Peter Fern

A better option than just hard-coding the new (current) value for itemType might be to query the itemType from videoversiontype for id 40400. Doing so will allow this to work...

What's your use-case for the additional text fields? I'm hesitant to add labels for the sake of it, unless they're useful for querying metrics.

So AIUI the only viable choice currently for working unixd is to leave `anonymous` enabled (including anon LDAP binds, etc), allowing user enumeration by anyone who can get a socket...

On this repository, with `Task version: v3.39.2 (h1:Zt7KXHmMNq5xWZ1ihphDb+n2zYLCo4BdRe09AnMMIgA=)`: https://github.com/delaneyj/datastar Running `task -w templ`: ```yaml templ: env: TEMPL_EXPERIMENT: rawgo generates: - "**/*_templ.go" sources: - "**/*.templ" cmds: - templ generate . ```...

Okay, weirdly, if the above `templ` task is modified to additionally include just files in the root directory like so: ``` templ: env: TEMPL_EXPERIMENT: rawgo generates: - "backends/**/*_templ.go" - "./*_templ.go"...

In concrete terms, how would implementing the API via JMAP improve the ability to develop/maintain the API vs evolving/documenting the REST API (via e.g. OpenAPI/gRPC/whatever)? Are these gains significant enough...

I can see the alignment for internal workflow/maint, and that may be enough to justify taking this route. For external consumers of the API though, it's likely that they will...

Alternatively, `hx-on` should natively support some way to subscribe to events from an alternative target, for parity with `hx-target`. Considering the value is raw JS, this probably needs to go...

> Personal opinion: hx-on is fine for simple usecases where you have a simple call, like a form.reset after a request or stuff like that. > I feel like as...

@lowinger42 we're lowercasing attribute names only here, not values, so the charset will be ASCII.