[consulting] Feedback on gold stars for Drupal Contrib modules

[Bill Fitzgerald]

A lower barrier to entry results in more contrib modules, and a low 
barrier and much code will (generally) result in more code that doesn't 
meet the "standard" -- and I put "standard" in quotations because the 
"standard" really hasn't been defined yet.

However, a lower barrier to entry results in multiple approaches to 
similar problems, and this can have the effect of pushing/forcing 
development in a positive way -- for a recent example, I'll hold up 
agentrickards aggregation api proposal --

One of the things that makes Drupal a great tool is that there is 
generally multiple ways to solve the same problem. One of the things 
that makes Drupal difficult to use is that there are generally multiple 
ways to solve the same problem.

As the current thread on the dev list about 3rd party code in contrib 
illustrates, we are all over the map with respect to what constitutes 
"best practice" -- Drewish's SoC project will go a long way toward 
starting us down this road, but there will always (I think, anyways) be 
a certain element of sorting contrib that is reputation-based. While 
this doesn't help new users get the best modules quickly, it does take 
steps to ensure that the diversity of opinion embodied in the variety of 
contrib modules will continue to flourish -- for better of for worse, 
the variety of contrib modules (and the varying quality of contrib 
modules) helps keep the project strong. And, how many developers look 
back on their first module as their best work?

In short, we need to take steps to ensure that the barrier to entry is 
low. This keeps new developers coming on board. Out of the new 
developers will come new approaches, and new talent, and that is a Good 
Thing. I'm not saying that Gold Star Cerification would necessarily 
reduce that, but it does have the potential.

Cheers,

Bill

Sami Khan wrote:
> Quoting Dries Buytaert <dries.buytaert at gmail.com>:
>
>   
>> Boris' suggestion certainly has merit but it seems to quantify the
>> developer -- and in specific, how well the developer's methodologies
>> match what is considered to be the current best practice.  The system
>> penalizes other (emerging or established) development methodologies
>> that produce equally good code.
>>
>> In other words, it is an interesting idea, but it might not be the
>> best system to rate the quality of the result (i.e. the final theme
>> or module). A specific methodology does not guarantee a good product,
>> and a good product doesn't imply a specific methodology.
>>
>> Nonetheless, I think this is something that needs to be communicated
>> on the project pages.  Not as a 'gold star' but in plain text:
>>     
>
> I meant to leave this as a comment on Boris' blog, but essentially
> putting more pressure on developers to meet certain expectation or
> rules I think will have an effect of increasing abandonment of
> projects and result in fewer contrib modules which might be available.
> It might produce a few good quality modules, but those would still be
> around regardless -- and the respective developers would as well. In
> that gold star modules and non-gold star modules are exclusive in
> terms of functionality, the net effect would be reducing the number of
> modules and functionality available to the general consumers of
> Drupal. In essence its the consumer who needs the proper information
> about the quality of the goods they're consuming, but no more then
> that... It's up to them to make a decision. Obviously, the experts
> would be able to get at the code regardless, but for non-expert what
> Drupal's fully capable of would be masked under such a system. I think
> Dries approach is spot on in terms of providing metrics and adding
> guidelines, leave the rest to developers and consumers to judge for
> themselves, for their commitments, etc.
>
> Regards,
> Sami
>
>
>
>
> _______________________________________________
> consulting mailing list
> consulting at drupal.org
> http://lists.drupal.org/mailman/listinfo/consulting
>
>   


-- 
Bill Fitzgerald
http://www.funnymonkey.com
Tools for Teachers
503.897.7160