I build https://t.co/ROVhnT9gvH
It’s a full replica of the https://t.co/YFzZDUXcHH world map, but with the added ability to zoom out fully and see all pixels at once.
@theo@yacineMTB Where as with graphql, or any kind of dynamic query language, the LLM can specify exactly which tables, joins and columns it needs to answer the users question very token efficiently, because (hopefully) no unnecessary data was being queried
@theo@yacineMTB What technology would you use to make LLMs be able to efficiently query data related to the users question?
If you would just expose the existing REST api a web app uses, the LLM might have to query multiple times and join „manually“ to answer the users question
@pioner_dev Also compared to react-native-nitro-sqlite, which uses a c++ unordered map/AnyMap and calls setProperty for each row object.
https://t.co/yZkHxdNbPw
@mrousavy have you thought about using a similar array buffer reader approach for nitro untyped objects?
@jarredsumner I think a phased rolled out (beta, stable) would probably be better, because there are very likely issues not covered by the test suite, but covered by real world workflows.
This way you keep your high reputation and will not get flooded by issues that say "rewrite was a mistake"
I build https://t.co/ROVhnT9gvH
It’s a full replica of the https://t.co/YFzZDUXcHH world map, but with the added ability to zoom out fully and see all pixels at once.
@rmoskovy@scriptjunkie1 Message content is never transmitted via notification payloads
Apple's server only receives a ping from the Signal server to the user's device
On the user's device, the Signal app downloads the encrypted message from the Signal server, decrypts it, and displays it as notification
I've built https://t.co/PQGFT8Kuos
Submit suspicious websites and emails to automatically scan them for phishing and report them to service providers to take them down.
You can also directly forward phishing emails to [email protected] as .eml file
https://t.co/EixOAkwjc2
Is there any service to which I can forward phishing emails and websites that will automatically contact law enforcement and report it to the service providers?
If not, I think I have a new weekend project
Why does it take this long to takedown a phishing site?
In the past I‘ve manually taken down phishing sites fairly fast, by getting the WhoIS information of the domain and sending a mail to the domain registrar.
That surely can be automated.
I've built https://t.co/PQGFT8Kuos
Submit suspicious websites and emails to automatically scan them for phishing and report them to service providers to take them down.
You can also directly forward phishing emails to [email protected] as .eml file
https://t.co/EixOAkwjc2
Is there any service to which I can forward phishing emails and websites that will automatically contact law enforcement and report it to the service providers?
If not, I think I have a new weekend project
@gautham_vijay_@mrousavy Awesome work 🚀
Do you use URLSessionUploadTask on iOS? This way it’s possible to upload in the background aswell
https://t.co/YTZwr6gBO5