Most active commenters
  • lr0(4)
  • hyperpape(3)

←back to thread

330 points todsacerdoti | 14 comments | | HN request time: 0.392s | source | bottom
Show context
the__alchemist ◴[] No.46235329[source]
If the community followed the author's guidance, we would have names like "Generic LLM wrapper 690" ("GLW690" if following the early programming language conventions.) or "Github clone with a different ideology 11"
replies(2): >>46235362 #>>46235446 #
1. lr0 ◴[] No.46235362[source]
Not at all. You don't name by category, you can name by function or approach. PostgreSQL isn't "Generic SQL Database 47" it's the successor to Ingres (Post-Ingres-SQL). If your "LLM wrapper" does nothing distinctive worth naming, maybe don't publish it. But if it specifically handles streaming, call it something like "llm-stream-client." If it focuses on prompt templating, "prompt-template-engine." The name encodes the actual value proposition.

I actually stated this on the post, but let me reiterate, I think that naming things in somehow fun way is totally okay as long as it stays relevant to what the tool actually does (you can have this achieved by play wording suffixes (Mongo"DB", Open"SSL", Ma"git" are good examples, all are better than elephant, dog, and beaver).

replies(2): >>46235525 #>>46235528 #
2. hyperpape ◴[] No.46235525[source]
> PostgreSQL isn't "Generic SQL Database 47" it's the successor to Ingres (Post-Ingres-SQL).

Indeed. This helps me know that I'm using a database more modern than Ingres. I chose not to use Oracle or SQL Server because they might have predated Ingres.

Just one question: what's Ingres, and why do I care about it? Of course, I don't, which makes Postgres no more useful of a name than "fluffnutz" or "hooxup". That said, over time, I've come to like the name Postgres.

replies(2): >>46235817 #>>46235864 #
3. gipp ◴[] No.46235528[source]
Sure, but how many LLM streaming clients are out there?

Namespacing, sure. But is "We use gh:someguy/openai/llm-streaming-client to talk to the backend" (x50 similarly cumbersome names in any architecture discussion) really better than "We use Pegasus as our LLM streaming client"?

replies(1): >>46235826 #
4. lr0 ◴[] No.46235817[source]
You don't need to know what Ingres is. "PostgreSQL" still tells you it's SQL-related, which is infinitely more than "fluffnutz" tells you. And once you learn it's a database, the name reinforces that knowledge forever. Good luck remembering what "fluffnutz" does in 6 months.
replies(1): >>46236657 #
5. lr0 ◴[] No.46235826[source]
Nobody says "gh:someguy/openai/llm-streaming-client" in conversation. You say "the streaming client" or "llm-stream" the same way you'd say "Pegasus." But when someone new joins or you're reading code, "llm-stream" is self-documenting. "Pegasus" requires looking it up every single time until you memorize an arbitrary mapping.
replies(3): >>46237532 #>>46238721 #>>46239034 #
6. indymike ◴[] No.46235864[source]
Sometimes names have great value at the beginning of the project. In this case it explains exactly what the project is and will be... That said, marketing decisions like naming a product often don't age well.
7. hyperpape ◴[] No.46236657{3}[source]
That's a really nice mnemonic. I wish I lived in an alternate universe where Postgres was called PostgreSQL so that it was easier to remember. Perhaps if we start using that, it will take over, like how everyone calls the Go project Golang.
replies(2): >>46237683 #>>46240405 #
8. debazel ◴[] No.46237532{3}[source]
This sounds awful, now you'll be reading some documentation or comment about llm-stream where they didn't mention the full namespace, so you have no idea which of the 50 different llm-stream tools they're talking about, and on top of that you can't even search for it online.
9. lr0 ◴[] No.46237683{4}[source]
https://www.postgresql.org

https://en.wikipedia.org/wiki/PostgreSQL

replies(1): >>46238974 #
10. wredcoll ◴[] No.46238721{3}[source]
I promise you, names are not self documenting. Not in any meaningful way.

This is one of those classic examples where things you've already learned are "obvious and intuitive" and new things are "opaque and indistinct".

We can go back and forth with specific examples all day: cat, ls, grep, etc are all famously inscrutable, power shell tried to name everything with a self-documenting name and the results are impossible to memorize. "llm-stream" tells me absolutely nothing without context and if it had context, pegasus would be equally understandable.

11. hyperpape ◴[] No.46238974{5}[source]
I know.

My point is that almost everyone refers to it as Postgres, because they do not actually value the descriptiveness of "PostgreSQL".

replies(1): >>46240971 #
12. Joker_vD ◴[] No.46239034{3}[source]
> You say "the streaming client"

"Which one?! There are seven popular projects with this exact name on GitHub that have >100K stars; which particular one do you use?"

13. necovek ◴[] No.46240405{4}[source]
When Google introduced the Go language, it was impossible to google for any content related to it. So community quickly pivoted to always saying golang ;)

(At least that's how I remember it as I was "why name a language like that when you know it won't be searchable")

14. halper ◴[] No.46240971{6}[source]
I always thought it was because it is more obvious how to pronounce "postgres" than "PostgreSQL".