dsPIC-EL or Explorer 16?

Back in June, I walked through a sequence of exercises with another amateur, W8LSS, on the dsPIC33FJ256GP710 using Microchip's Explorer 16 Development Board.

More recently, my thoughts have been along the line of building a "dsPIC-EL" board for the student.  But in organizing the various repos of code associated with this project, I got to wondering if perhaps the Explorer 16 might not be a better choice.

There are two main disadvantages, and one more minor, to using the Explorer 16 instead of a purpose built board:

1. Cost - the Explorer-16 would cost the student about $50 more than the estimated price of a dsPIC-EL, perhaps a little more.  Since the dsPIC-EL is estimated to cost more than the PIC-EL and does not include a programmer, the project cost to the student is already fairly high.  Adding to it seems to be a poor plan.
2. The Explorer 16 uses 3.3 volt parts.  The intent with the dsPIC-EL is to use a 5 volt part which is somewhat more hobbyist-friendly.  Also, although dsPIC33's and PIC24's are available in DIP, students seem reluctant to move past the part they learned on, so that could be an impediment to later experimentation.
3. The LCD display on the Explorer 16, although it uses more or less standard code, has some features that are lacking in typical LCDs.  This could mislead the student.  Worse, not all Explorer 16s have the same LCD.

But there are advantages, too:

1. The Explorer 16 is available already.  There is no design work needed, no new circuit board to be developed.   Part of our thinking is that we might talk AA0ZZ into building and marketing a dsPIC-EL.  Although he seems receptive to the idea, we still haven't heard that he is actually willing. (In fairness, we haven't pressed, either, preferring to have more content available before moving ahead).
2. The Explorer 16 comes with both a dsPIC33 and a PIC24, giving the student the ability to explore both families.
3. Microchip provides low cost "PICtail Plus" prototyping boards, giving the student the opportunity to haul out the soldering iron and add his own ideas.  In at least one of the experiments I did with W8LSS we did just that.
4. Because of the large pin count format, the board has more peripherals to play with.
5. Microchip provides a number of additional PICtail Plus boards for a variety of purposes which might interest the advanced experimenter.
6. I already have a pretty good series of exercise programs in place and tested with W8LSS.

I think I'm still leaning toward the dsPIC-EL idea, perhaps because it seems like a logical follow on to the original PIC-EL, which was amazingly successful.  But at this point there is still nothing cast in stone.