I implemented a receipt scanner to Google Sheet using Gemini Flash.

The fact that it is ”intelligent" it's fine for some things.

For example I created structured output schema that had a field "currency" with the 3 letter format (USD, EUR...). So I scanned a receipt from some shop in Jakarta and it filled that field with IDR (Indonesian Rupiah). It inferred that data because of the city name on the receipt.

Would it be better for my use case that it would have returned no data for the currency field? Don't think so.

Note: if needed maybe I could have changed the prompt to not infer the currency when not explicitly listed on the receipt.

> Would it be better for my use case that it would have returned no data for the currency field? Don't think so.

If there’s a decent chance it infers the wrong currency, potentially one where the value of each unit is a few units of scale larger or smaller than that of IDR, it might be better to not infer it.

I think most tools in this space do the "infer a bunch of data and show it to the user for confirmation", which lowers the pain of a miss here.

> Would it be better for my use case that it would have returned no data for the currency field?

Almost certainly yes.

Except in setups where you always check its work, and the effort from the 5% of the time you have to correct the currency is vastly outweighed due to effort saved from the other 95% of the time. Pretty common situation.

Even in those setups it's better to leave the currency field blank instead of hallucinating something.

If your observability stack works and you are fine with it, do you need to update it?

I understand updating some front facing service due to a vulnerability... But for a thing that it's internally accessible?

You can do no-downtime deploy of a web service with:

- Kamal

- Docker compose with Caddy (lb_try_duration to hold requests while the HTTP container restarts)

- Systemd using socket activation (same as Docker compose, it holds HTTP connections while the HTTP service restarts)

So you don't have to buy the whole pig and butcher it to eat bacon.

> - Systemd using socket activation (same as Docker compose, it holds HTTP connections while the HTTP service restarts)

Nit: it holds the TCP connections while the HTTP service restarts. Any HTTP-level stuff would need to be restarted by the client. But that’s true of every “zero downtime” system I’m aware of.

"There are privacy implications as the email transmission informs the mail service the applications the user is using and when they used them."

Not really, as I can enter any email on a service login page that uses magic links for auth. The owner of that email will receive the login link but that doesn't mean they tried to login on that system.

Not really indeed. You're right that false positive are possible with such a system, but false negatives are not. That means that you're leaking information about when a user didn't use a service, as well as partial information about when the did (which you could combine with other data to tell you something meaningful).

Forget about performance, specially in a 2d game like this one. Focus on making it fun (really-really fun is possible).

I have played games that didn't performed so well because they were so fun. Games are about fun, anything else (narrative, performance, sound...) is secondary.

If frames are dropping, it's likely not fun.

For technical questions the agreeableness is a problem when asking for evalation of some idea. The trick is asking the LLM to present pros and cons. Or if you want a harder review just ask it to poke holes in your idea.

Sometimes it still tries to bullshit you, but you are still the responsible driver so don't let the clanker drive unsupervised.

I did something similar but as a Chrome extension using Gemini 2.5 Flash (or Flash Lite) for summarizing.

On the page it shows an extra TLDR button near the like button.

You can change the prompt to modify how the summary looks and has an optional mode with links to specific timestamps.

Coincidentally yesterday I was reading the Cybernetics Wikipedia page and discovered the relation between it and Kubernetes.

I'm having a Baader-Meinhof effect moment right now :)

Don't you find liberating that any human, no matter how powerful they may be, how good or bad they are, cannot escape from death?

Maybe it sounds a little dark or edgy, but this thought gives me peace. Imagine what an immortal tyrant could do to humanity...

Yet he was surrounded by his family by the very end. Pretty much died under his conditions, unlike all the other lives he affected

There's apparently an old Japanese saying that goes "Asleep, one mat; awake, half a mat." It refers to the space on a mat that everyone, even the Emperor, occupies.

Straight from The Great Dictator.

"The hate of men will pass, and dictators die, and the power they took from the people will return to the people. And so long as men die, liberty will never perish."

Until you learn it's not individuals, but groups of them with ideas that persist for multiple generations.

“Behind this mask there is more than just flesh. Beneath this mask there is an idea... and ideas are bulletproof.”

Tomorrow is the 5th of November after all.

> Imagine what an immortal tyrant could do to humanity...

I'd imagine an immortal tyrant would do nothing to humanity since humanity would be insignificant to him.

The great equalizer

Just suggest to him to implement or supervise the creation of a system like that ON HIS RESPONSIBILITY. That is, if the system fails and loses company/client money he has to pay it from his own account.

Then tell us what how he sees that 5% error rate.