How was the Webmethods certification like

How easy it is to pass the certification? How do I review? Can anyone help?

Joan,
I’m also interested in this - you’re allowed three tries at the exam. I sat it one morning, around six months ago, in half the time, just to see what it was like, and missed out by 1 mark. I wish I’d spent more time on it …

 Besides working through the fundamentals of I.S. (package management) & Developer (Flow, etc.) you have to understand xml DTDs and Schemas, so practicing them by hand seems like a good idea. They also ask questions about forms of EDI/e-Commerce XML formats which I did not know very well - OAGIS (?), etc.. 

I'm not certain that there is much help in readying yourself for the exam. I'll give it another shot after my current wM project and a little more reading on the B2B landscape. 

Nick

Last year, I worked as an independent consultant. One of the main check points for consideration by many of the clients was the issue of certification. So, one Sunday after dinner, I sat down for 21 minutes and took the exam. In my opinion, the exam was easy because I had the practical experience to back me up. If you do not have “real” experience, then you will need to use good test taking skills.

I didn’t ace the entire exam. For one, I’m not a java programmer, but I can hack around. I missed a little bit there. On the other hand, I worked on a major project relating to XSDs, DTDs and XML files that allowed me to get all those parts right.

Overall, the guide that webMethods provides on their web site is pretty comprehensive but very broad. They ask some off the wall questions, based on “it’s all stuff you should need to know.”

My opinion is that the test does not cover their server environment but is based on philsophical grounds. it covers important material but not exactly relevant. If I had not worked on the big schema/xml project, I would have not passed the test.

I have a guy I used to work for who just read all the manuals and took the test. He never touched the software.

I would put the WM test somewhere between a candy-coated Microsoft Certification and an Oracle Certification Exam.

Hope this helps.

Ray

Hi,
If you go thru the documentation that webMethods provided for the developer, server and TN, with a little bit of java and xml knowledge, you should be able to pass.
Best of Luck!

I agree with Ray’s assessment that the test doesn’t really cover material pertinent to wM software (e.g. knowing what X12 is, which is one of the questions, is next to meaningless for indicating an understanding of wM software). On the other hand I don’t want the wM test to turn into what the Java certification exam is (e.g. “what statement below will compile”–isn’t that why we have a compiler?).

When looking for staff, wM certification is close to meaningless. But, some companies still want it so it can be worthwhile to have.

Of course that’s just my opinion.

To pass the exam: read the wM docs thoroughly; have a good understanding of XML, DTDs and schemas, have a good grasp of Java, work through all wM tutorials and samples.

I found the exam pretty easy after about 1 year of hands-on webmethods installations, upgrades, and developing. I had also read the pdf documentation a million times during the learning curve which probably helped.

The topics are broad but if you have a strong base you should be ok.

Will

I agree with Rob’s take on the test. It doesn’t cover enough specific information. Also, I worked with many MS certified types who proved to me that cert tests are nearly worthless as a tool to weed out “good” vs. “bad” developers. What happens if you are a poor test taker but a genius in all other respects?

Ray

As the person responsible for the certification test itself, let me say a couple of things. The purpose of the test is not to determine if you’ve been thru training or if you’ve memorized the documentation. It’s designed to measure if you can actually complete a webMethods integration.

This is why the “off the wall” question come into play. However, the question selection algorhythm ensures that the focus is where it should be: on the software itself.

The target audience is someone who has been thru training and has approximately 6 months hands-on experience. However, I have known people who passed with all hands-on and no formal training. I have also seen the opposite.

In the near future, we will be launching the webMethods 6 certification test. We will definitely consider your comments as we finalize the test.

Please forward any comments or suggestions on the test itself directly to me.

Thanx,
Theo

Theo:

Thanks for chiming in and providing the information about the goals of the certification exam. You certainly have a challenging task! (Side note: Did Matt Klug contribute some material to the current exams? Is he still a wM’er?)

The challenge as I see it is that there are two goals that tend to pull against each other:

  1. Measure knowledge of the specifics of webMethods tools.
  2. Measure the ability to complete integration projects.

Focusing too much on 1 tends to favor “book” knowledge. Focusing toom much on 2 tends to get away from the toolset (e.g. you end up with questions that seemingly have nothing to do with knowing Integration Server, or how to configure an adapter, etc.). Finding a balance between the two can be very difficult–but as you mention, the question selection algorithm leans toward 1, which seems right in my opinion.

“…designed to measure if you can actually complete a webMethods integration.” is a very elusive target. The only true measure is to have someone actually complete one. Multiple choice questions are a very difficult way to measure this capability.

Question to the group–if wM offered a controlled-environment (think college entrance exams), hands-on certification process, would you take it? What cost would you be willing to bear? Would varying levels of certification be of value? I would think if such a program were in place, certification would mean a bit more than it does now.

Perhaps the answers to some of these questions would be very enlightening. I’m particularly interested in hearing from managers responsible for webMethods [STRIKE]resources[/STRIKE] people–what’s of value to you?

I believe, as with all testing, the best kind of test contains flexible, subjective ‘word problems’, rather than the kind of testing that is ruining our public schools (but that is for another forum).

i.e. Here is an integration problem. How would you solve it?

There are some nice “integration patterns” defined in GEAR, so one part of the answer could be to define the pattern to use. A second part of the answer would be what webMethods components are used to best implement that integration pattern. There are multiple correct answers, and therefore, may require justification to get full credit.

Grading such a certification exam would be resource intensive, so maybe it would be “Architect Certification”.

Rob - IDK if Matt contributed any questions or not, but yes he’s still around as a Cleveland-based SE. In fact, I spoke to him last week.

Our current consensus is to try for a 70/30 weighting of “book learning” vs hands-on. Measuring the hands-on experience is tough with the current multiple choice format. We are looking at other options for the future. I am very interested how others answer your questions. Perhaps a separate thread would be appropriate?

In light of this discussion, I thought I’d chime in my thoughts.

Certification is only worth the value that is put upon it.

There is definitely a value in the webMethods Certification Exam. The exam is challenging enough to require study and/or project experience. Many people have failed the exam and many have passed.

The webMethods certification exams are not cake walks by any means. They don’t, however, require “trench work” to pass them.

This is not a bad thing, in my opinion. I even know of several non-technical Sales Force-types who are certified B2B Developers. The designation is important to help those salesmen establish credibility to their customers.

It is my plan, though, to offer an expert webMethods certification exam through the webMethods User Community. I have run the idea past a few key webMethods folks who are in support of the idea.

The wMUsers.com webMethods Certification exam will be difficult. I already drafted some sample questions and wrote the first phase of the exam software. My test subjects did not fare so well and that was a good thing.

The certification offered by the community will be a challenge and it will test for a comprehensive knowledge of the webMethods Platform. This includes theory and actual application. One already-drafted question requires an applicant to properly identify a file as server.bat versus server.sh.

Esoteric? Perhaps, but what better way to separate the Experts from the Knowledgeable?

I am expecting many people to fail the wMUsers.com webMethods Certification exam because of its difficulty. There is value in certifying Sales Force-types, but a salesmen will have no chance of passing this exam unless they have spent considerable time architecting, coding, and debugging on the webMethods Platform.

Most of the details are hazy at this point, but I expect to release the certification exam in the (United States) Summer, 2003.

I expect the wMUsers.com Certification Exam to be a supplemental certification and not a replacement for the various webMethods-offered certifications. The current exams are invaluable for testing Theoretical Application of the webMethods Platform and that is a core skill that all webMethods stakeholders should have – especially the certified experts.

This is great new Dan!

I passed the cert tests within days of finishing my first week of training. It helped that I had a pretty good understanding of XML type concepts going it, I guess.

On the other hand, there were entire sections that I knew absolutely nothing about, not even recognizing the terminology (I think they were all BI questions) I got enough of the other questions right to pass the exam…

My opinion - I shouldn’t have passed the exam when getting 0 right on a couple of sections. I think it would be possible to weed people like me out by making the test longer, and being harsher on people who do very poorly on whole sections. No amount of Java knowledge should be able to make up for zero knowledge of (for example) Business Integrator.

I’ve got my IS certification up on my cubical wall though, right next to the ES certification.

I like the idea of different degrees of certification. The web-based one has a value and should be retained. If nothing else, it’s ease and cheapness to take encourages people who might otherwise not have the time or strong belief in certificates in IT. Once complete they become more enthusiastic on getting more certification.

At that point though, I it would be great if there was another certification for more experienced people and ideally it would be practical. Ie. you are given a couple scenarios, you are given access to all manuals (ie like in real life) and you spend 1/2 - 1 day basically doing an integration project in simulated environment under monitored supervision. From that you would be graded. In addition you could plug in a “theoretical” part of exam so that immediately following the practical one you would then write an essay on a particular question. Not multiple choice, an actual open ended question that really challenges original thought and understanding of the integration solution space.

The negatives would be the cost in deliverying such certification since even marking would be manual and time consuming. But if someone asked be best case certification, that is what I choose. Even though I did the web certification and passed, I fundamentally don’t see enough true value when the test is more a memory exercise and a true exercise in problem solving and creativity with all tools at disposal as in real life.

“No amount of Java knowledge should be able to make up for zero knowledge of (for example) Business Integrator.”

This raises an interesting point. Last summer at a user group meeting, a wM SE asked how many people in the audience of 50 or so had looked at or were using BI. Not one hand went up though there were obviously wM IS skilled/experienced people in the room. Part of this can certainly be attributed to the newness of BI at the time. I wonder what the BI penetration is at this point?

A couple of questions come to mind: How much knowledge is required before one is considered an expert? Does one need to know every tool in the suite? Is BI considered “core”? What about Workflow? How about Mainframe Integration Server? If you’ve never used the File I/O adapter, can you be considered an expert?

Perhaps commonality is the driving factor. If so, then perhaps IDOCS/SAP BC knowledge is a requirement–there seems to be a LOT of work in that area judging from the number of posts in the forums.

Are there core questions that, if missed, would cause one to be failed regardless of how one performed on other questions? For example, if a person doesn’t know what pub/sub is, should that person fail?

(On a side note, given a choice between 2 people–one knows BI and is new to Java, the other is a Java veteran and new to BI–I’d take the Java veteran every time.)

Where to draw the line in the sand…

Good point, but there’s “not being an expert” and there’s “I’ve never even heard of it”

A certifiable (I like it) person should have at least heard of BI, or FileIO, or [insert core product here]

If BI isn’t core to being an IS developer, then it shouldn’t be on the certification exam at all (IMHO) If a product is tested on the cert exam, posessing the certification should indicate at least some knowledge of that product.

A certification should represent a baseline of knowledge. It’s pointless to have a certification that doesn’t specify what it means beyond “I passed the certification test.”

A potential employer should see a certification on a resume, and be able to find out what skill set that means the person has - otherwise, what’s the point?

You’re right though, where do you draw the line? I think it’s too lenient right now - almost as if webMethods wants to be able to market with “there are thousands of certified developers out there” or something (there goes my conspiracy theory hat again )

I saw a message in webmethods certification page.

* What's on the webMethods 6 Exam? coming soon 

When is it going to be alive?

Also, Dan, is there a “tutorial” for your certification exam? Is there a “COST” factor associated with it?

Thanks
MKR

Hi, MKR.

I am moving this thread to General webMethods Topics and will address your questions in a new thread.

I am responding now…

Just to share my thoughts about WM exam -

3 years Java & XML Projects + Just 3 months of webMethods experience + 2 days of Documentation study = 63/75 (Passed the exam in less than 30 mins) :slight_smile:

Sunny