Other advantages to generating content: (1) fewer copyright issues. (2) No creators to pay, just GPU bills scaling with the use of the platform. (3) a much smaller critical mass.
Given these advantages I expect the current "social media" to be replaced with a new one, rather than them pivoting. The next big thing after tiktok might be something that only has generated content, where a last final bit of "social" is taken out of "social media".
Generating stuff is very cheap compared to building and training the model. When you have your model done you're incentivized to use it as much as possible. Maybe even considering the sunken costs.
Why can't it be the actual tiktok? By simply winning competition with humans i.e. whereas vast majority of humans see their pay go too low to bother to continue?
Yeah, I've been griping about LLM overconfidence for years, as somebody who is racked with self-doubt and second-guessing. On one hand, my own low opinion of myself made me a terrible mentor and manager, because having a similarly zero-trust policy towards my colleagues' work caused no end of friction (especially as a founder where people looked up to me for validation). On the other hand, i don't know very many top-tier practitioners that don't exhibit significantly more self-doubt than an off-the-shelf LLM.
Hence this blog post. I will say I've got a dozen similar tricks baked into my Claude config, but I'm not sure they've helped any.
I relate to this a lot— I treat my colleagues' work with suspicion and distrust not because I don't trust them but because that's also my stance toward my own work, like "what is this BS? Is it absolutely necessary? Can it be half the length by leveraging a library or ignoring error handling in cases where a panic/crash is no worse than a controlled exit?"
I find working with copilot is just catnip to someone like this because it's endlessly willing to iterate and explore the problem space, certainly well past the point where a normal person would be like are you for real can we just merge this and move on.
"Let it crash" is a sentence that gets attention. It makes a person want to know more about it, as it sounds controversial and different. "Let it heal" doesn't have that.
It also has a deeper philosophical meaning of unexpected software bugs should be noisy and obvious instead of causing silently corruption or misleading user experience. If monitoring doesn’t catch the failure, customers will and it can be fixed right away (whether it’s the software, a hardware error, dependency issue, etc.).
A web service returning a 500 error code is a lot more obvious than a 200 with an invalid payload. A crashed app with a stack trace is easier to debug and will cause more user feedback than an app than hangs in a retry loop.
When I had to deal with these things in the Java world, it meant not blindly handling or swallowing exceptions that business code had no business caring about. Does your account management code really think it knows how to properly handle an InterruptedException? Unless your answer is rollback and reset the interrupted flag it’s probably wrong. Can’t write a test for a particular failure scenario? That better blow up loudly with enough context that makes it possible to understand the error condition (and then write a test for it).
One of the biggest usability problems with Python dependencies is that the name you import might be different from the name that you use to install the package.
So if you find some script on the web that has an `import foo` at the top, you cannot just `pip install foo`. Instead, you'll have to do some research into which package was originally used. Maybe it's named `pyfoo` or `foolib`.
Compare that to for example Java, which does not have that problem, thanks to Reverse Domain Name Notation. That is a much better system.
"install name == import name" cannot work in Python, because when you `pip install foo`, you may get more than one top-level package. Or you may get a single-file module. Or you may, validly per the spec, get no Python code whatsoever. (For example, you could publish large datasets separately from your data science library, as separate wheels which could be listed as optional dependencies.)
The lack of good namespacing practice is a problem. Part of the reason for it, in my estimation, is that developers have cargo-culted around a mistaken understanding of `__init__.py`.
I don't think becoming vegetarian helps. From the article:
> The study found evidence that food processing is a likely source of microplastic contamination, as highly processed protein products (like fish sticks, chicken nuggets, tofu, and plant-based burgers, among others) contained significantly more microplastics per gram than minimally processed products (items like packaged wild Alaska pollock, raw chicken breast, and others).
If you look at the actual data, tofu had the third lowest concentration of microplastic particles per gram (0.03, vs 0.02 for pork loin chop and 0.01 for whole chicken breast).
The press release does a disservice to the study by referring to the highly processed group as a whole and not excluding tofu. For reference, the breaded shrimp and fish sticks were measured to have 1.2 and 0.26 particles per gram.
Regardless if a processed food like tofu has minimal plastic concentration, I would assume minimally processed whole plant food like beans and nuts would also have low microplastic exposure. The study found little total plastic from packaging, their evidence pointed towards the processing.
Did you account for calories? Beef has ~3-4X the calories per gram of tofu.
I would guess plants have more microplastics as they just pull things out of the soil and store them (why many plants have a ton of heavy metals). Animals at least have some systems for filtering and processing unwanted items.
Given these advantages I expect the current "social media" to be replaced with a new one, rather than them pivoting. The next big thing after tiktok might be something that only has generated content, where a last final bit of "social" is taken out of "social media".
reply