Loading ...
Sorry, an error occurred while loading the content.

Re: [json] New JSON superset -- RSON

Expand Messages
  • Gregg Irwin
    Hi Patrick, P I am developing a superset of JSON called RSON (Readable Serial P Object Notation), at http://code.google.com/p/rson/ I know there was a pyson
    Message 1 of 4 , Mar 16, 2010
    • 0 Attachment
      Hi Patrick,

      P> I am developing a superset of JSON called RSON (Readable Serial
      P> Object Notation), at http://code.google.com/p/rson/

      I know there was a pyson project, but if it's dead, maybe you could
      use the name. I think PySON is a much better fit for your notation.
      It's not that your notation isn't readable--it's very much like REBOL
      in many ways--but the indent option gives it a very Pythonic feel.

      I agree on YAML. It seemed great at first glance, but I found it was
      more complex than I cared for once I went to write a parser for it.
      The goals are different for YAML, so it's not a blanket criticism.

      --Gregg
    • Patrick Maupin
      ... I agree that it came out that way. And I do code in Python, but frankly, I started using YAML, and then changed it until I got something I like (which
      Message 2 of 4 , Mar 16, 2010
      • 0 Attachment
        On Tue, Mar 16, 2010 at 2:46 AM, Gregg Irwin <gregg.irwin@...> wrote:
        > I know there was a pyson project, but if it's dead, maybe you could
        > use the name. I think PySON is a much better fit for your notation.
        > It's not that your notation isn't readable--it's very much like REBOL
        > in many ways--but the indent option gives it a very Pythonic feel.

        I agree that it came out that way. And I do code in Python, but
        frankly, I started using YAML, and then changed it until I got
        something I like (which might be because I code in Python, but who
        knows?). While I feel the syntax does mesh well with Python, I would
        prefer not to give the impression that it is restricted to Python. It
        would be easy to create a parser for this in almost any language.
        Besides, I would like for RSON to "set fire to configuration files all
        over the planet." (Although helping out in the goal of World
        Domination would be good too..)

        > I agree on YAML. It seemed great at first glance, but I found it was
        > more complex than I cared for once I went to write a parser for it.
        > The goals are different for YAML, so it's not a blanket criticism.

        Yes, a YAML parser is very much not a thing you can knock out in an
        afternoon, and while YAML is very powerful, explaining its more
        powerful features to non-technical people can be quite a challenge.

        Regards,
        Pat
      • Tatu Saloranta
        ... Yes. YAML is a good candidate to explain concept of Too Much of Everything (or, American Ice Cream syndrome :-)). It is unfortunately quite common to
        Message 3 of 4 , Mar 16, 2010
        • 0 Attachment
          On Tue, Mar 16, 2010 at 8:21 AM, Patrick Maupin <pmaupin@...> wrote:
          > On Tue, Mar 16, 2010 at 2:46 AM, Gregg Irwin <gregg.irwin@...> wrote:
          >> I know there was a pyson project, but if it's dead, maybe you could
          >> use the name. I think PySON is a much better fit for your notation.
          >> It's not that your notation isn't readable--it's very much like REBOL
          >> in many ways--but the indent option gives it a very Pythonic feel.
          >
          > I agree that it came out that way.  And I do code in Python, but
          > frankly, I started using YAML, and then changed it until I got
          > something I like (which might be because I code in Python, but who
          > knows?).  While I feel the syntax does mesh well with Python, I would
          > prefer not to give the impression that it is restricted to Python.  It
          > would be easy to create a parser for this in almost any language.
          > Besides, I would like for RSON to "set fire to configuration files all
          > over the planet."  (Although helping out in the goal of World
          > Domination would be good too..)
          >
          >> I agree on YAML. It seemed great at first glance, but I found it was
          >> more complex than I cared for once I went to write a parser for it.
          >> The goals are different for YAML, so it's not a blanket criticism.
          >
          > Yes, a YAML parser is very much not a thing you can knock out in an
          > afternoon, and while YAML is very powerful, explaining its more
          > powerful features to non-technical people can be quite a challenge.

          Yes. YAML is a good candidate to explain concept of "Too Much of
          Everything" (or, "American Ice Cream syndrome" :-)).

          It is unfortunately quite common to get carried away by desire add
          just 'one more feature', instead of carefully considering how much is
          enough, what is really needed, and where is 80/20 point.
          Not everything has to be done at the very lowest level, syntax of data
          notation; oftentimes optimal place is somewhat higher in processing
          stack. It's just necessary to have basic hooks so other layers can
          tackle other things, such as reference resolution, inclusions, type
          information handling.
          And at some point increasing theoretical expressive power comes at
          expense when users can't figure out how to use the thing. ;-/

          For what it's worth, I think JSON is slightly underpowered, personally
          (lacking comments is an absolute killer, but there are couple of other
          small but important omissions), but it is quite close to optimal. So I
          understand why defining supersets has some appeal, especially for
          configuration definitions.

          -+ Tatu +-
        Your message has been successfully submitted and would be delivered to recipients shortly.