Message Boards Message Boards

Avoid trial and error with Wolfram|Alpha?


Often I try to test some Wolfram Language (WL) expression within Wolfram|Alpha (Just the standard Web-Interface or WinPhone App). Its hopeless concerning use of WL and structuring expressions. Some few WL-expressions are understood. Most of them are not understood. Always it ends up in an endless trial and error process, since WA does not understand what I mean.

Unfortunately there is no Wolfram|Alpha documentation since it claims to be self-explaining. But it is very far from that. Is there some link where I can at least learn how to use WL within Wolfram|Alpha?

I can enter


I can enter as well


But I cannot enter


This is misunderstood.

Of course I tried the same with natural language input. Something like

(subsets of {1,2,3,4,5} with size 3) union {{3,4,5}}

This doesn't work either.

POSTED BY: Werner Geiger
1 year ago

Simply said, you can't (generally speaking) use wolfram language directly. Though it would be easy to change wolfram alpha to accept Wolfram Language code. ToExpression[input] basically it could be 'hazardous' as you easily do infinite computations or computations that delete files or connect to other computers or run other commands.

Generally I would not recommend to use W|A to learn WL...

POSTED BY: Sander Huisman
1 year ago

Of course I do not use W|A to learn WL. I use Programming Lab for that. But W|A is handsome for getting some quick answers on small problems. But it's really tedious since it's unpredictable what W|A understands and what it doesn't understand.

Accepting WL-expressions within W|A seems natural to me. And some of them are accepted in deed. Your "hazardous"-argument does not convince me. Of course you can do stupid things with each and every programming language.

POSTED BY: Werner Geiger
1 year ago

Group Abstract Group Abstract