM is, I'd say it means getting a computer to emulate human language. The word "language" does not imply truthiness anywhere near to the extent that the word "intelligence" does.
You could reasonably describe it as "human language emulator" back when people were using GPT-2 and the likes to compose text. But what we have today doesn't just emulate human language - it accepts tasks in that language, including such tasks that require reasoning to perform, and then carries them out. Granted, the only thing it can really "do" is produce text, but that already covers quite a lot of tasks - and then of course text can be an API call.
Interesting perspective. I'm still learning about what it really is, and I'm having trouble marrying the thoughts of a parent commenter with yours:
> ... does what it is engineered to do pretty well, which is, generate text that is representative of its training data following on from input tokens. It can't reason ...
versus
> ... doesn't just emulate human language - it accepts tasks in that language, including such tasks that require reasoning to perform ...
Maybe a third party can jump in here: does ChatGPT use reasoning beyond the domain of language, or not?
Nobody can definitely answer this question because we don't know what exactly is going on inside the model of that size. We can only speculate based on the observed behavior.
But in this case, I didn't imply that it's "reasoning beyond the domain of language", in a sense that language is exactly what it uses to reason. If you force it to perform tasks without intermediate or final outputs that are meaningful text, the result is far worse. Conversely, if you tell it to "think out loud", the results are significantly better for most tasks. Here's one example from GPT-4 where the "thinking" effectively becomes a self-prompt for the corresponding SQL query: https://gist.github.com/int19h/4f5b98bcb9fab124d308efc19e530....
I think the real point of disagreement is whether this constitutes actual reasoning or "merely completing tokens". If you showed the transcript of a chat with GPT-4 solving a multi-step task to a random person off the street, I have no doubt that they'd describe it as reasoning. Beyond that, one can pick the definition of "reason" that best fits their interpretation - there is no shortage of them, just as there is no shortage of definitions for "intelligence", "consciousness" etc.
> Most humans do not have sufficient background to understand what they're really being presented with, they will take it at face value.
Trust in all forms of media [1], and institutions [2], is at an all time low. I'm not sure why that distrust would go away, with reading the output from a company that's clearly censuring/fudging that output.
I think any "damaging" trust would, clearly, be a transitory phenomenon, since the distrust in media and institutions is from an ability to see BS. I don't think some sentences on a screen will be as destructive as some think, because people don't appear to be as stupid as some believe.
You always have the option of creating a map style and using Mapbox's Raster Tile API - although that does mean you probably can't dynamically adjust your raster images in time for a request to be served.
How are you using the Static Maps API currently where you would be able to send a POST request and display the image back to the user? Most use-cases for the Static Maps API involve the HTML <img> tag or something similar, which all send GET requests.
My company is in a similar situation - have large teams in several eastern European cities (primarily Minsk), and a few long-term remote teammates in Ukraine. A couple have fled (from Kyiv) with children, others are worried about being conscripted. It makes me want to vomit just thinking about it. I'm amazed many of them are still on Slack and making PRs.
We're fortunate some of our leadership has quite close ties to the area, so our company's response so far has been extremely understanding.
> “I'm amazed many of them are still on Slack and making PRs.”
I was talking with a friend yesterday about his teammates from Ukraine (we’re both outside Ukraine). They said that working is the only thing keeping them sane. This allows them to stop focusing every second at the war outside their homes.
Economics has also already given us lots of information on how much money people are prepared to sink into transportation - the owners of the self-driving cars/rental companies will attempt to extract that value.
I don't doubt there will be a shift in the overall economic picture, but it may not be as large for the car _user_ as we would hope. It will need to be better/cheaper in some dimension for it to get adopted, but the rent-seeking vultures are always circling, spoiling the utopia :-(
> For example,
an under-bark temperature of –37°C will
kill 50 per cent of a mountain pine beetle
population, even in mid-winter; however, a
low temperature of –20°C in the fall, be-
fore the beetles are prepared for winter,
or in the spring, when beetles are start-
ing to become more active, will also kill
beetles if it is preceded by temperatures
above 0°C. The relatively warmer temper-
ature causes the larvae to start to lose its
natural antifreeze.
I would love to see something like this written by someone with a young family. I find myself living JIT subject to the whimsical demands of small children for extended periods, and often wonder if it's even possible to re-arrange life in this context to try to re-capture some sanity.
Usually I just give up and bathe myself in the chaos, hoping that it's just a few years and then some semblance of sanity will return naturally.
Self-employment is the only answer I have seen provided. From memory, some people said they were able to transition into it, some others implied they felt they only had the chance at contract work due to a long onramp they made for themselves working at it earlier in their life. Hopefully we'll get some anecdotes.
Well, I suppose there is teaching your children discipline; depends on what you mean by whimsical demands.
You're raising a great point. While I'm not living together with my partner, we do have a little family and I'm very much with you that it's super hard to get anything done with kids around.
Unless you get up super early, stay up super late after kids are in bed, or use the time when they aren't around (e.g. kindergarden), I also think that it's almost impossible to fully focus on something for a longer period.
I guess it kind of boils down to finding the right routines. That certainly doesn't make it easier...
The sad news is that the credit card scammers have ruined this - automated credit card testing often looks for small donation pages like you're suggesting, then uses them to check if stolen card numbers are valid. This leads to chargebacks, loss of reputation by the donatee in the eyes of the card processor, and a whole bunch of wasted time. One of the "services" that organizations like Patreon actually perform is to take on this risk on behalf of their clients.
I guess bitcoin/etc payments are still OK, but except in some small niche subject areas, aren't generally accepted enough to make the direct donation setup worthwhile.
Im 28 and have a vague recollection of cheque books. I can assure you no person under 21 in my country can say they have actually seen one. Do they still exist?
They are quite common in the U.S. Vast swaths of our financial system are still stuck in the mid-20th century or earlier. A check is the only way (besides cash) to transfer money in the U.S. without incurring some sort of fee or other overhead.
It must be because I live in a developing country or maybe the lack of infrastructure 30 years ago means we can hop and skip straight to almost fully digital. These days going to the bank is akin to corporal punishment. Just give me an app or you're dead to me.
this. in one of the projects I am working on, we have this problem exactly. automated credit card testing problem for small donations.
we use Stripe API there. And Stripe is just suggesting to pay more for their radar check, which in our opinion, put more burden on both donors and the organization.
Edit: looks like redis has some built-in support for data sharding when used as a cluster (https://redis.io/docs/latest/commands/cluster-shards/) - I haven't used that, so not sure how easy it is to apply, and exactly what you'd have to change.