Perl "Certification" - still Snake Oil
I think "certification" for most software is snake oil.
And, despite the participation of fellow Perl trainer Peter Scott (whom I have the highest respect for), and my primary publisher O'Reilly (whom I also have the highest respect for) in a new "Perl Certification" program, I still think this is snake oil.
Therefore, I will be discouraging individuals from taking such courses, and HR people and clueless managers from looking for such certifications, particularly demanding them to be considered for an application. I will continue to work hard with my clients and my fellow contractors to have actual track records be considered, not some test one has managed to pass and pay for.
Please, people. Don't feed the trolls.
print "Just another Perl hacker"; # the original!
Why? What if the "certification" was only secondary? I was looking at it as a way of improving my Perl skills (I would be a novice) in a semi-classroom environment for a reasonable amount of money. The cert would allow me to have my company pay for it is all. I am certainly not going to shout "Look at me! I am Perl certified!" from the rooftops.
"The cert would allow me to have my company pay for it is all."
That is exactly the behavior I'm arguing against. If your company is giving you training only if there's a "certificate" at the end, they're hopelessly confused.
There's plenty of Perl training out there (disclaimer: I do Perl training as well). There's no need for only some of it to result in a "certificate", and especially no need to balkanize the workforce into those who have taken a particular test vs those who haven't.
Yes, I know you do Perl training. I might be a novice but I know about Stonehenge. :-)
My company might be hopelessly confused but a cert is a "marker" of some sort of achievement even if we who program don't take a "Perl certification" seriously. If I take a college course and want my company to pay for it I need to get a "C" or above and I have to prove that "C" or above to them to get reimbursed. It is the same principle really.
The O'Reilly thing would let me do it at my pace and on my time and really for my reasons. If Stonehenge did the same thing I would do it but to get reimbursed the company is going to want "something" that shows I did it because I am not a programmer to them but I can use this as a CPE for a certification I already have in which they have a vested interest in having me continue.
Maybe I am an edge case though and don't see the whole market as you do.
If it's just a piece of paper you need, Stonehenge has worked in the past with the training departments of our clients to produce whatever sort of meaningless thingy at Kinko's that they seemed to need.
But what matters to me, as a trainer, is that my students can perform and think differently at the end of the class, to be more effective and useful. No piece of paper makes that better or worse.
It really depends on what the certificate is. I didn't think the University of Illinois was certifying anyone as a Perl programmer. They give you a certificate proving that you took and met the requirements of the class. It's the same same difference between being graduated from law school and passing the bar. One proved that you showed up, while the other says you're competent to do the work. Professional development programs mostly do the former.
Stonehenge provides certificates of attendance for students when they request it. All that says is that they took the class, even if they just surfed Facebook all day.
Is this an "O'Reilly certificate" or a certificate that any establishment that passes a certain criteria can produce? Is it being done with The Perl Foundation?
If it's a matter of "we want more business by having the only Perl Certificate(tm) in town", I can understand the major issue with it. It's unfair competition, IMO. If it's a "training companies can now set a bar of knowledge or ability recognition" thing, I can see a positive side to it.
The O'Reilly School of Technology offers these certificates through the University of Illinois. It's fully disclosed on the Perl Programming Certificate page.
Hi All,
Thanks for your comments. Just to be clear, the Perl Certificate OST is offering is a Certificate of Professional Development from the University of Illinois. This simply means that when you successfully complete all four courses, the University sends you a certificate.
Each course is considered a noncredit, continuing education course administered through the O'Reilly School of Technology and sponsored by the University of Illinois Office of Continuing Education (which is in turn sponsored by the U of I CS Department). Our courses are experiential, meaning you build a portfolio of real-world Perl projects as you learn, and you have one instructor ensuring that you master each concept along the way.
I want to clarify that this is *not* an industry-type, exam-based certification like Sun or Microsoft. There are open-ended quizzes along with the projects in each lesson, but no certification exam (or any type of multiple choice exam) that has to be passed.
Personally, I tend to agree with Randal on that subject.
Thanks,
Trish
Trish, thanks for that clarification. I admit, I have a short fuse about this subject, and didn't research this much further than "OMG, Perl Certification!". It's good to know that at least the people involved are not looking at it like that.
I'm still worried that industry will start preferring this one course over others though, since it is offered by O'Reilly, which I'm still very much opposed to (despite it being Peter Scott and O'Reilly).
But as long as we're all clear, and it's out in the open and googleable and linkable, I'm happy to move on to the next topic. :)
Since I'm tagged here, let me give my own take, with some overlap to brian and Trish's good words. That this course has the dreaded 'C' word attached to it is about all it has in common with what you abhor. It's more like a degree (the UofI affiliation shows the direction they're going in).
There are quizzes and homework at the end of each lesson, but they aren't marked by machine; they're evaluated by a human instructor who doesn't assign a grade but instead offers feedback and keeps working with the student until they understand the topic (unless they reach the very rare point of having to refer the student for a refund because they aren't ready for the course). I purposely assigned homework that mostly requires thinking and giving written answers as opposed to multiple choice; Google won't help much.
The goal is not to fail people on a curve but to get everyone who enrolls the knowledge and ability that they came looking for; we want people to win. If it has the certification stigma associated with it then I expect this series is more likely to partly improve the image associated with that word than anything else, because this is as good as it gets in asynchronous online training.
What any prospective employer does with the information about someone taking that series is independent of the value of the teaching; they can misuse that data or they can use it usefully to know what topics they can quiz the applicant on.
As long as the certification course is good and meets its objectives (as this seems to (not surprised here since ORA has always produced good learning tools)), then its main purpose is to prove one was exposed to the topic. There is not a defacto "guarantee" that the person sporting the certificate is creative, good to work with, or productive. My key point being: certification only "proves" one was exposed to a topic; it does not guarantee a good/great worker.
This is all mostly "obvious" but since or industry seems to be accepting certification as a panacea for proving a potential employee will be good, we in the industry still need to consider the usual "caveat emptor" when hiring someone with certification and use due diligence to ensure they'll meet our needs.
Given the feedback above I think we all agree with the slant you are taking on this Randal. :-)
Can I get a Facebook certificate if I stare at Facebook all day?