Embarassing, I agree with that.

With so small amount of new packages - only one, maximum 2 a year, one would expect them being nearly perfect or at least well tested.

Say, GraphTheory Package has a couple of procedures screwed - but others may be OK (well, OK for Maple). But in this case - only 2 procedures! Not working in Classic and 1 of them not working right even in Standard - it's 50% of the package. Were they tested at all?

In general, I used to like Maple 10-12 years ago. But starting from Maple 6 it is degrading and degrading. Is there anything there that is not broken by now? New packages are broken, integration is broken (as usual), simplify is broken, solve is broken, ArrayTools are broken, evalf is broken, and even some integer procedures are broken!

Alec

For 1 and 2, there should be an additional to Document and Worksheet mode - something like Development mode, or Programming mode, with usual IDE things like color coding, indenting, parentheses matching etc. (and 1-D input of course.) Currently, trailing tildes can be hidden by setting interface(showassumed=0);

3 - I think, it is done like that in the Standard Maple - epsilon_0 looks subscripted and it is not an indexed variable.

4 - can not be done for backward compatibility (some older code would break.) gamma, actually, is not the worst in the list - D, for example, is worse.

Alec

Yes, I thought there about assigning it to NULL, too, but then I would have to use indices, and I tried to write a clean code (meaning that as an instructional example :) Both assigning to NULL and using indices would make code more obfuscated than I wanted. Generally, using indices seem like a good idea. I think, I saw that in someody's else post on this site recently - either Jacques Carette, or Robert Israel, or acer.

I've asked about map and op because it is used in convert/list. Using your version, with nolist, would be faster there.

Alec

Joe,

Is nolist faster than map(op,{indices(T)}) or map(op,[entries(T)])?

Alec

I am not aware of a good independent review - and it might not exist. Being a Java programmer (and beta-tester) for a long time, I, probably, could write it myself. The main idea is that originally, when Java started, it had a big enthusiastic following - it was a time of many .com launches using Java. After the .com bubble blow, the enthusiasm had diminished, but Java still was used on some web sites for interactivity. With years coming, Flash and Ajax replaced Java applets practically everywhere. .NET creation (and mono in Linux) seems to mean the end of Java (at least for me). Everything that could be done with Java, can be done much better with .NET.

_______________
Alec Mihailovs
Maplesoft Member

I am not aware of a good independent review - and it might not exist. Being a Java programmer (and beta-tester) for a long time, I, probably, could write it myself. The main idea is that originally, when Java started, it had a big enthusiastic following - it was a time of many .com launches using Java. After the .com bubble blow, the enthusiasm had diminished, but Java still was used on some web sites for interactivity. With years coming, Flash and Ajax replaced Java applets practically everywhere. .NET creation (and mono in Linux) seems to mean the end of Java (at least for me). Everything that could be done with Java, can be done much better with .NET.

_______________
Alec Mihailovs
Maplesoft Member

Yes, Jacques!

Some people at Maplesoft seem to be listening to you !

That gives some hope...

_______________
Alec Mihailovs
Maplesoft Member

Yes, Jacques!

Some people at Maplesoft seem to be listening to you !

That gives some hope...

_______________
Alec Mihailovs
Maplesoft Member

I agree.  I use C# a lot, and F# is one of my favorite things to play with. It looks as if F# is one of the best possible choices for writing a new CAS, if there was such a demand. At present, it seems as if SAGE (and Python) dominate though - it would need a serious effort (say 50 to 100 capable people for a year or two) to create something (in CAS area) that could approach that (I don't think that SAGE and Python could be beaten in a year or 2).

_______________
Alec Mihailovs
Maplesoft Member

Swing is bad, but AWT was much worse - it created a window for every button, or other element, for instance - as a result, it used much more memory. I used to program in Java myself (and teach Java programming) - and, frankly, I don't think that Java has any future, except, maybe for 5-6 years while people taught only Java and nothing else in college, learn something else.

_______________
Alec Mihailovs
Maplesoft Member

Well, Java is obviously a failure. Early or later (probably, later) it will be replaced with something usable (.NET perhaps).

_______________
Alec Mihailovs
Maplesoft Member

Well, Java is obviously a failure. Early or later (probably, later) it will be replaced with something usable (.NET perhaps).

_______________
Alec Mihailovs
Maplesoft Member

Axel,

I was informed in a personal email that posting on this site is not forbidden by Maplesoft management. More than that - it is encouraged. (Probably, not enough :)

_______________
Alec Mihailovs
Maplesoft Member

Axel,

I was informed in a personal email that posting on this site is not forbidden by Maplesoft management. More than that - it is encouraged. (Probably, not enough :)

_______________
Alec Mihailovs
Maplesoft Member

"Perhaps this option will be included in an update."

I think, you are too optimistic. Besides, what good in having just one of such usual in other programming editors and IDEs things? There are a lot of other things that should be done - automathic indenting, text coloring and highliting, tooltips, debugging etc.

Look at it this way - Maplesoft developers (well, most of them) are not even interested enough in their job to post on this site. Why would one expect them to do something that goes ahead of their assigned duties?

_______________
Alec Mihailovs
Maplesoft Member