Odd Question

If the length of the round slot is long enough (judgement call here) then I always switch it to a square slot, for a couple reasons.  One being that it takes less hits to measure a square slot than a round one (if you do it right, anyway, 6 hits is NOT enough for a round slot, 3 at each end, while a square slot, width measurement, is 6 hits).  Two, when working with pierce sheet metal, the actual radius at each end can be slightly boinked, so by measuring it as a square slot, you get a better length result over round slot since the square slot (width measurement) measure the flat sides first, then takes hits at the mid-line of the sides on the ends of the radii.

I work with castings and solid pieces of steel or aluminum, very few sheet metal items. Normally, I do 4-6 hits on each end and 2-4 along the flats. A part was checked and all the slots reported as .005" below nominal. I knew that the CMM result was correct because the machinist told me he had a wrong tool and confirmed it with hard gages. This new guy was going to write off the .005" deviation as "machine error" because he claimed PC Dmis cannot calculate round slots correctly. I could see maybe .0005"-.001", but .005" seems a bit excessive.

This guy is a younger programmer and is constantly criticizing older programs that our other programmer and myself wrote and have been using for years. At first, it was just we were not using the latest and greatest features. Now, he is telling us things like PC Dmis cannot calculate round slots or Keyin dimension can be used to insert text/comments in reports.

I'm trying to figure out if he is full of it or there is something I was unaware of. It's been about 10 years since I've been through and formal CMM training.

I work with castings and solid pieces of steel or aluminum, very few sheet metal items. Normally, I do 4-6 hits on each end and 2-4 along the flats. A part was checked and all the slots reported as .005" below nominal. I knew that the CMM result was correct because the machinist told me he had a wrong tool and confirmed it with hard gages. This new guy was going to write off the .005" deviation as "machine error" because he claimed PC Dmis cannot calculate round slots correctly. I could see maybe .0005"-.001", but .005" seems a bit excessive.

This guy is a younger programmer and is constantly criticizing older programs that our other programmer and myself wrote and have been using for years. At first, it was just we were not using the latest and greatest features. Now, he is telling us things like PC Dmis cannot calculate round slots or Keyin dimension can be used to insert text/comments in reports.

I'm trying to figure out if he is full of it or there is something I was unaware of. It's been about 10 years since I've been through and formal CMM training.

Not sure if your new programmer is full of anything, but I can confirm 1.  PC-DMIS can measure round slots and 2. The Key-in dimension item is a handy tool and works as he describes. 

Step 1: Roll up your morning paper
Step 2: Wait for the young buck to say something about your old programs
Step 3: Smack him on the nose with your rolled up newspaper and firmly say "No!"
Step 4: Tell him to sit down and explain himself with small words and show you exactly what he is talking about
Step 5: Brace yourself because the lil *** might be right! 

Once you have completed steps 1-5 a few times though, you will both like each other more and both have learned a few things. 

Yes you can measure round slots.
Yes you can key in dimensions and have them displayed on the report, comments as well. 
Yes .005" is too much to shrug off as "machine error" (see steps 1-5) (however if the alignment is poor, or the slot is far enough off location .005 "error" could easily happen)

"Hey, remember that time you hit me with a rolled up newspaper? Yeah, anyway, how's the wife?"

I have always used comment input, under report commands, for operator inputs other than dimensions. I did a lot of programming in C++, C#, and Java prior to learning the CMM. The comment input was similar to stuff I did in earlier programming languages. I will have to look into key-ins more.

You can always verify the slot (measured (auto)) with measuring each end of the slot with circles.  Add the 2 circles (distance) + add Radius to verify.

The trick with operator inputs (and maybe key ins? I have not really used them) is to convert that into a generic feature so that it doesn't mess up the flow of your report. Then it can have tolerance and look like any other dimension. 

Machinist Aproxometers are even faster for double checking things like that!  

The thing is, I am not entering in just dimensions. I have reports that have to include the name of the operator running the program, the gage ID and calibration due date of the gages used for hard checks, part serial and lot ID codes, and other things using letters and/or symbols. So I use comment inputs, capture the input as a variable, and put it in the report as a report comment. 

I have had 3 people, outside of this forum, tell me that I can do all of this with dimensional key-in. Yet, I have not figured out how and none of the 3 people that told me I can, could show me how.

Look into trace fields :) 
Or baring that a custom form so you can easily collect the data in one place (I am not a form or tracefield guru) 
We have a form at the start of the program here that I have edited some that captures serial number, has a drop down for operator and shift, along with several fields of optional data. 

Asking on the forums is a great way to find out what you need too! 

I would have to imagine they mean Tracefields as ArkansasBeeper had suggested. Key-in Dimensions only accept numerical inputs to my knowledge. But you can do things like set an assigned variable to be the measured value, and it will display the Variable's value when the Key-in Dimension pops up and not just "V1". It must also be numerical, or it will just display 0. Either that or they simply mean standard Input comments. I'd certainly like to see their answer if they can ever truly give you one.