What do you think it's holding Odin back?

i had thought about this and honestly, it didn’t seem like too bad an idea lol

only problem i could think of was lsp usage

See how the discussion has already slid from “sqlite” to “sqlite and postgres”…

1 Like

Slip down that slope, baby! Odin has sdl and glfw so why not sqlite and postgres? Still, I shouldn’t emphasize postgres. Sqlite is a much better candidate.

3 Likes

We started slipping after the Box2D package. We suddenly got so many physics libraries like Chipmunk2D, Bullet, Godot Physics, Jolt Physics… THIS IS INSANE!??

1 Like

From a language point of view I think Odin is fine. I’m not that big of a fan of the #+feature stuff and would rather the language were more opinionated and just outright banned literal maps, using statements and whatnot. I also worry a bit about how the odin language only specifying base and then letting core being “compiler specific” will pan out.

But other than that, I think the language is quite good and I don’t have much to complain about with the language and its tooling. What I think is “holding it back” is lack of a library ecosystem and of open source programs written in it.

I think having more good open source projects that are actively maintained and useful to people and would also convince more people to learn it so that they could contribute to said projects is very valuable. But a lot of people using odin just write one off projects or closed source projects.

2 Likes

It’s not something inside Odin that’s holding it back.
Every new language has the same problem.
Without support from big companies, or some luck, it’s hard to grow.

Success doesn’t always come from quality or maturity

For hobby work, new programming language and joy of programming sound great.

For boring systems — and the people in charge — they don’t mean much :joy:

Regarding SQLite, I think that the concern is misplaced.

People might request more databases but SQLite is the only right answer to a small, on device or in memory, relational database. It has virtually 100% market share in that segment and you have to have a REALLY good reason to not use SQLite if SQLite is an option for your usecase.

PostgreSQL in comparison has competition at least on paper. Saying “If I add PostgreSQL, I have to add MySQL and somebody will ask for MariaDB specific features and then somebody is gonna ask for MS SQL” is a 100% legitimate concern and there’s room to argue and all the stuff you want to avoid.

If you put your foot down and do SQLite and nothing else, I don’t think anybody can legit argue against that except for “ is also SQL” but that’s the weakest possible argument you could have.

I do think the SQLite API is very stable though so also I don’t necessarily see a good reason to not just generate your own bindings. Like, in the database sphere I think SQLite is a no-brainer but should you invest your time into databases in the first place when there’s other stuff to do?

2 Likes

Hi all. Brand new here, so I can’t comment on language-specifics, but I can say two things:

  1. the Odin homepage could be more clear about what Odin is like. (See my comment in the marketing thread.)

  2. Activity in Discord is not as visible (or searchable) as activity in the forum. Prospective users will look at forum activity as a proxy for how active the user community is.