Using Perl in the Real World (Cookbook)

I've sort of carped and blogged about the lack of well documented, well supported Perl interfaces to some of the newer services available to developers today. For example, interfacing Perl and Amazon web services, payment services, etc, etc.

I may not have done enough homework, but I think what might be lacking is a Perl cookbook for real world applications. I'm not talking about how to connect to a database using DBI or such trivial cookbook staple, but rather some meaty material that show me how to use Perl and connect to the Abbyy Fine Read OCR web service, Amazon S3/ECS/SQS/SNS interfaces, Stripe, etc.

I think a collaborative effort of all Perl programmers out there to share their experiences with doing something more than ETL or coding yet another database application in the form of a cookbook (maybe a wiki?) would be cool.

I see this as one step up from the traditional cookbooks that only show core programming idioms. I envision an expanded cookbook that delves into application specific areas that are more relevant to the kind of services available today. Perl developers need to see examples of using Perl to interface with services that enable mobile apps as well as the kind of RIAs that users are demanding today.

We don't need any more cookbooks that show me how to interface with Oracle. We need cookbooks that show how to store a note on Evernote, add records to a Mongo database, a file on Dropbox, process a payment with Stripe or spark up an EC2 server.

Anyone agree?

8 Comments

I think it's a good idea, but I don't think we need such a resource in a dead-wood form. For example, a blog site dedicated to such articles would be good, or even a category/section on blogs.perl.org.

I like this idea. I think it would work well as a wiki with an index of sites/services.

I've had a lot of frustration interfacing with online services. A lot of the modules on CPAN decay as APIs change and I have to try out several until I find one that still works. Often some debugging is required, sometimes even taking network traces. It would be great if there was a resource to go to see how other people have dealt with such situations.

It would also be much easier if more sites published a SPORE description of their API.

Step 1 is a starting list of the services we want to describe.
Step 2 is a wiki, one page per service (even just a TBD will do) and one main page with the list of pages.
Step 3 is volunteers.

Yes, definitely!

What about packaging documentation as CPAN distributions in a common namespace and hosted on github? It's not an ideal fit, but it would give easy versioning, bugfixing, example code and tests, use of Perldoc to view the documentation, etc.

Leave a comment

About rlauer

user-pic I blog about Perl.