afair I've mostly only used fold when doing maths not covered by the standard sum or product. Fold is similar to map reduce but it's just one expression.
the point seems to be making a small binary to run language model.
How useful is that?
From a functional perspective, I guess not very but the model can be improved.
From a performance / cost perspective, also not very because most of the cost is in training the model and small binary isn't necessarily indicative of quick.
I guess it's just kind of interesting that it doesn't take that much code to run the model.
That won't help. Their TOS and policies are vague enough that they can terminate all accounts you own (under "Use of multiple accounts for abuse" for instance).
I used to be "[email protected]" in the early days of Hotmail, of course I was also a 6th grader (true story). One day they unceremoniously closed my account without any possibility to recover mails.
That day I learned an important lesson: pimpin' ain't easy.
My buddy lost his Gmail account because of a heart attack followed by a string of events that google ‘AI’ considered too risky to allow the account to live.
If their fraud AI is wrong there is now human to talk to.
Even if it is warranted on their part, the 1% false positive will be detrimental to those affected. And we all know there is no way to reach out to them in case the account is automatically flagged.
I asked Gemini about banning risks, and it answered:
Gemini: Yes, there is a potential risk of your Google account being suspended if your SaaS is used to process inappropriate content, even if you use Gemini to reject the request. While Gemini can help you filter and identify harmful content, it's not a foolproof solution.
Here are some additional measures you can take to protect your account:
* Content moderation: Implement a robust content moderation system to filter out inappropriate content before it reaches Gemini. This can include keyword-based filtering, machine learning models, and human review.
...
* Regularly review usage: Monitor your usage of Gemini to identify any suspicious activity.
* Follow Google's terms of service: Make sure that your use of Gemini complies with Google's terms of service.
By taking these steps, you can minimize the risk of your account being suspended and ensure that your SaaS is used responsibly.
---
In a follow up question I asked about how to implement robust content moderation and it suggested humans reviewing each message...
So a convenient blah blah blah about all the nice things you can doto avoid Google's brainless algorithmic wrath, but which may simply not work anyhow because even by following all rules in good faith, you still get banned one day, as has happened to many, many people with zero recourse.
As an attacker, instead of DDoSing a service we could just upload a bunch of NSFW text so Google kills their infra for us.
Other providers, like OpenAI, at least provide a free moderation API. Google has a moderation API, that after the free 50k requests it is more expensive than Gemini 1.5 flash (Moderation API costs $0.0005/100 characters vs Gemini 1.5 flash $0.000001875/100 characters).
The first point is reasonable enough, but the point still stands you can't find the same size house in the city for the suburb price.
Most cities simply don't have more than a handful of spacious houses with big yards.
Your second point is invalid, as you're arguing against his assumptions. It's only possible to argue againt someone's logic, arguing someones assuptions is the same as calling names. I like woodwork and have alternatives.
You don't need the same space. That is the point. Yes, if I lived in a suburb I would also want more space because everything else would be harder to do.
I'm not arguing against their assumption. I said most, that isn't them. This is exactly why I didn't want to elaborate, so I won't.
it loses trust with customers when the simple setup is flawed.
S3 is rightly built to support as much egress as any customer would want, but wrong to make it complex to set up rules to limit the bandwidth and price.
It should be possible to use the service, especially common ones like S3 with little knowledge of architecture and stuff.
> it loses trust with customers when the simple setup is flawed.
S3’s simple setup (which denies all public access) is not flawed in the manner being discussed here. Allowing public direct access to an S3 bucket is a supported option, but for years has been both non-default and strongly recommended against.
are there debugging tools specifically for situations like that?
do you just write code to test manually?
How do you ensure dev builds don't break stuff like that even without considering debugging?
The most useful tool is a full tracing system (basically a stream of run instructions you can use to trace the execution of the code without interrupting it), but unfortunately they're quite expensive and proprietery, and require extra connections to the systems that do support them, so they're not particularly commonly used. Most people just use some kind of home-grown logging/tracing system that tracks the particular state they're interested in, possibly logged into a ringbuffer which can be dumped when triggered by some event.
You ensure dev builds don't break stuff like that with realtime programming techniques. Dev tools exist and they're usually some combination of platform specific, expensive, buggy, and fragile.
printf and friends are fantastic when applicable. Sometimes the cost to even do an async print or even building in any mode except stripped release is impossible though, which usually leads to !fun!.
