I wrote an editorial on the MCA certification referencing a blog by Simon Munro. He's not written a new blog entry referencing my editorial, which references his blog. Just to continue to circular reference, here's a new editorial referencing his blog referencing his, well you can fill in the blanks.
I think Simon has a good argument for the certification and since he's putting up the $10,000 for the program, he's putting his money where his mouth is. I applaud him and I hope he sees many times the return on his investment. And I'll certainly be one of the first to congratulate him when he achieves the cert.
I'm still not convinced that Microsoft needs to charge that much and if developing the program means it costs $50,000 or $100,000 to get top architects in a room, they should eat it. Maybe the community support from Microsoft isn't as strong as we've thought.
But I do agree that this isn't a certification for most of us. And that there are other certifications more suited to the rest of us technical folk. I think the latest generation of Microsoft certifications is a step in the right direction, but there's still a long way to go. The tests need to be more focused (think an SSIS or a replication test) and harder, requiring you to solve real world problems like the CCIEs must do. There's a lot of brainpower up in Redmond and they've done some amazing work on interfaces. I'd think it would be worth it for them to build a testing engine that could allow someone to actually solve a problem.
Or at least back up a database as a test exercise.
I tend to agree on the cost of the cert. I've been an infrastructure architect (according to MS' classification) now for the last 5 years. And while I'd love to pursue the MCA, the fact of the matter is the cost is more than my company would bear (we won't even get into the time commitment) and it's certainly not something I can afford out of pocket.
Since in our field circular references should be avoided (almost at any cost) maybe it raises an eyebrow as to our suitability for certification, writing certification books or anything beyond simply switching our machines on.
The Microsoft Certified Architect certification is very, very different from any of the other Microsoft certifications. It requires not only putting together a case study (with the help of a mentor), but also meeting a board of application/infrastructure architects (not all Microsoft, either) and not only presenting your case, but also fielding any and all questions related to architecture they may choose to ask. It's not exactly an easy certification, even if you can pony up the money.
Microsoft Certified Architect Program
It's admirable that Simon is willing to put his money where his mouth is. However, I think it would help a lot of people researching this certification if he could explain why he's doing it? And what he expects to get in return for his investment?
That would be a lot more persuasive (and impressive) than the "soft-marketing" on the official MCA website, MS' cost-recovery plan, or unsubstantiated claims that this certification is the "equivalent of an MBA degree".
I'm a .NET developer and a Sql Server developer, and I'm still back and forth on the costs/benefits of any and all Microsoft certification exams. Correct me if I'm wrong, but it seems that with so much of the material on these exams becoming obsolete so quickly that Microsoft has us on a hamster wheel that benefits them greatly.
Are we suckers or am I too cynical?
Yes, and yes.
Seriously, though. Certifications, whether we try to discount them, do give some level of credibility or people wouldn't pursue them. Whether it's an HR person looking at our resumes or a manager who pauses because you've got a cert that says you know databases, it does help, even if in a subconcious way. I know I qualify for certain government jobs just because I have a GSEC, a cert many people haven't heard of. Did I get the cert to qualify for certain government security jobs? No but it's nice that I have that as a side benefit.
Also, whenever you seriously undertake certification, you are forcing yourself to take a harder look at a discipline and your knowledge of that discipline. That's a great time to learn something new, to fill in gaps, and to correct mistakes in the knowledge. And while some of the material may seem obsolete, it isn't. There are still a lot of tried and true principles you learn getting an NT 4.0 MCSE that apply to managing Windows Server 2003 boxes today. Yes, a lot has changed (AD, firewalls, automatic updates, new built in accounts, etc., etc., etc.), but a lot has stayed the same (the way shares and NTFS permissions work, what to look for in the event log, had to configure networking using the GUI, etc.).
So while Microsoft surely is making something off all of these certifications, we get some benefits, some very tangible, out of the whole thing, too.
Thanks for your positive comments - especially from people respected within the SQL community. Although my blogs are littered with my personal motivation for walking the MCA road, perhaps a post which is a clearer reasoning checklist is in order - it should even crystallize some of my own thoughts on the matter from a first-person perspective. I'll add it to my 'List of things I want to blog about' and post it soon. It will at least come after at least one article that I have promised to Steve which I think I should get to since he spends so much time sourcing content for us - and I need to earn a bit of SQL street cred!
I don't know how Steve manages to churn out a daily editorial... I wouldn't get any other work done.
Thanks for the response bkelley.
Your point regarding credibility is sound.
Also, I understand that tried and true the principals do not become obsolete. However, generally speaking, changes in implementation seem to outweigh unchanging principles and unchanging implementation processes/tools ("seem" is the keyword).
With regard to the benefits of gaining knowledge, again, I don't disagree. But I believe that such benefits can also be realized via other, less expensive means. In addition, I would argue that such knowledge can be more tailored by avoiding MS certs. Not to diminish all who hold MS designations, but I've worked with uncertified developers whose performance was consistently superior to that of certified developers (and visa versa). Anecdotal and subjective, I know.
I'm convinced on the credibility front, but I remain on the fence on the knowledge-base front. Trying to keep an open-mind though. I want to believe, and I'm looking forward to Simon's post.