State of Drupal Certifications in 2010

Transparency

Submitted by Benjamin Melançon on Mon, 2010-09-27 11:01.

Why isn't the certified to rock algorithm public? (If it is, please point me to it.)

Anything claiming to be a Drupal certification that isn't transparent about it makes me a little uncomfortable. That said, i completely agree with the approach that certification should not be about a test, but about publicly verifiable contributions.

The Drupal Guilds idea also proposes qualifying standards, and there too i urge contributions be the gold standard– which don't have to be code, as Dani and others point out there, and most contrib skills like documenting and organizing and more are worthy jobs skills. I'd be interested in your thoughts on this wider array of rocking skills, some of which could also potentially be measured from d.o data.

secrecy is a feature, not a bug

Submitted by Greg on Mon, 2010-09-27 14:07.

See the last few paragraphs on this discussion of problems with certiifcation programs

At this point it's a feature, not a bug, that the algorithm is private.

I appreciate that having private things in an Open Source project with an open community feels weird, but I think once people understand learn why we are doing that they will come to understand and appreciate it.

low barriers to entry ++

Submitted by Benjamin Melançon on Mon, 2010-09-27 11:05.

Low barries to entry and community orientation key also, i meant to say. Thank you. Again, a lot of discussion around this at the Drupal Guilds concept.

Certified to Rock

Submitted by John Q on Mon, 2010-09-27 12:23.

I had completely forgotten about CTR. Thanks for posting about it. I've taken Microsoft and Novell certifications in the past but many people can study and do well on a test. Real-world contributions in a given field are a very good indicator of competency.

Any certification not certified by the Drupal community isn't worth much. CTR seems like the most semi-authorized one at this time.

I was very pleased to learn that my CTR score is a 4. I'm going to try to get it to a 5 as soon as possible.

John Q

no wonder

Submitted by chx on Mon, 2010-09-27 18:08.

a traditional certificate would need to be retaken every year and completely worthless after two. This is a breakneck pace, almost impossible to keep up both on the company running the program and people apply for the cert.

bygones?

Submitted by Heather on Tue, 2010-09-28 04:02.

whimper You're not gonna let me get away with that one, are you? Jason had made the joke the night before after a few beers. I probably had impaired judgment when I RTd that. Alas, wasn't very funny. I haven't made a joke on Twitter since that morning.

Good to have a round-up and pull interested parties out of the woodwork.

Acquia has no official position or plan for certification currently. We are reviewing options. The knowledge we need does not lie within the Drupal community. It's time to look outside - to other open source and also proprietary software to see how they managed certification.

Some more notes at: http://heather.drupalgardens.com/content/bsd-certification-group-working...

BTW, You didn't mention Lullabot gives "certificate of attendance" or similar.

no intention to dwell on it

Submitted by Greg on Tue, 2010-09-28 08:21.

Sorry if it feels like I'm harping on that tweet - I have no intention to dwell on your retweet, it's just the most public facing and obvious thing about an Acquia Certification in a few years.

Good point about Lullabots certificate. I'll update the post...

Rock on

Submitted by Moshe Weitzman on Tue, 2010-09-28 06:36.

I have no insider info at all, but I bet CTR is pre-empting the efforts of commercial training firms, including Acquia. Which was really the point, I think. So many kudos to GVS for conceiving and especially for executing so well.

definitely hope to shape the market, if not pre-empt it

Submitted by Greg on Tue, 2010-09-28 09:53.

We definitely do hope to help shape the market if not preempt it. I feel concerned that a traditional certification would be damaging to Drupal, or at least not helpful. Our goal is to create an environment for certifications (and guilds, or whatever else is valuable) that is beneficial to the recipients and the community alike.

Leave No Drupal Dev Behind

Submitted by Sean Larkin on Tue, 2010-09-28 12:40.

I share concerns of many on this thread about Drupal certifications. This prompted a blog post for me:

http://thinkshout.com/blog/2010/09/sean/leave-no-drupal-dev-behind

I don't want to divert the conversation here, so I'll repost below:

The increased discussion of Drupal engineering certifications concerns me. I question the motivations behind this movement, the efficacy of such certification programs to precisely identify talent within our developer community, and its potentially skewing effects on vendor selection in the the Drupal service provider industry.

I am not a Drupal rockstar engineer. I'm a humble 4 on the Certified to Rock scale. I've asked way more questions on the forums than I've answered. But I've personally logged probably 3 to 3.5 thousand hours of paid Drupal development and I've project managed another 3 thousand hours of such work. I consider myself a pretty solid Drupal technologist who's made 70-80 clients happy over the last 5 years.

What makes me a passable Drupal technologist doesn't have much to do with code. It has to do with being an English major and being somewhat talented at communication. It has to do with having lived in the developing world and understanding what true crisis is (it is not a website launching a month late). It has to do with having spent a lot of years working in the vertical industry (in my case nonprofits) that I serve.

Now, my motivation in writing all this isn't to self-aggrandize (though it may sound that way), or to feel sorry for myself that I'm only a "CTR 4". There are way better developers, project managers and firms out there (which is awesome, because I learn from them). My point is that as seen with testing programs in U.S. public school, a written test of an engineer's expertise with Drupal provides very limited information when making a hiring decision - while simultaneously bringing with it the possibility of much misinformation and many negative implications for our community.

I think that the following points need to be considered:

1. What are the motivations of the companies that seek to guide (or control) these certification processes?

The people driving this movement are good. They love the Drupal community. But they need to pay themselves. Managing the certification process has financial rewards. At this point in the process, you don't have to have the best certification process, you have to be the first to gain traction in the market. Do we want the quality of our fellow Drupal developers to be determined by the marketing that backs such-and-such certification program? Do we want talented Drupal developers to spent time (and money) on certifications rather than contributing organically to our community?

Certified to Rock is a great response to this. It provides a fun, silly, low-key way for Drupal devs to bump chests at the bar after a day at a DrupalCon. There's no financial reward (that I'm aware of) for those that manage the algorithm. Hopefully if it becomes too serious, it will go away - because it will stop being fun for us geeks.

2. It's not that hard to vet a developer's technical chops.

Engineering certifications provide a single point of data in the hiring process. But that's the one point that's easiest to ascertain in the hiring process. With just two or three well-asked questions and 5 minutes of research on Google, I can get a good-enough sense of a Drupal developer to know whether or not I should interview her for a job. It's the interview process that's the art - and that process is affected by organizational culture and other factors that can't be reduced to a written test.

3. We are flooded with ourselves.

We are largely a self-taught community. Part of why Drupal is so popular is that it's an easy framework to pickup and learn without familiarity with other development tools. If we focus on hiring developers strictly based upon Drupal experience, rather than plain-old software engineering experience (or God help us, rather than with folks that have been working in Rails or other equally-to-more robust development frameworks), we will miss learning opportunities and we'll become the stereotypical Royal Family - and a little too close to close to kissing cousins.

Case in point, when OpenSourcery (where I project managed for 2 years) hired Jonathan Hedstrom, he had next to no previous Drupal experience. He'd made relatively few upstream, public commits on any open source projects. But he'd been doing incredible engineering work for years on an application that was intended to be open sourced in the future. He knew source control inside and out, he knew his development tools, he knew test-driven development. He is brilliant, and he instantly became a Drupal star and our senior Drupal engineer. If we'd chosen another applicant with a higher Drupal "score", the entire Drupal community would have missed out on his talent, let alone OpenSourcery.

4. Certifications will become a pissing contest between development firms.

Sorry for being crude, but as a vendor this is my primary concern - and not because I'd win or lose that match, but because our clients will suffer as a consequence. It's already overwhelming for clients, particularly in the nonprofit space, to understand what makes a reliable, effective Drupal shop. It's not about badges on a vendor's website. It's in large part about engineering talent. But it's mostly about making clients happy. It's about translating technology into tools. It's about exceeding expectations. It's about communicating clearly. It's about integrity. It can't be bought with certifications or badges - but it can be hidden in the strobe effect of such flashing marquees.

Even if such certifications are meant to be a single point of data, clients can miss that - particularly in an RFP process. If we want to do right by our clients we need more engagement with them to help them understand the subtleties of working with technology providers - not given them our "scores", saying "Good luck, you ought to pick me."

. . . .

In conclusion, our Drupal community is unique because of it's grassroots nature - and the personal interactions that we share as geeks, engineers, artists, activists, and so forth. As Drupal becomes a main-stream enterprise platform, that culture is poised to change. In some ways, it has to change. But we need to be conscious of the unintended effects of the "meta companies" and "certifications" and "service level agreements" (all potentially Cathedrals) that are springing up. We need to honor our roots in the Bazaar.

You spoke my mind

Submitted by Jacob Singh on Wed, 2010-09-29 07:17.

I love this comment. Brilliantly written; that English degree was worth all those thousands in debt :)

Here's a few rambling afterthoughts:

The kissing cousins bit is especially important. We need to grow this community, not weed out the pretenders at this point. Training is 100x more important at the moment than certification IMO. Certification will inspire people to pay for training though, which is a good thing.

I agree that CTR (I think I'm a 7) should be public if it wants to be taken seriously. I don't see why being secretive is a feature in this case. I'm guessing it is secret because greggles, et all doesn't want people complaining about the recipe :) I can understand that, but if it is anything more than useful for pissing matches, I'd need to know what's going in the pot.

Maybe a more effective (albeit labor intensive) certification is to have people give detailed patch reviews or code reviews of various difficulties on d.o. That would truly demonstrate someone's ability to read Drupal code and contribute back.

followup

Submitted by Greg on Wed, 2010-09-29 20:05.

I posted a followup over on sean's blog that might interest you.

I'm not sure why you say it has to be public. No other certification program shows the whole world their test. And there's no need to guess at why it's secret, I've stated it multiple times including a few days ago in this thread.

NobleProg

Submitted by Greg H on Wed, 2010-09-29 07:01.

I actually used to do a bit of training work for them - I know the owner. The whole self-styled Drupal certification thing was a terrible idea, and they've now gone off to focus on management training, etc. anyway... I think they lost interest in Drupal as a training business, even though their site is Drupal still.

Their certificates were never any real proof of competence, or at least not while I was doing work for them. It was just confirmation you sat there for 2 days and listened! ;-)

That said, they did train some high-profile clients in Drupal in the UK, Poland and further afield (they have quite a smart remote training set-up with a dedicated VNC server in Germany).

Anyway - just thought that might fill in a few holes.

NobleProg Certification

Submitted by Bernard S on Wed, 2011-06-29 08:10.

I think Bazaar model doesn't really work well with certification (I do not know a single instance when community can easily design it), as well there is little awareness of the cost related to certification development and upgrade.

THERE IS NO MONEY IN CERTIFICATION compare to other activities and it is sad that instead of support from the community (which on the other hand works wonderfully with the Drupal development itself) we had more criticism than any kind of support.

Developing certification programmes requires a lot of money (usually more than it is possible to get from exams fees) and few companies are interested in sponsoring the effort. I suppose that is the reason why there is still no certification available.

As to the remarks from Greg H, they are not really up to date (a couple of years old at least) and have very little to do with current reality.

My thoughts

Submitted by Brian Gilbert on Wed, 2010-09-29 23:27.

While I can understand some people wanting the algorithm public, I like that it isn't.

I think that some people would end up focusing their efforts specifically on trying to raise their CTR rating if it was public.

With the algorithm unpublished people are genuinely focused on contributing in any way they can, which is better for Drupal.

Personally as a current 4 it gives me motivation to keep contributing and elevate my score, which I believe benefits both Drupal via my contributions and myself with what I learn in the process.

Guilds vs. Cathedrals vs Diebold vs. Fun

Submitted by Sean Larkin on Thu, 2010-09-30 00:49.

Just to be clear, I think that there are many benefits to this algorithm remaining private. What I worry about (well, one of the issues that I worry about) is the conflict of interest between the keepers of said blackbox and their own paid consulting work. Playing ref and goalkeeper can get tricky.

I do think that CRT is a much better solution to paid cert programs. As pointed out above, it keeps folks focused on contributing to the community (albeit contributions potentially skewed by thinking contrib modules are worth more than contrib themes, etc. - and points potentially bought by just committing contractors' code...), and cuts out the financial barriers to entry that the paid programs create (the same ones that turned me from Microsoft platforms to open source in the first place...).

I am more interested though, in the Guild concept that Aaron Winborn posted about yesterday:

http://aaronwinborn.com/blogs/aaron/transcript-drupal-guilds-presentation

Regardless, Greggles et al, I am more than impressed with your overall transparency in CRT. Thank you for being so open to debate, standing your ground, and doing all that you can to make our community stronger and more sustainable.

Cheers,

Sean

Coming in waaay late, but...

Submitted by Tom Geller on Fri, 2011-01-14 13:21.

Like Lullabot, lynda.com provides printable certificates to those who complete their Drupal video courses.