in reply to Re: Inside-out Objects: Public Discussion in New York City (POSTPONED)
in thread Inside-out Objects: Public Discussion in New York City

And indeed it is rescheduled for our January meeting: Tuesday, January 17, 6:15-8:15 pm. NYPC User Group Office Suite. 481 8 Ave (Ramada New Yorker Hotel building), btw West 34 & 35 Sts, Manhattan. Room 550.
  • Comment on Re^2: Inside-out Objects: Public Discussion in New York City (RE-SCHEDULED)

Replies are listed 'Best First'.
Re^3: Inside-out Objects: Public Discussion in New York City (SLIDES)
by xdg (Monsignor) on Jan 18, 2006 at 15:40 UTC

    I've posted the slides from the talk. (PDF format)

    -xdg

    Code written by xdg and posted on PerlMonks is public domain. It is provided as is with no warranties, express or implied, of any kind. Posted code may not have been tested. Use of posted code is at your own risk.

[reply]

      Some comments based on the slides:

      • Page 8 section 1 should mention that locked hashes provide a good solution to this problem. But are restricted to post Perl 5.8
      • The point on page 16 about refaddr having a cost to calculate is IMO a bit of a red-herring. Returning the refaddr of a SV doesn't really involve any calculation, simply the creation of a new SV and the copying of the relevent address. I would guess that this is roughly equivelent to dereferencing a ref to a scalar in terms of processing. Hashes are going to be slower however because the conversion will be something like void * => newSvIV => newSvPV (actually I think its a SvUpgrade or whatever :-). But this isnt refaddr's fault. For instance it wouldnt be a too difficult to put together an XS routine that did all of this without the intermediate steps. At some point I bet it would be competitive with the scalar index dereferencing.

      Otherwise id say your slides are great and are well worth reviewing for a good example of how to do a good presentation on a subject. Great work++.

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

[reply]

In Section Perl News