1. yawaramin 1 day ago

   OK. My definition is that it should be able to add two integers together and give you a result somehow. So in SQL:

       select 1+1; -- Result: 2
   

   In HTML: not possible.

   That's the key difference.

2. yawaramin 2 days ago

   No, Apple can't suddenly start doing attestation in the future by default because that would instantly kill all the passkeys that have already been created on Apple devices without attestation. It would be as if a home security company went around and changed all the locks they had installed on their customers' front doors. It would be instant suicide as a trusted vendor.
3. yawaramin 2 days ago

   Not really. ORMs have defining characteristics that hand-rolled SQL with mapping code does not. Eg, something like `Users.all.where(age > 45)` create queries from classes and method calls, while hand-rolled SQL queries are...well..hand-written.
4. yawaramin 2 days ago

   Passwords are also simply phished, and many people have discovered those sharp edges by getting their accounts hacked.
5. yawaramin 2 days ago

   As I said earlier, this is functionally impossible because Apple devices don't offer device attestation data.
6. yawaramin 2 days ago

   Apple doesn't do attestation, so effectively this feature is dead in the water.
7. yawaramin 3 days ago

   Learning SQL basically launched my career as a professional SWENG. Once I knew SQL, I found ways to apply it in even non-technical jobs.
8. yawaramin 3 days ago

   What is your definition of 'programming language'?
9. yawaramin 3 days ago

   'The original microservices'
10. yawaramin 3 days ago

    For fun, I did the same for OCaml:

    Ex 1

        let say = "I love OCaml"
        let () = print_endline say
    
        (* Requires linking in the 'str' library *)
        let say = Str.replace_first (Str.regexp {|\(.*\)love\(.*\)|}) {|\1*love*\2|} say;;
        let () = print_endline (String.uppercase_ascii say)
    
        let () = ignore |> Seq.init 5 |> Seq.iter (fun () -> print_endline say)
    

    Ex 2

        module StringSet = Set.Make(String)
    
        let cities = StringSet.of_list [
          "London";
          "Oslo";
          "Paris";
          "Amsterdam";
          "Berlin";
        ]
        let visited = StringSet.of_list ["Berlin"; "Oslo"]
    
        (* Requires the 'fmt' library *)
        let string_set fmt v = Fmt.Dump.list Fmt.string fmt (StringSet.to_list v)
    
        let () =
          Format.printf "I still need to visit the following cities: %a\n"
            string_set
            (StringSet.diff cities visited)
    

    Ex 3

      module Greeter : sig
        type t
    
        val make : string -> t
        val salute : t -> unit
      end = struct
        type t = { name : string }
    
        let make name = { name = String.uppercase_ascii name }
        let salute t = Format.printf "Hello %s\n" t.name
      end
    
      let g = Greeter.make "world"
      let () = Greeter.salute g
    

    Obviously, OCaml is a much lower-level language than Ruby or Raku–notably, regex support is not as great, and we have to explicitly tell it how to print values of custom types. Still, I find its lack of syntax sugar makes it easy to read nearly any OCaml code I come across in the wild!
11. yawaramin 3 days ago

    See https://rakujourney.wordpress.com/2025/03/30/the-harc-stack/
12. yawaramin 3 days ago

    I am assuming that if the user selected a specific brightness mode, they want sites they visit to respect that theme. Call me crazy but this seems like common sense.
13. yawaramin 4 days ago

    One would think so, except we are constantly seeing people creating React apps that should have been a few simple HTML files and some JS.
14. yawaramin 4 days ago

    That's why you just have the HTML in the backend server codebase...which can also make sure it's cached properly with HTTP caching techniques like last modification time, ETag, and so on.
15. yawaramin 4 days ago

    The 'often loaded' static html won't help if the 'often loaded' JS bundle is re-deployed (eg, when the devs make a 'small' change) and the cache is invalidated. In that case all the users will be forced to reload the giant JS bundle over and over again, as opposed to just reloading a single resource on a single page like they would with the htmx approach.
16. yawaramin 4 days ago

    Which browsers are not doing this?!
17. yawaramin 4 days ago

    https://htmx.org/examples/active-search/
18. yawaramin 4 days ago

    See this subthread https://www.hackerneue.com/item?id=46315559
19. yawaramin 4 days ago

    It doesn't have history support out of the box. I consider that essential for hypermedia apps. When I'm loading fragments into my page I need to push the URL of the fragment I just loaded so the user can reliably come back to that specific resource later.
20. yawaramin 4 days ago

    > it genuinely does make the...users happy

    But you have no idea if it does that, you just have the word of the PO who's not actually building anything, they're at best just copying what others are doing (ie being derivative) or at worst just doing guesswork.

    How about offering an alternative: a UI/UX that takes the web as it is, a primarily document-based format with navigation and data entry? A lot of cool stuff can be built on top of that.

This user hasn’t submitted anything.