I don’t get it, to be honest. Maybe I missed something. There are three main features listed:

Routing: The standard library does routing in net/http. Since Go 1.22 it even provides some syntax sugar to make things concise and easy.

This example:

 if second == "routing" && third == "" && c.Method == "GET" { /* ... */ }

Is written in the standard library like this (assuming I understood it correctly):

http.Handle("GET /routing", /* ... */)

SQL / migrations:

SQL is covered by database/sql in the standard library, though of course you have to bring your specific DB driver. I’m not a fan of ORMs, so not going to comment on how well that is done. Migrations seem explicitly tied to the ORM component, so same thing.

WASM:

WASM is supported through the standard library’s syscall/js package. I didn’t get what Feedback does here, the FAQ links only to a few small files with some helpers for calling basic JS functions.

Lastly, HTML templating is implemented in the standard library’s html/template package and it seems like that’s the same templater used here in the examples.

I don’t want to be overly negative, but I feel like I have to, because some comments here are overly positive and I find it uncanny.

To anyone considering using this project, don’t. It makes tons of bad choices, breaks anything standard about Go, and is overall completely non-idiomatic:

• the routing alone is absolutely terrible, making conditions on each part of the route makes zero sense, and this explanation HAS to be some kind of prank. Just use Gin, chi, or net/http.

• about sql support (?):

We support sqlite and postgres. And we make heavy use of: map[string]any This means you do not have to make a struct for each table.

Throwing out type safety is NOT a good thing. At this point just use something else than Go. It might be quick for toy apps, but it’ll fall apart as soon as you need any kind of business logic. Just use sqlx, sqlc, scanny, or simply database/sql.

• I didn’t go into details on migrations, but having a custom format not based on SQL seems like a recipe for disaster at this point. Just use go-migrate.

Regarding the codebase itself, it’s not great Go code. It’s all over the place and un-architectured, ie. probably unmaintainable by anyone other than you today (no one in 3 months). As a seasoned Go developer, I have no idea where to start from a Quick Look. Why are there libs making use of AI? For generating images apparently? What does it have to do with the lib? Why are so many stdlib helpers wrapped in unsafe helpers? Why are there python scripts? What’s this?

I apologize for being harsh but you have to realize this is absolutely not anywhere close to production ready for anyone. Maybe some kind of all in one framework like Rails has its place in Go, but please make it work with Go idioms, not with idioms from somewhere else. And you might want to focus on why you’re doing it.

Unsolicited feedback;

• what’s the name of this framework?
• what are its strengths? Limitations?
• how does it compare to existing frameworks?
• what’s the roadmap?

Took me a while to realize the name is “Feedback”. I thought you were asking for feedback. Given the URLs, I assumed it was called frame. Are there any other names you like? I see some markdown links but can’t tell if that’s a style choice. Weird formatting & typos hurt the vibe, just a bit. There’s an error in the JS console.

The overall aim is unclear. FAQ says “go on rails” but no mention of ActiveRecord or even controllers or anything. “No npm” but it’s in the simple example. Seems like everything’s written in Go, which is great, except the Python scripts(?). So it’s a tailwind/gotemplate-based framework on Postgres with manual routing, SQL helpers, & something for migrations. Is that right?

I like how the style reminds me of Cap’n Proto, but I want paragraphs of rationale. Especially with code examples, like the nodejs homepage or rails homepage or the zig homepage .

Is the router limited to “second” and “third”? What about “fourth” and so on? Seems like an odd way to do things; I’m really confused.

My go-to mental exercise when evaluating a tool is as follows:

Consider a web application that has been in production for some time and allows users to submit links and comments. Now, a new feature is required where users can submit text and others can add comments to it. This changes the comment structure from having a direct foreign key relationship with a link to a polymorphic relationship that supports both features. How would you implement or solve the following steps?

1. Add both “target_type” and “target_id” columns to the comments database table
2. For each row in the comments table set the “target_id” to be the same value as it has for “link_id”
3. For each row in the comments table set the “target_type” to be “link”
4. Drop the “link_id” column and foreign key index

I did not dive deep into it, so just a quick note: the series of if statements in the example on the front page can be written more concise and more to the point with a switch. Only one expression will be true. Also, there are predefined constants for the HTTP Methods:

switch {
  case second == "" && third == "" && c.Method == http.MethodGet:
    handleProject(c)
  case "second" == "new" && third == "" && c.Method == http.MethodGet:
    handleProjectNew(c):
  ...
}

congrats on the launch!

i really like the minimalistic approach, for instance that migrations are additive only. also the combo server-side templates + tailwind + daisyui works so well for shipping fast.

i’ve been building something similar myself over at https://www.plainweb.dev but decided to embrace htmx instead of client-side scripting. i wonder how the wasm story looks like in feedback? from the website it kinda feels like “there you go: it can run go in the browser, figure out how to build an app with it”. maybe worth looking into htmx so most of the things can be done in go?

i’ve been looking into templ + htmx in the past and it looked like a nice way to do things in go. but ultimately decided to settle for typescript and jsx because of the flexible type system.

regarding communication:

i get the appeal for using “x on rails” but i wonder whether this truly highlights the benefits of feedback. when i opened the landing page my immediate thoughts were “minimalism” and “simplicity” maybe also “constraining by design”. more than “feature packed”, “battle-tested” or “rich ecosystem”, which is how i would describe rails.

i mention rails, django and laravel briefly in the plainweb.dev docs as well. haha i guess it’s hard to avoid comparing yourself to the big ones! good luck with feedback, looks really cool.

Extremely non obvious UI/UX browsing the posts included link; unclear about what it does, how to use it or what makes it better/different than standard library or other relevant web based go projects.

I would Suggest having site have a prominent link to the repo on GitHub after updating the github repo README with well written overview and brief description of main features and a how to use it / install & quick start guide (if needed, get an idea from this as an easy to understand project README example if unsure how to start https://github.com/bigskysoftware/_hyperscript/blob/master/README.md )

wow, -1 off-topic this couldn’t be more on topic to go! love the lobsters community.

https://i.imgur.com/isKZJfm.png

don’t those count for anything not to get downvoted for a go project I’ve pour my hear and soul into for 5+ years?