The Utilities add-in for the Prizm also contains a calculator locking feature that enables you to "lock" the calculator with a password that's different (and can be stronger) from the one used to protect the user information.
---
amazonka: Casio made Geometry, Picture Plot and etc. add-ins and not built-ins because they specifically want them to be erasable, as some schools and exams may not allow their functionality. Otherwise, and just to mention one example, the Conversion function could be always on, without the need to check for the presence of that silly Conv.g3a.
While add-ins can be removed, they can also be added later, and thus there is no functionality loss. Your argument could also be used against the exam mode included in the newest French versions of Casio calculators, "data storage and add-ins were advertised in the packaging so activating the exam mode constitutes a loss of functionality". Indeed it does, but it's a temporary loss, and an intended one.
After thinking a bit, you'll also realize that if Casio were to make all add-ins as built-in apps, then the OS wouldn't support add-ins, and then there would be no easy way to run custom code in it. In fact, this is what Casio did with the later versions of the Classpad, where the only "add-in" ever made is in fact akin to Conv.g3a on the Prizm, which is only used to enable functionality already coded into the OS:
TeamFX (in an email to me) wrote:
The compiled Algy code is included in the ClassPad II OS and Algy2.c2a is a fake add-in. Only the name and file size are checked by the OS.
As for installing add-ins to outside of SMEM it would be even harder than modding the OS to automatically start an add-in (traditionally installed in storage memory) on startup, which would increase the security of lock add-ins while not bricking the calculator. People could still use the bootloader to erase the flash/update the OS, which is a good thing since you'll be surprised at how many people forget their passwords, or have their friends pull a prank on them by changing the password to one that is not of their knowledge.
But even this solution is hard to implement with the current state of things.