I've been experimenting with using ChatGPT for worldbuilding, including NPC dialog and stuff. I was rather satisfied with the results, that is until I saw your comment. The text it generated for you is very similar to what it gave me. The style is immediately recognizable, the structure is extremely similar, and in case of "For I am Grimgor Blackfist, the most feared warrior in the land, and no one can stand against me." I literally got the same sentence with a few words changed.
I wonder if it's possible to customize the prompt in order to make the output more unique otherwise everyone who is using ChatGPT for fantasy writing will end up with very samey and super recognizable style.
I consider obligatory 20% tipping as that weird American thing along with guns and (absence of) healthcare. People have elaborate arguments and discussions about them, and the side in favor can look very rational and plausible, and their hypotheses can even have a lot of explanatory power, but they are easily refuted by looking at basically every other developed country.
I travel a lot and eat out a lot and I haven't noticed any discernible difference in service between the countries with obligatory, optional and prohibited tipping.
That's interesting you consider tipping an American thing, I'm guessing you aren't very well traveled as you clearly haven't been to Canada or the Middle East.
Same, never got any headaches from any screens. Regardless of the type of the screen, lighting or viewing position. Have been using computers for almost 25 years, with many 10h days and occasionally 15+ hours.
As I got older I started getting neck pain though, if I sit too awkward. But still no headaches ;)
I have been developing a hobby project (AI powered document search) for a few months and was in sore need of a frontend. My frontend development skills however are stuck in late 1990s and I have zero skill with anything but plain HTML and a little bit of JS. Several times I tried learning React, reading tutorials, watching videos, but the whole idea of it was very removed from how I learned to code, so I gave up every time.
Today, I asked ChatGPT to develop the React app for me. ChatGPT guided me through the entire process starting from installing npm and necessary dependencies. The commands it suggested sometimes didn't work but every time I just copy-pasted the resulting error message into ChatGPT and it offered a working solution. I gave it the example of JSON output from my API backend and it generated the search UI which, to my surprise, worked.
My wet dream for the past few months was to implement infinite scrolling for my search. Again, after hours of google searches, tutorials, etc. I just gave up every single time. Not today. I asked ChatGPT to add infinite scrolling to my app. It wasn't easy. It didn't produce a working app immediately, it took a couple hours of conversations: I had many questions how different parts of React worked, how to fix errors etc. etc. In the end however, I had my working search app, and with infinite scroll to boot!
I haven't done a single google search or consulted any external documentation to do it and I was able to progress faster than I have ever did before when learning a new thing. ChatGPT is, for all intents and purposes, magic.
It doesn't sound that much different than going through Google and Stackoverflow though, is it? In a few hours of googling you can probably get something working if you are an experienced dev.
The crucial difference is that at no point I felt I was stuck. I could paste any line of code into ChatGPT and ask it to explain it. Practically every time I got a meaningful and valuable explanation, moreover the explanation was in the context of my code. Similarly all functions it generated were matching the context of my code so I could just copy and paste it and it just worked, most of the time.
Rather than going through Google and Stackoverflow it felt like working side-by-side with a moderately competent developer. Mind you, I have tried the google-and-stackoverflow method before for the exact same thing, and failed every time ;-)
It's very different because it will answer the question you asked, rather than answering a question that matches a substring of the question you asked like Google will.
Google apparently uses BERT to actually answer the question you asked … and an obvious incarnation for this sort of tech is probably going to be further integration into google . Makes sense doesn’t it .
BERT is a simple model that is not capable of answering questions in this manner. For very simple things it might help with that answer box at the top, but that's not what I meant.
Yeah I had a slightly similar experience as OP, though simpler. I asked it to automate a basic task, something I hadn't done before. I managed to do it with ChatGPT only, no other resources.
That said - ChatGPT did make mistakes, there were inconsistencies in its instructions, it didn't recognize certain bugs (I had to find them myself). _But_ there was something about the chat-based interaction that to an extent helped me preserve flow (maybe a bit like pair programming?).
I do think that if I had set my mind to it, I would have been faster solving the task with Google, and to some extent I went through this exercise just to test ChatGPT.
ChatGPT helped me close several business deals. I am now a mega millionaire thanks to ChatGPT! Before, I wasn’t able to find the basic info on how to close multi million dollar deals, and I tried all kinds of stuff. But ChatGPT helped me through that. On the calls - whenever I didn’t know what to say next, I would just read off what ChatGPT was responding to the customers, and to my surprise, it matched what they wanted to hear! And they started responding back and forth with it as if it was always in the plan! In the end, they didn’t exactly say “shut up and take my money”, but they did seem to express deep concern that I wouldn’t have availability for them, and essentially agreed to all the upsells very quickly.
I recommend ChatGPT to anyone who wants to close customers or save their marriage. Just say whatever ChatGPT is telling you… even if that means using one of the new “personal” beamed sound into your skull things. You’ll have superhuman ability to vibe with anyone and outcompete everyone who relies on just “their own experience”.
- Written by ChatGPT in response to a prompt.
In the end it added, “no one will ever believe you” in all lowercase.
Hopefully ChatGPT doesn't refuse to answer my question because of some reason appreciated only by people who get too much pleasure from the StackOverflow moderation game
That “it’s just a different search front end”… but I think after more experience with it I disagree.
At its worst, it’s “multiple searches” at once.
Example1… I wanted to find a CAGE for code a specific military mfg. I only had the last 3 digits. I asked for CAGE codes that match and got all the answers instantly. I could have searched this, but it would have been multiple searches.
I asked for the etymology for the Swahili word for trapezoid… again, multiple searches. If I could have found links to the Arabic root of some Swahili words at all.
That’s it’s worst case, convenient multiple searches. The better case is the UX of a conversation is powerful for the user, in a way we are just learning the words for.
But OP explicitly said they had little experience in this area. They also presumably have a technical career and are awash in the ways of Google. I'm in a similar situation to GP and have gone down that very path with React and whatnot. It's like you're starting a rodeo off the bull and have to figure out how to get back on. It's a terrible experience and you're left infuriated at a faceless collective that carelessly makes getting started so difficult.
An experienced dev can work something out even if its not his main stack. I could probably get something very basic done in Swift or Android despite never doing it. Experienced devs are just good in reading documentation and having a general understanding of how things should work.
Totally, and it seems like ChatGPT almost does the experienced dev work here for a junior developer - impressive.
But much like you need to cause some stress to a muscle to cause it to grow, junior developers historically needed to get experienced at finding some of the solutions to their own pain to become really experienced developers...
It seems like ChatGPT may cut that form of growth out of the cycle...
I wonder about the implications of this... Junior devs will progress more quickly, but they will also grow less of their own skills and be very reliant on ChatGPT - like an exoskeleton for their development skills.
I guess that will be great for OpenAI if they can charge a hefty monthly fee...
I'd still rather max out my own skills before relying on an exoskeleton (once I've maxed myself out sure, give me the exoskeleton, and let's see what it can do), but maybe I'm too old fashioned...
replace chat GPT with slide rule and calculator and you have the endless arguments made against calculators in the 70s. change it to typewriters in word processors and you have all the hand ringing in the early '80s about how writing was going to be destroyed by easy copy paste. That isn't a proof that your argument is wrong of course, but it is very suggestive to me.
I typed this with text to speech, another thing we were confidently told would never work
This is blatantly untrue. Venezuelan bolivar has lost 100% of its value (within a rounding error) - current exchange rate is something like 800 trillion of old bolivars to USD. Argentine peso is doing somewhat better than bitcoin this year, only losing 40%, but you don't have to go far back for it to become much, much worse. In the entire history of bitcoin there are very few periods when you'd come out on top holding argentine pesos rather than bitcoins.
Evidence for existence of a (il)liquidity premium in the stock market is very weak. What this means is that the market has been liquid enough for a long, long time.
Consider high frequency trading. Does it really matter for capital allocation that you can sell your Google stock for a fair price in 10ms instead of 20ms? There are hundreds of PhDs working on algorithms and billions invested into infrastructure to make those 20ms into 10ms, which I'd argue is a byproduct of how exchanges work and doesn't serve humanity in any way.
> Evidence for existence of a (il)liquidity premium in the stock market is very weak. What this means is that the market has been liquid enough for a long, long time.
That's not entirely true. Liquidity is often used to mean 'ability to exit your position at the most recently traded price', or something similar. However, i'm using a slightly expanded definition, which means: exit your position at its intrinsic value. The ability to sell your stock for what its actually worth is extremely valuable, and statistical arbitrageurs help you do this.
> Consider high frequency trading. Does it really matter for capital allocation that you can sell your Google stock for a fair price in 10ms instead of 20ms? There are hundreds of PhDs working on algorithms and billions invested into infrastructure to make those 20ms into 10ms, which I'd argue is a byproduct of how exchanges work and doesn't serve humanity in any way.
Here I agree with you, although I think it's important to understand that a lot of high frequency trading is actually making markets efficient. That is to say, the service being provided by HFTs is truly valuable. What is not super valuable is squeezing out that last marginal millisecond. The energy poured into the last millisecond is indeed deadweight loss, but I think the amount of energy being spent there, while very large in absolute terms, is not so large relevant to the financial industry writ large.
So yes, I agree that the competition at the margin in HFT is probably not productive, it is a byproduct of the necessary service that HFTs provide and the dynamics of the environment in which they provide it. I also think that any 'solutions' to the problem of that energy expenditure are likely to make everyone worse off over all, not better. You really want the HFT game to be a winner take all market, in the way that it currently is, precisely because competition is deadweight loss.
If you do things like introduce purposeful stochasticity into order submission (for instance), you're going to make the competitive equilibrium more multipolar, so that instead of one dominant firm taking it all, you have a bunch. And that's actually worse for everyone. What we want is basically 1 HFT firm that is the best and does a good enough job that nobody else tries to compete, and earns a reasonable profit for providing the efficiency and liquidity that they do. I think the competitive environment is actually converging on that, even though it may not look that way from the outside.
I usually don't go out of my way to pay with bitcoin, but recently needed to buy a PSN prepaid card and every website I tried either declined my 2 credit cards and paypal outright or requested some ridiculous verification (record myself on camera holding a passport to my face? To buy a $10 prepaid card? No thanks)
In the end I stumbled upon a shop that accepted bitcoin and got my PSN card in the time it took to get one network confirmation.
I highly doubt it really mines bitcoin because with the current difficulty a typical modern graphics card can expect to mine less than $0.01 worth of bitcoin a week if left running 24/7 on full load.
Edit: what I meant is that EpicScale in all likelihood mines another cryptocurrency, not bitcoin. Especially that it seems to use CPU which is utterly useless for bitcoin mining (you won't even get $0.01 in a year with the latest i7!)
Solving math problems for weather prediction, physics simulations, cryptography (including cryptocurrency mining) and more has real world value. We solve these problems on behalf of our trusted partners, and donate proceeds to your favorite charities.
The key words there being including cryptocurrency mining.
I wonder if it's possible to customize the prompt in order to make the output more unique otherwise everyone who is using ChatGPT for fantasy writing will end up with very samey and super recognizable style.