[OS X TeX] New Macros, new Engines, new TeXShop versions, and all that

[David Messerschmitt]

> 
> How is not doing anything to change things automatically not totally transparent. The ``lazy'' user doesn't want anything to change as far as I can tell and if they do want the new stuff they can install it. Lets see... there was a macro that I used and suddenly it disappeared when I updated TeXShop. And I've got to do something to NOT have it disappear!?
> 


Thanks for the valuable discussion, but please read my proposal carefully. Under my proposal, once a user adds a macro, replaces a macro, or deletes a macro, then that action would be preserved through an upgrade. This, I argue, is exactly what a power user wants. So...my proposal meets your stated requirements (as a power user).

The lazy or naive user, on the other hand, is (almost by definition) one who is happy to delegate the choice of macros, upgrade of macros, etc, to the developers. So this user would prefer to just install and use, and still benefit from new macros, improved macros, etc. If the developers decide that a macro is no longer worthwhile or can be replaced by something better, then this naive user is happy to see that change blindly introduced and to live with the consequences. Thus, I disagree with your characterization of the needs of the naive user.

My proposal attempts to meets both of these requirements. However, there is an inconsistency in that a new macro cannot be automatically added for the naive user without also adding it for the power user. There are two solutions to this quandary: Either the power user will have to manually delete an added macro if they dont want it, or there will have to be two installation processes, one for power and one for naive (each user would have to self-identify once, say as a preference item). This is the only inconsistently, fortunately.

There is absolutely no reason that the needs of both the power and naive users cannot be accommodated, if we can agree on their respective requirements.  (This forum is a good place to debate these issues.) One of the (thus far) flaws of open source software is that it has difficulty meeting the needs of naive users. See my debate with Eric Raymond on precisely this point in IEEE Software:

http://www.eecs.berkeley.edu/~messer/PAPERS/03/Point-counterpoint.pdf


David Messerschmitt