dalys
Joined 16 years ago
Joined 16 years ago
According to Wolfram Alpha that is:
Single-spaced document: 1500 pages
Double-spaced document: 3000 pages
Book: 1028 pages
So around 1-5 books.
I'm assuming they're using OpenAI's tiktoken tokenizer. (??)
Idea one: Captchas are to become pretty useless as a "is this a human" tactic soon. Maybe it already is, I don't know. What other things could we think off to prove someone is human? I was watching Lex Fridman and Max Tegmark and they were remarking on how Twitter using payment as a differentiator between human and bot is actually really good. And maybe the only way we can reliably determine if someone is a human or not right now. Just by the virtue that having thousands of bots doing something, that suddenly costs $5 per event will deter most attacks. Integrating online identification systems from various countries could be one tactic (such as https://en.wikipedia.org/wiki/BankID that we use in Sweden to log in to basically any online service). New startup: Un-botable authentication as a service.
Idea two: Since captchas are useless, we'll be able to do bots that can do almost everything on the web. No need for writing automation scripts, headless browsers, regexp etc. Just feed real visual data from browser to GPT-4 (or MiniGPT-4 or similar). Give instructions like "You need to accomplish this task: Go to facebook.com and create a user account and be friends with 100 people and act like a human. Follow the instructions on the website.". Then let the bot figure out where to move the mouse and send click events, keyboard events etc. Obviously much slower than a scripted bot, but would be very hard to detect as a bot. Good for E2E testing also? New startup: WebGPT: Bots that can do any task on any website. TestGPT: E2E testing service. Just write test instructions like "Make sure the login works on /login" and you're done! And you thought flaky tests were a thing of the past... Would be kind of cool for UX testing also. "You are a 43 year old truck driver in Spain, please go to news.ycombinator.com and see how easy it is for you to register an account, log in, go to any news story and make a new comment. Report anything you think is confusing."
I used to also love piste maps as a kid, and still go skiing a few times per year. One of the best activities I know. Outdoor, exercise and spectacular views for a whole day. What a dream!
I think for a brand that does as simple and cheap cars as possible there's Dacia. They have the electric Dacia Spring which is also a small car. It does have a touch screen though. Only available in some select European countries.
Edit: Actually, the touch screen in the Spring is only the "Plus" version, so you can be touch-screen free! :) But the Plus version also has DC fast charging, so you probably want that anyway.
Because you know, HN is a marketing site for companies in Y Combinator and Y Combinator itself.
Every landing page you have ever visited is marketing.
Every GitHub README.md is marketing.
There's not a single business in the history of mankind that have made money without marketing. Because they wouldn't be allowed to show or tell anyone they existed.
I would just target "Surfboard (sporting goods)", "Surfing (water sport)" and similar and then let the FB algorithm do it's magic. I.e. create a very broad targeting for your ad and then let the ad run for a while and get results, and those results will let FB automatically fine tune the targeting for you. It costs some money but it will probably be better in the end. Unofficial rumours says you need around 50 conversion events on ad for the ad to be fully "trained", which can mean a couple of hundred dollars.
Remember the targets you add have a OR function between them, unless you use the "Narrow audience" button in the targeting.
If you really want to narrow down you could do "Surfing (water sport)" and narrow the audience and find some interest that is AND <insert teaching-related interest targeting>. People that work as surf school teachers could possibly have been identified by FB with some kind of teaching-interest.
Step 1. Get a huge email list
Step 2. Grep that email list for peter
Step 3. Do a custom audience in FB and run a Peter t-shirt campaign
I don't know any other way to do anything "dynamic" in Facebook Ads. And FB doesn't really like that advertisers creep out their audience, even though they can. I assume it's the reason behind their policy that you can't call out personal attributes. E.g. you can't say "Hey hetereosexual white person! Click here", but you have to write it as "Here's a fun website for hetereosexual white people".
I just double checked the policies and using a first name in the ad is actually not allowed, BUT I think you can get around that by having a general ad like "Cool name t-shirts" and then the image happens to have "Peter" printed on the t-shirt.
Source: https://transparency.fb.com/policies/ad-standards/objectiona...
1: Drench yourself in water.
2: Use soap where you have apocrine sweat glands ONLY, i.e. groin and arm pits.
3: Rinse soap.
4: Rinse hair and massage with your hands simultaneously.
5: Done.
I used to cover my whole body and soap carefully and rinse it off, but after realizing the foul smelling sweat glands only are in the groin and arm pit area, I realized it's unnecessary.
What about shampooing the air? I don't. I usually only do it before going to the hair dresser, once every two months or so. Does it get super oily? No, you can train your hair not be, if you don't shampoo. Here's how (for people ok with short hair). Just do it after cutting your hair really short. Stop shampooing and instead just rinse it really good. After the shower it's important to rub the hair with a towel really good though. It's much easier with short hair. You can also train long hair by just extending the period between shampoos slowly, there's a ton of YouTube videos about it.
Ex-engineer turned marketer.
Contact me at [email protected] I live in Stockholm, Sweden.