[consulting] Coder Chops-Test

[Bram Moreinis]

I hesitate to continue this on-list, but as I composed my email to Brian 
(and cc'ing Sami) I thought I'd take one more step.

 > What type of solution did you have in mind? Patching the module 
directly (ie, submitting to the issue queue), or working from
 > an external module with hook_mail_alter() (we have our own module 
that needs to change this behaviour)?

I was thinking of patching...I have hacked my own version of 
authorcontact (slightly different purpose) but they are sufficiently 
similar that I could apply the patch to my version once I saw how it was 
done.

Now that I think about it, though, it might make more sense (and be more 
community friendly) to turn my hack into an add-on to author contact (so 
as to continue to benefit from their updates).  The maintainers last 
updated on 1/25/2010 so there seems to be a commitment on their part 
that should be respected.

 > What did you have in mind for the testing framework?

I'd found a PHP function 
(http://www.php.net/manual/en/function.htmlspecialchars-decode.php) but 
couldn't apply it because Authorcontact used Drupal API and required an 
API remedy.  Sami found the Drupal API function 
(http://api.drupal.org/api/function/check_plain) - thank you, Sami!
Since I've always hired coders I never hit the wall of "where PHP ends 
and Drupal API begins" before.

So one way to show skill is to see the structure of the first Google 
search (did they search site: api.drupal.org like Sami, or 
site:drupal.org as I did?),

How to see this? I was thinking "over the shoulder" (in person or a 
screen-record).  What would that show which is generalizable?  If we 
broke it into phases, there's:

    * What discovery questions did the coder ask? (e.g. your question - 
patch or external?) [tests ethos / system knowledge]
    * What did the coder already know without searching? [tests personal 
experience, not best practice, but that is important]
    * How did the coder format your search? [tests parity between 
personal structured knowledge base and drupal structured knowledge base]
    * Once the coder found a result, how did s/he apply it? [tests 
hacker vs. coder]
    * Once the coder found a solution, how did s/he document it in the 
code? [tests best practice]

-Bram