User talk:Dan Bron

From J Wiki
Jump to: navigation, search

You can leave comments, notes, messages, etc for me here. Or you can contact me via the J Forums or directly via email (though note I haven't been good at keeping up with email recently).

J explainer

J explainer like could be done in JHP with client-side JavaScript. Very simply: word formation and lookup to one-line definitions and links to dictionary.
-- Oleg Kobchenko


Regarding recent changes you made to RosettaCode:

1. Exponentiation operator is a very nice little essay. I'm really glad you elaborated it as you did, and I'll be pointing others to it.

    • Thanks!

1. Sorting Using a Custom Comparator still does not sort correctly. The sort must be both longest-to-shortest, and alphabetical among entries with the same length. The original version met that requirement, clunky though I may have been.

    • I'm not sure this is true. I'm pretty sure strings \: ((#&.>strings),.lowercase&.>strings) will sort in the way you suggest (which is what the code does). I'll have to look at it later today.
    • I investigated, and you were right, the the code (above) was sorting by size and reverse alphabetical. Fixed now. The new method is essentially strings /: ((-@:#&.>strings),.lowercase&.>strings)

1. Write float arrays to a text file does not meet the specification, contrary to your assurances. Look at the file it produces and compare it with the output shown in the requirements. Your code manages digits to the right of the decimal, not "digits of precision".

    • Ah. I read "digits of precision" as "digits after the decimal", and disregared the exponential notation (e.g. e+05) as a limitation of the native formatting of the language of the tasker (i.e. Python). I'll rewrite it later today. Won't be as pretty.

1. I look forward to working together on more entries, as time allows.

-- Tracy Harms (aka TBH)

RosettaCode redux

Here are the Rosetta Code tasks that leave me most curious to see a solution in J, but which are beyond my current capabilities. The first three of these baffle me because they involve establishing new types and J has fixed data types.

Formal Power Series
Evens Sum To Even
Defining Primitive Data Types
Dragon curve
Lucas-Lehmer test
Numerical Integration

-- Tracy Harms