Ross Attrill

• About: Keeping it simple with Perl

Recent Actions

• Commented on When Must You Test Your Code?
  There are several things that I like about this article. I like that the article acknowledges technical debt is sometimes appropriate and provides some suggestions on managing that. Your concluding paragraph says it all....
• Commented on By all means, try out $LANG, but also give Perl a real chance (a.k.a.: Good tools)
  A great article. There is a lot to be said for picking good tools and sticking with them. I think good new tools do come along from time to time (eg. Git was a step forward) but it is not...
• Commented on What about a Perl example?
  I agree with the observation and share the concern. Unfortunately I think it is a fashion thing and fixing this symptom is difficult....
• Commented on Chicago.PM - Dependency Injection (also: Beam::Wire)
  I am very interested in the combination of DI with Moose Roles. The presentation is very interesting - are you able to show any more detail on where you were going with the DAO type role as on slide 31?...
• Commented on Method::Signatures: here a tweak, there a bugfix
  Buddy - have their been any performance comparisons for MS vs. other modules providing similar if less comprehensive functionality eg. Method::Signatures::Simple....
• Commented on Method::Signatures: here a tweak, there a bugfix
  Method::Signatures must be the most important new module for Perl right now....
• Commented on DBIx::DataModel - Elegant Database Interaction with Perl
  Steven, Laurent has provided some performance benchmarks for raw DBI, DBIC and DBIDM at https://github.com/damil/compare-ORM In that study DBIDM is consistently faster than DBIC but raw DBI is the fastest. I have found DBIDM to be much simpler than DBIC...
• Posted DBIx::DataModel - Elegant Database Interaction with Perl to Ross Attrill

  DBIx::DataModel (DBIDM) is an Object Relational Mapper for Perl that is very elegant, simple and effective. It helps you to efficiently interact with your database.

  This article discusses three of the great features of DBIDM and …

• Commented on On the version number succeeding Perl 5
  Great work is already happening with perl5i: https://metacpan.org/module/perl5i Method Signatures or similar must be included for consideration in any future evolution of Perl 5: https://metacpan.org/module/Method::Signatures...
• Commented on Method::Signatures: where and when
  Congratulations Buddy on this release. Great to see Method::Signatures working on Windows. Thank you!...
• Commented on Plan (software) to live forever
  I agree with the value you place on documentation. We hear a lot about automated testing and refactoring but documentation and design sometimes seem to be valued less than they should be in the modern (agile) SDLC world....
• Commented on My Top MySQL Gripes
  Given the ownership of MySql I wouldn't invest much hope in seeing it improve....
• Commented on Why I'm considering dropping Perlmonks
  I feel that Perlmonks would be healthier place if anonymous posts were not allowed. That said, I keep away and would use stackoverflow if I had a Perl question....
• Commented on A Method::Signatures Retrospective
  Buddy - are you be able to comment on the plans for Windows support for Method::Signatures? The dependency on Devel::BeginLift is currently a problem on Win32. I note there is an issue to fix this: https://github.com/schwern/method-signatures/issues/39 It would be great...
• Commented on Comparing YUI - Dojo - Ext JS - jQuery
  Did you consider Apache Flex at all? Flex works very nicely with a Dancer based services....
• Commented on YAPC's Not Over!
  Hi JT, The site http://www.yapcna.org/ has never been available to me. I don't think it is being blocked by anything locally. I am outside of North America - could that be a cause? Error 137 (net::ERR_NAME_RESOLUTION_FAILED): Unknown error. Regards, Ross....
• Commented on Hidden unhelpful reviews on cpanratings.org
  I am happy for unhelpful reviews to be hidden. This isn't stopping constructively critical reviews - they may still be helpful. I just don't want to see the unconstructive, unhelpful comments that plague the net....
• Commented on Perl is more viable for web development than ever!
  I very much agree that the new simple frameworks such as Dancer and DBIx::Class are really giving Perl a renaissance. I am very impressed with Dancer....
• Commented on ORMs and Their Alternatives
  Great article. I appreciated the use of stored procs which are a valuable tool missing or ignored in most ORMs....

Comment Threads

• zaki commented on What about a Perl example?

  One thing that I have noticed is that the CPAN modules have the opportunity to become better documentation than the documentation from the API provider. APIs for Perl can include a readable manual rather than just a list of functions. Also, if you include complete usable examples for common tasks, you include them under an examples directory and MetaCPAN will show them.

• Steven Haryanto commented on By all means, try out $LANG, but also give Perl a real chance (a.k.a.: Good tools)

  Yup, I think my life (well, to be more specific, the way I store and manage code and data) has changed because of git. So thanks again, Linus et al.
  

• mascip commented on Method::Signatures: where and when

  Hi Buddy, thanks a lot for Method::Signatures.

  One naive (unrealistic?) idea/request:

  Could there be any way for a user to tell in a Moo(se) class: “don’t do any type check for any subroutines of this class” ?

  Maybe in the same way that Debuggit skips useless debuggit() function calls ? (“Performance Considerations” in Debuggit::Manual - implemented with B::Deparse)

  We could then develop/debug with type checks on, and not use type checking in production.

• mascip commented on Method::Signatures: where and when

  I should probably have written "in a package", rather than "in a Moo class".

• Buddy Burden commented on Method::Signatures: where and when

  > Could there be any way for a user to tell in a Moo(se) class: “don’t do any type check for any subroutines of this class” ?

  Sure. This is being discussed right now, in fact, as GitHub issue #73.