As discussed earlier, the current listings have both the 1252A and 1252B. Looking at the posted datasheets the only real difference for the A from the B, other than extra frequency support in the high AC ranges, seems to be the higher input impedance at low voltages.
Yep, that's about the gist of it. I haven't looked to see if there are any design issues as a consequence of the differences between the A and B.
1) are there any practical issues that would arise from the impedance difference? I'll admit that I'm not too knowledgeable about what that means for the DMM's capabilities.
The higher impedance is useful if you'll be measuring circuits that have high impedances. You want the meter's impedance to be much higher than the things you're measuring in order to minimize the meter's effect on the circuit. In general, though, the lower impedance of the B should not be a problem. It's still plenty high. The higher impedance of the A might make it more sensitive to noise, but I haven't heard of any complaints. Some bench multimeters have even higher input impedance.
2) is there anything else I'm missing? If the minor datasheet differences aren't important, it looks like the A might be a good choice -- seems to come with a few extras and despite it's being older I'm getting the impression that the cal dates on the B's are around the same time by now.
I haven't heard anything specifically about the A models from nmori (no purchases mentioned here yet), so I don't know how they compare to his B models. I quite like the color of the A models, though I got a B one before the A's showed up.
Also despite rereading the last few pages I'm having trouble telling whether the current batch of B's is having problems or not (and who knows about the A's). I see that the seller is going to stand behind the products but I'd like to avoid any defect problems if possible. Should I be worried about anything here?
It seems a couple of units had issues and are being taken care of by the seller. There's no way to know in advance, apart from opening and testing, if a unit might have a defect. For the bargain price on new-old-stock, you take some risk, but it's not such a big deal since the seller is responding to any issues. Many of us bought them.
Finally, (hope I'm not asking too many questions here) does the older cal date have any practical consequences for a hobbyist use-case?
Unless you need to measure something and know for sure that it is X, not X+0.001 nor X-0.001, for example, there's no practical consequence for hobby use if the meter has drifted a little.
For most electronics, knowing that a value is in the right ballpark is often sufficient. Whether your 5V power rail measures 4.9, 5.0, or 5.1 volts usually won't have an adverse effect on your circuit. If your project requires a meter that is calibrated, you'll know it because you designed it that way, it's documented with such requirements, or you have to adhere to some standards that require it.
I like the high count and accuracy figures for these meters, but if they've drifted too far out of spec that's somewhat moot (and I don't have the equipment to do a nice calibration myself).
Being out of spec doesn't render useless the meter's high count and resolution. You'll still be able to maintain more decimal places at higher measurements than with a lower-count meter. You'll also still be able to detect smaller changes in values (i.e., the difference between two readings) than with meters that have lower resolution. What you lose with significant drift is accuracy, i.e., the ability to know the absolute value of a measurement relative to a known standard.
It sounds like most people have been getting good results but the cal dates appear to be sliding backwards as the seller goes through stock.
Yes, that does seem to be the case.