in reply to Wat was the Architecture used in perlmonks

Just to clarify some of the other replies. PM is based on an older version of the Everything engine. It has been heavily hacked since then and represents its own fork of the Everything development track.

I dont know that anyone currently knows exactly how different the two engines are now, especially as the code PerlMonks uses is not open source. Personally ive always kinda wondered how difficult it would be to migrate some of the PM infrastructure to other Everything setups.

---
$world=~s/war/peace/g

  • Comment on Re: Wat was the Architecture used in perlmonks

Replies are listed 'Best First'.
Re^2: Wat was the Architecture used in perlmonks
by castaway (Parson) on May 24, 2005 at 14:01 UTC
    Right.. well, you could try, had you the time.. ,)

    Since I'm in the process of fiddling an Everything pre1.0 site into life, I maybe have a perspective on that.. I feel quite a bit of PM would not be too hard to port: create special tables, ie those for approval, re-create nodetypes, port the many superdocs, nodelets, containers and so on, create the whole XML business (displaytypes etc).. And then theres the renamed functions (eg isApproved is now inGroup)..

    Ok, so maybe not *hard* as such, but probably would involve a lot of poking around to find out which functions have been renamed, or moved, so it would not be a quick job, anyhow.. And then if one wanted to make use of new features, like nodemethods (though I've personally found them slightly more annoying, so far), which replace the maintainance nodes, among other things, it would be even longer..

    Well, maybe when someone has a month or 4 to spare..

    C.

Re^2: Wat was the Architecture used in perlmonks
by wazoox (Prior) on May 26, 2005 at 13:41 UTC
    BTW, why on earth isn't PM code open-source? Does it make sense?

      The Everything engine it is based on is open source in some form. However the heavily patched code that actually runs here is not. The reason generally is a matter of the PITA that it is to publish the material, combined with minor issues over site security. Hypothetically it would be nice to be able to completely open source the code, but realistically its difficult and we restrict access to the pmdev group just to ensure that those with code access are more or less responsible types who will tell us about problems they find and not somebody else.

      ---
      $world=~s/war/peace/g

        "Security through obscurity" is pure concentrated hi octane heavy-duty "cargo cult programming"... But I admit the "publishing is a PITA" argument : laziness is the first programmer's quality after all :)