samanator

Excel formula to postgres SQL compiler. Reminiscent of Salesforce formulas.

Demo uses postgres compiled for WASM so demo runs on an actual postgres db.

https://skamensky.github.io/postgres-formula-compiler/

Yummy

Interesting. The same thing happened to me. Was offered a refund (graciously, as I had forgotten to cancel the subscription). And after thanking them and agreeing to the refund, was promptly ignored!

Very strange behavior honestly.

Cool to see you here!

We discussed this very thing on supabase

https://github.com/orgs/supabase/discussions/18061#discussio...

https://archive.is/BrssU

The two top comments since the November 7th updatee have thousands of upvotes because of affected users. Comments:

ThSAWistheLAW WC (11/7/24) "I like this app but it is not functioning correctly at the moment. If your video that you're viewing must be interrupted by a message notification or if you simply turn your phone right side up to read the messages or comments the video will stop playing and it will not start no matter how many times you hit the play button. Other than that it's a fine app."

---

michael. jacqui (11/8/24) "New update is awful. It no longer saves where you left off on a video. If you swipe down and leave the video going to search for a new one it automatically paused the video. When you go back to play the video it cannot play unless you select a new video and then have to reselect the video you want to watch to get it to play. The app is horrible and extremely frustrating."

I thought it was only me, but when I went to update the app I saw that the most recent reviews all describe serious UI bugs and have thousands of likes.

Good luck! Given your track record, I'm sure you'll do great things with the funds .

Thanks, apparently the article was taken down...

I just tested o1-preview on the "How many r's are in strawberry?" question. It answers correctly!

Yes it can be done with rules, inheritance, triggers, and updatable views.

See my answer on stack overflow

https://stackoverflow.com/a/78694335/4188138

However, in my experience, it's more easily maintainable to use normal deletes, have a trigger to move it to a schemaless table (jsonb column type that contains all of the data at the time of delete), and have manual restore logic via a good UX.

You can't use cancellation or heavenly bodies:

"He is far away, or in some strange location, and he cannot see anything we can see. For instance, we cannot say, “Look at Ursa major; now see how those stars are arranged. What we mean by ‘right’ is …” We are only allowed to telephone him."

I wonder how many petabytes of leaked memory there is in the world at any given time from open chrome tabs

The server got the HN hug of death

Mind share isn't necessarily an issue when you build a platform based on standard tools (e.g JavaScript, graphql, etc.).

In my experience those limitations are not just nuisances,they are expensive. They require a lot of custom work and and less reliable since you're leaving what's well known and supported. For example, many companies hit the "100 active concurrently scheduled jobs" limit very quickly. Then they need their own scheduler, or deal with the consequences of stale data.

Those limitations are not just nuisances, they change real business workflows.

Yes, there are definitely good business justifications for choosing Salesforce, but I definitely think there is real business value to be found in improved CRM systems.

Salesforce is slow (the lwc frontend I mean), and has legacy (aka governer) limits. E.g. you can't have synchronous code run for more than 10 seconds, you can't update more than 200 records in a standard transaction, the query language has only very basic joins, and much more.

If you can build a crm without these limitations, why not get your 80% without the limits?

Very cool and promising.

Currently the biggest blocker, even for simple use cases that don't have to add automation (there is no clear way to add customized automation), is the lack of permissions. Everyone has the same access to all the data!

User permissions: https://github.com/twentyhq/twenty/discussions/209

As mentioned by Andreas, it's because SerenityOS does not depend on third party libraries and part of the new ethos of Ladybird is to decrease "not invented here" syndrome.

https://news.ycombinator.com/item?id=40561408

Yay! Newlines in f-strings!

python, but if other people would like to use it, I'd rewrite it in go for the sake of easy distribution.

I haven't published anything. If you're interested I could clean it up and send it to you. Let me know. My email is in my profile.

I've written a locally running app that is a lightweight version of this. You can download all your emails, it loads it into sqlite, and you can browse and sort by aggregate email counts (how many emails each sender sent). Once you select a sender, you can archive all of the emails it sent you.

I'd have loved to have made it a full fleshed app like this. Kudos!

Apparently a misattribution.

According to this

https://klangable.com/blog/quantity-has-a-quality-all-its-ow...

It was not Stalin who said that

Where is this from? I've been looking for a term for this...

Same thing is happening to me

The last graph shows decreasing sentiment over time for all metrics. I wonder if HN has become more negative over time…

I remember rowy from a while ago. It said then that other databases (aside from Firestore) are coming soon. Any news on that? I'm looking for a Postgres frontend.

This update also seems to have patched all working DAN "jailbreaks".

None of the prompts at https://github.com/0xk1h0/ChatGPT_DAN/ or https://www.jailbreakchat.com/ work anymore with gpt4. Some are still working with gpt3.5

Yup, I've been on the waitlist since May

The paper's implementation is to essentially append memory text as part of the prompt.

Why don't they use a storage/retrieval system that doesn't consume context window tokens? E.g. Storage could be to automatically categorize data with tags at insertion time (i.e. upon user prompt) and retrieval could be a query that filters using a tag guessed by the LLM (before responding to the user).

With a few initial rules like hard coded tag names/styles, my intuition is that this would produce great results.