Everything is a Freaking DNS problem

There are a lot of pros and cons and the discussions can go forever.

I think they are needed just to be clear from begining on my point of view .

I value all the points expressed and i say, why not make it better than the other did it? One common point i see is that it should be linked to the involvement in the community. It can take the form of a gamification process on drupal.org. The overall goal should be increasing involvement and quality of our work.

Let's not forget that certifications have a strong value in recruiting process where the first layer of selection is not technical.

It's clear the positive experiences of certification are very few and far between. The level of Cisco certification we admire is very costly and time consuming, thus difficult to scale and distribute.

I hope you'll read about the previous discussions on this topic.
http://groups.drupal.org/node/172439

Please also read this page about definitions of terms for clarification:
http://groups.drupal.org/node/172419

And if you want to continue the discussion, please join:
http://groups.drupal.org/curriculum-and-training

I think many people would benefit from assessment: "Where am I and how can I develop my skills?" I get such emails directly from individuals who want to know how they can develop their skills. Right now, there is nothing available beyond me speaking to them individually and pointing them in the right direction. I can't scale that, heh.

So I do think assessment is needed in the community. In some cases to validate someone's skills/experience in ways that can't be done in the issue queue, and also to help form a plan for personal development. Guidance could direct people to specific training programs and core initiatives to get involved in.

And of course, as mentioned above - to help clients and partners HIRE and expand their teams. Please see some documentation about this:

http://drupal.org/node/51169 - How to hire a Drupal developer
http://drupal.org/node/1245640 - Skills of a successful back-end web developer
http://drupal.org/node/1245650 - Skills of a successful front-end web developer

(The above inventory of skills are transferable between different applications - this can help an employer identify someone who would be able to learn Drupal.)

Walking around Munich, I saw "certifications" in the shop windows of bakeries. To say certification is important in some countries would be putting it mildly. Acquia has clients and partners in western Europe, India, Asia that need certification to get contracts for jobs. They are literally begging for it.

I was very against it before, but I came around to the idea that it is necessary and needed in some cases.

Again, if you want to continue the discussion please join
http://groups.drupal.org/curriculum-and-training

I was involved with the team investigating sysadmin/SAGE certification around 2000. Like Kris, I could not find any interest for this topic.

To my mind, only really CISCO has managed to build a certification programme that anyone believes in. I never heard of anyone with a Red Hat cert, and MS is mainly a money spinner, etc. Certification is a business rather than a form of education. Badges of honour are not really looked upon seriously in our field because it is what we do that counts, not which club we belong to.

In 2003 I established the first Masters degree in system administration to encourgae deep learning. I think that kind of accomplishment is a better approach to setting industry standard, because i) it is NOT a business, and ii) college education is a depth and breadth exploration. Certification only captures some soundbites.

After starting the company CFEngine to support our "next generation" CFEngine 3, we did a similar trial. Again. not much interest in certification. Many people do not even want support -- there is a strong mentality of individualism in system administration. So far, that makes certification mainly an issue for vendors -- like us -- to standardize our own internal knowledge management. We want to certify our own materials for QA and not pass fake judgement on people.

So the obvious answer is, these projects should create a certification that you can earn by contributing to the project. At least for Drupal this would obviously work: If you can develop in PHP/Drupal, you can prove that by contributing. There could also be a certification for contributing themes, graphics, or even content on drupal.org.

Perhaps for other situation this doesn't work as well: It's not obvious that a acompetent Red Hat sysadmin can contribute anything meaningful to the Linux kernel. Otoh a certified developer can.

I've not believed in certifications since I met my first MCSE many, many years ago. That was followed by lots of other "certified" people. I still don't believe in them. In fact you probably know what mcse stands for right? ;-)

The way I deal with the issue of "Anyone can hire themselves out as a 'Drupal expert'" is checking their level of commits, patches, and experience on drupal.org. That's the first step to telling if someone is actually a Drupal expert that lives and breathes the community or a regular old php dev doing one-offs with a little structure.

Personally I think the way to go is an apprenticeship model. I'd much rather hear from someone that they helped work on Module X with established developer Y then the fact that they have certification from any vendor. This continues the fallacy that a degree = competent employee. A degree could an indicator of a potential good hire but references and experience are far more valuable.

Forget the financial implications for a moment - important, but not relevant to my first point.

The problem right now that I see is that Drupal risks gaining a bad reputation as being "unreliable" or "difficult", when the problems are really related to the people trying in install or maintain it.

Anyone can hire themselves out as a "Drupal expert", and assuming they can market themselves as such, can get work doing that - when they're really not qualified to do it. As a result, they end up creating a site that either doesn't perform well, isn't stable, or has any number of other problems. And, when the customer / client complains, those people then put the blame on Drupal rather than acknowledge their lack of skills.

I've worked with enough people who have had to go in and fix "bad" sites to believe this problem is only going to get worse unless *something* is done - acknowledging that certification is only one possible solution.

I do believe that this is an issue that needs to be addressed somehow - and it needs to be heavily promoted on drupal.org, such that a prospective client / customer has confidence in it.

I think there's also a substantial difference between the need for Drupal certification and certifications for things like MySQL or RedHat - and that relates to the target customer.

Generally speaking (knowing that like any generalization there are lots of exceptions), it's going to be larger companies or corporations that would / should be looking for certified MySQL or RedHat expertise. In those situations, the 'experts' are usually working with in-house staff who can end up properly evaluating those people's skills.

But, in this area (Baltimore/Washington), there are a _lot_ of small businesses that want to establish a presence on the web. Those companies are the ones that tend to see their money wasted on bad implementations, and end up believing it's Drupal that is the problem. The individual contractor moves on, leaving behind yet another business believing that Drupal isn't ready for prime time.

Great you walked away from the MySQL-exam.
I'm rather negative about certifications.

In the closed source world, certification is an extra income-stream for the software-company. They sell licenses to users and sell certifications to developers. Good for them.
Usually getting certified is insanely expensive, where you are required to follow a series of expensive proprietary courses.

In the past, I evaluated some certification-programs about project-management. But most questions of the test-exams were self-referential, and thus did not refer to actual knowledge.

In the open source world, you prove your skill-level in a very different way: by contributing to the project.

I'm an Oracle-certified MySQL for both, dev and dba. I did it only because I received majority of my experience doing private stuff, so it wasn't easy to document it otherwise.

Just a certification use case.

I have a similar experience. Got RHCE certified a decade ago and got surprisingly little leads from Red Hat or RHCE focused interest from the market.

My not-so-current RHCE is for 7 too. It will be a good laugh when people start calling how we already got certified for 7 when it's one day on the market. Maybe it will actually create some new business.

So friend or foe? It's a friend because the RHCE education was a really good experience. It's a foe because the market is oblivious what a RHCE/RHCA really is and what a reasonable rate is for an experienced RHCE and RHCA. Examples of demanding a RHCE for 35 euro/hour to do low-end sysadmin work are too common... Unfortunately the Enterprise demands some kind of certification so you can't really be without it. In the end I guess it's a friend for the wrong reasons.