That's what I was trying to say, but after a couple of glasses of Burgundy.....

add( isprime(n), n=1..10000 ); returns 8771*false+1229*true the coeffs being the number of priomes, not their sum. However, coeff( add( n*isprime(n), n=1..10000 ), true); does do what is required reasonably quickly! Toby

add( isprime(n), n=1..10000 ); returns 8771*false+1229*true the coeffs being the number of priomes, not their sum. However, coeff( add( n*isprime(n), n=1..10000 ), true); does do what is required reasonably quickly! Toby

Tim, Yes, it's now writing using API calls, which is good. What is not good is the choice of writing to LocalSettings since that location is intended for data specific to the particular machine and so it is not backed up to the user's roaming profile. So if you save your preferences on one machine on a network and then move to another machine they are lost. It seems a bizarre decision by Maplesoft, unless there is some way round it or I have misunderstood the situation. Toby

I am told that with Maple 11 something has been done to fix the problem described above, but a poor choice has been made. So now Maple 11 tries to write to the User's profile, but it writes to the "LocalSettings" area. This area is intended for settings particular to one machine that one does *not* want replicated back to the user profile. Consequently, users still can not make changes to their preferences that survive to a log on with a different machine. Why was this done, and is there some way round it? Toby

So it does! I've passed it on to my colleague. Thanks, Toby

So it does! I've passed it on to my colleague. Thanks, Toby

Because I am at the mercy of the Uni's annual upgrade program.

Because I am at the mercy of the Uni's annual upgrade program.

It is Maple as Microsoft might do it - supposedly user friendly but it doesn't work. I spend all my time getting students to use Maple properly to switch their preferences to worksheet and input to Maple input. Unfortunately because of some oddity of Maple's implementation, Maple doesn't save preferences properly on our system - groan. Why Maple could not have introduced document blocks for those who want them without switching the basic defaults I'll never know. Anyway, if somebody is too stupid to use normal Maple input, they probably shouldn't be doing mathematics at all. Rant Over.

It is Maple as Microsoft might do it - supposedly user friendly but it doesn't work. I spend all my time getting students to use Maple properly to switch their preferences to worksheet and input to Maple input. Unfortunately because of some oddity of Maple's implementation, Maple doesn't save preferences properly on our system - groan. Why Maple could not have introduced document blocks for those who want them without switching the basic defaults I'll never know. Anyway, if somebody is too stupid to use normal Maple input, they probably shouldn't be doing mathematics at all. Rant Over.

Indeed yes, but that doesn't change the validity of what I said about the usual convention for the meaning of the square root symbol when applied to positive real numbers. For real numbers it is possible to make a continuous choice, and the usual convention is to take the positive one.

Indeed yes, but that doesn't change the validity of what I said about the usual convention for the meaning of the square root symbol when applied to positive real numbers. For real numbers it is possible to make a continuous choice, and the usual convention is to take the positive one.

Joe, Precisely. I was about to answer that but fortunately you got in first with your very clear explanation. The more I think about it, the more it seems a mess to me. As you imply, neither output nor Q are reserved words in Maple. Logically then if you should protect one then you should protect the other. So the help page should suggest 'output=Q' not output='Q' and our version should be 'output=:-Q' which will I guess work in all circumstances. Alternatively, one could take the view that the documentation is correct and the implementation buggy - how difficult would it be to write JordanForm so that 'output=Q' worked whether or not Q was a defined local variable. I begin to have sympathy with some of my colleagues who condemn Maple as a syntactic muddle. Toby

Joe, yes I guess you're right and that's the reason. I'd have thought that once one had strings then that would be clearly the syntactically clean way to go - it seems to have taken a while to notice this. I've not seen the Diff Geom package yet - we're still on Maple 10. Regards, Toby