CLD Grading shifting to emphasize functionality

[Fisel]

Beginning November 1, 2019, CLD grading will shift from the current 40-point scale to a 100-point scale, with a point distribution of:

• Functionality 50%
• Style 30%
• Documentation 20%

To allow time for candidates to adapt and prepare, from October 1, 2019 to November 1 we will grade all CLDs using both the old and the new scale/distributions, and candidates will receive the higher of the two grades. After November 1 only the new 100-point scale will be used.

Thank you,

[crossrulz]

I assume it is still a 70% to pass?  But I still see the need for only about half of the functionality points to pass.  I am not arguing for or against the change.  Just and observation that it is not really that much of a change.

[Fisel]

Yes, still 70% to pass. Thank you for your feedback, we're hoping for discussion.

This is just a tweak. The CLD serves the LabVIEW community well, so we're not aiming for major changes at this time. We still want to emphasize Style and Documentation enough to encourage candidates to develop good habits while preparing. Just want to nudge Functionality higher so fewer exams are carried to "near pass" by extra-functional aspects.

[JÞB]

---

@Fisel wrote:

Yes, still 70% to pass. Thank you for your feedback, we're hoping for discussion.

This is just a tweak. The CLD serves the LabVIEW community well, so we're not aiming for major changes at this time. We still want to emphasize Style and Documentation enough to encourage candidates to develop good habits while preparing. Just want to nudge Functionality higher so fewer exams are carried to "near pass" by extra-functional aspects.

---

OK, I'll bite into the topic. by answering the hypothetical question "What do I want from a new CLD sitting in my office?"

Frankly, If their code actually WORKS I would expect that to be a bonus! I expect they will write bugs (I'll train them not to do that eventually.)  Is it clean? Is it documented? Does it have a list of what they tried to get it to do? Did they choose a correct pattern or template or did they miss that day in class? 

Style, Readability, Documentation, and a proper Design Pattern are absolute musts!  I can help them debug their code from there rather easily. 

That's an argument for de-valuing functionality to perhaps a 40/30/30 with higher weight on BD, icon, Con pane and VI Properties with a lower weight for FP style.

[Gregory]

Hi Fisel, is there any chance you can post the distribution of grades during the month of October for both grading styles?

[Fisel]

Thank you, Jay, interesting ideas. We'd like to hear from others who agree, and from those who do not.

I'd like to suggest that increasing the Functionality points actually helps us evaluate the things you've mentioned are important. It's easy to get trivial functionality points with almost any design pattern, the more complex functionalities are where design patterns distinguish themselves. And since Style and Documentation are graded subtractively, they don't mean much on overly simplistic code.

Breaking it down: Using the old S/F/D breakdown of 15/15/10, it is possible to pass with only 20% of the functionality requirements met. Your suggested 30/40/30 ups that to only 25% functionality. If we move the balance to 30/50/20, they'll need 40% of the functionality to pass.

It seems like 40% functionality is pretty close to what you're suggesting we aim at. It shows they've recognized the problem in its full complexity and are taking a reasonable approach to solving it. Of course, if their Style and Documentation are bad, they'll need a lot more functionality to compensate.

Does that make sense?

Elwood

[Fisel]

Hi Gregory

We won't be posting the numbers, but we are keeping a close eye on them. This change will drop the pass rate, at least until the change is fully absorbed by the community and incorporated into candidates' preparation and assumptions. At this point we believe the overall balance will be positive.

[piterx]

Hello,

I'm preparing for CLD exam and luckily I saw that in time (in December 2019) to revise my assumptions about the exam, but shouldn't it be reflected in CLD preparation guide?

Best wishes,

Piotr Golacki