Modern Perl

DBIx::Class::Migration and the CPAN ecosystem

One of the things I've found most gratifying while working on DBIx::Class::Migration is how in using the CPAN ecosystem I was able to engage many authors to solve bugs / issues I uncovered in shared code.  For example, I noticed a edge case in the more recent versions of YAML that got solved rapidly.  I also uncovered some issues with DBIx::Class::Schema::Loader that the author was willing to take on based on my test case.  And in DBIx::Class::DeploymentHandler I worked with the author through several patch sets which solved bugs and improved the functionality.

In doing so the entire ecosystem was improved, not just the one project I am working on.  Improvements to DBIx::Class::DeploymentHandler will benefit anyone using that code.  The fixes to YAML are significant since that module is broadly uses.  And DBIC::Schema::Loader is a core part of the DBIC ecosystem.

Sometimes when you go to work on a project it might feel simplier to say, "I'm going to roll it all myself." Or to just copy paste stuff.  In some ways that can be faster.  It can sometimes be a slower process to work with many authors to convince them to accept your patches.  But it is always better for the community overall.  I believe that this contract amongst CPAN authors is part of what is going to allow Perl to continue to grow and continue to be a top notch solution for application development. 

 

MooseX-Types-Paramerizable Updated on CPAN

MooseX-Types-Paramerizable

This is a small patch by contributor (chocolateboy++) to remove some unneeded dependencies and clean up the pod a bit.  If it's working for you I can't consider this a required update but it shouldn't hurt anything either.

 

 

Catalyst-TraitFor-Model-DBIC-Schema-QueryLog-AdoptPlack Updated for Trial Catalyst

The most recently released Trial version of Catalyst fixes the way the Plack %ENV is exposed to your application.  As a result, I have updated Catalyst-TraitFor-Model-DBIC-Schema-QueryLog-AdoptPlack to work with the newer version as well as maintain backward compatibility.

If you are using Catalyst with Catalyst::Engine::PSGI, it is HIGHLY recommended you update to a newer version of Catalyst that bakes PSGI support right in.  However if you are currently stuck, please let me know if this update to the querylog debug panel busts anything for you.

Great Picture of Catalyst Dispatch Flow

Was on IRC today and someone posted:

Which is a pretty handy guide to Catalyst dispatch flow.  ++ to whoever did it!

 

Wikified Links to the DBIx::Class::Tutorial

Several people have told me they are playing with DBIx::Class::Migration from the github repo, but have all run into the same questions.  Rather than keep repeating my answers, until I can get that code on CPAN here's a link to the wiki that has links to the internal tutorial, which you might not see due to the way github is organized.  You'll need to navigate with the brower back button for now, but this is just temporary until I can work out the patches to DBIx::Class::DeploymentHandler with the author, and move this to CPAN.  Until then I hope the tutorial is useful!

Also, a link to the FAQ, which has answers to the most common error messages I am hearing about.

 

New Version of Test::DBIx::Class

Just a quick announcement, there's a new version of Test::DBIx::Class out there on cpan.  This is a bit late to be published, just needed a bit of time to fully review some pull requests (lecstor++) and I've had my mind on DBIx::Class::Migration quite a bit last few weeks.

I've been really happy to see that the last few revisions of this code have really been pushed along by great community support.  It is truly gratifying to me to know that code I've opensourced is not only useful to other, but is useful enough that other people blog about it and contribute.  This is the part of my career I enjoy the most.

How SOPA and PIPA could affect my (and your) OpenSource work

I'm blogging to say I oppose the SOPA and related PIPA acts, or anything at all like it.

I write a lot of free software, nearly all on my own time.   I give it away with the hope it helps others.  I do this because I feel that I have personally benefited from open source / free software and as a programmer that has made a lifetime career using open source I feel some sense of obligation to do what I can to help others who were in position my 15 years ago.

I wish that people working in content creation would have a more humble sense of their position in history.  No story today is unique, everyone has built upon the creative might of the ages.  Just as I have benefited from the monumental work of great programmers before me, so do the Hollywood storytellers of now benefit from from the creative genius of thousands of years of human culture.

I agreed that one has the right, if one chooses, to be paid a reasonable sum for reasonable work.  I make a good living writing the 'special sauce' for companies on top of solid open source components.  We live is a free market economy and I begrudge no one the right to make an honest living.  But the right to defend oneselves does not extend to preemptively blocking the wellsprings of human knowledge.  It seems to me that there are those in the content and human knowlege industry that would put 99 innocent people against the wall in order to shoot one guilty man.  They have an inflated sense of their own importance and an inappropriately inflated sense of entitlement.  SOPA and PIPA reflect that overblown ego.

These are the same people that would extend copyright forever.  These are the same people that have blackmailed thousands with blanket lawsuits.

For the short term, join me in opposing SOPA and PIPA. For the long term we need to kill the beast at its roots, by reducing our consumption of mass popular media products.  Personally I don't even own a TV and haven't bought a CD (or online digital media project) or dvd for years.  Think about it.

Thanks for you attention!

DBIx::Class::Migration Preview

If you're using the DBIx::Class ORM and building websites in Perl using Catalyst or similar modern frameworks, you might be interested in my take on packaging up DBIx::Class::DeploymentHandler and DBIx::Class::Fixtures into a neat set of classes and scripts designed to give you a standard workflow for database migrations.

The goal of the project is to help you get started fast using migrations and to help take away some of the painful setup effort it takes to get started with an DBIC project.  Over time I am hoping to build a set of standard practices that are community tested and well documented around these tools.  Lets give some of the other popular daabase tools for migrating and deploying something to worry about!

Currently this is a Github only project since its pending review and integration of several patches to DBIx::Class::DeploymentHandler, but if you are curious have a sneak peak and feedback is very welcomed!

Catalyst Advent Day Six

Just a quick plug for the Catalyst Day Six Advent article I recently wrote.  If you are interested in some of my thoughts on how awesome it is to have Plack as a core Catalyst technology, give it a read!

Things on the Minds of IT College Students

Last week I was interviewing a group of cooperative students who are seeking job placement for the Spring 2012 semester.  All in all they were a pretty solid group, although I should probably blog a bit regarding some advice I have for them :)  What I want to write about now is a few observations I have regarding what I heard the students say about IT, Perl and their overall interests.

Although this blog was inspired by my recent experiences at the University of Waterloo, I have also interviewed students and Stevan's University and NYU Polytech, so these ad hoc observations are a bit of a summary of several interview sites.

First of all, I always ask students what got them into programming and what programming languages are they learning from a young age as well as in College.  Many students tell me they started programming early in highschool or even earlier.  This jives with my personal experience (I stared programming in 7th grade or so, basically as earlier as it was possible given the technology of the late 1970's)  So it seems IT people still are a pretty early selecting group, many people play with a computer at a young age and just fall in love with programming.

Early programming languages tend to be Basic and related.  I was surprised to not hear Logo; I guess that has fallen out of favor?

In terms of what they are learning in college, I here C/C++ and Python a lot.  Many students tell me they used PHP in highschool but seemed to prefer Python once they got into college.  So it looks to me like the Python communities dedication and focus on being a learning language of choice has really paid off.  I am told lots of teachers are using Python and recommending it. 

Things students like about Python is its syntax and the depth of available libraries.

Regarding opensource / freesoftware, most students indicated a pragmatic preference.  They liked the fact that the tools are free (as in beer) but didn't usually express this preference in philosophical terms.  This makes me wonder if the open/free software movement has either moved into a sort of post political phase, where open/free software is basically a given, or if instead the quality of free (beer) tools and open(ish) documentation has to a large degree co-opted the philosophical underpinnings of the movement?  For example, I heard a deep interest in developing for platforms that are far from Free (such as the Apple iDevice ecosystem.)  To some degree this is enabled by the relatively inexpensive cost of entry (lots of free tutorials, tools are free or very easy for students to steal.)  Partly this has to do with what kind of jobs are the students thinking about, I would imagine.

Quite a few students told me they participated in a number of statewide / national programming contests while in highschool.

In terms of the Perl community, we are pretty lucky that a percentage of these students end up learning Perl one way or another, given they don't get it in highschool or college.  From what I can tell through copious interviewing for Shutterstock (as well as many other companies) people come to Perl typically through a job or because of a particular project they are interested in.  This experience left me wondering what, if anything, we could usefully do to reach people at a younger age (at least here in the USA / North America area?) and if putting effort into such an outreach is a good use of our limited advertising / community building time?