in reply to Re^3: WebPerl (eLearning)
in thread WebPerl Regex Tester (beta)

apropos a general training site and not necessarily with what the WebPerl site offers, here is an idea: assess students by presenting them with an old question from the PM without the answers. Then slowly reveal answers from the lowest voted to the highest as hints. Or give them a random paragraph from an answer as a hint.

That could also become a "how good at Perl you are" test. PM is a treasure trove for material, don't know who has the copyright.

the WebPerl Regex tester is great and looks great too! well done.

Replies are listed 'Best First'.
Re^5: WebPerl (eLearning)
by marto (Cardinal) on Sep 05, 2018 at 14:27 UTC

    This doesn't make much sense in practice. There are replies which are really good technically at solving the problem which don't have as high a reputation as other responses, somehow gauging them all (since things change, TMTOWTDI and the longevity issue), ranking them then asking a new user/student to choose the least worst solution out of a selection of answers seems to me like a really bad way of trying to teach anything to anyone. It specifically doesn't address "how good at Perl you are", when you're just asking someone to make a choice between other peoples work, ranked by (often) arbitrary voting.

[reply]
      when you're just asking someone to make a choice between other peoples work, ranked by (often) arbitrary voting.

      not make a choice between other people's work but present other people's work (the answers) as "hints".

[reply]

        Again, I stand by what I said. Basing any learning system (even if it's just hints) on answers ranked post reputation is going to cause problems.

        "Or give them a random paragraph from an answer as a hint."

        That would just throw more confusion into the pot. Choose 10 random SoPW threads, start randomly selecting sentences/paragraphs from responses.

[reply]

In Section Cool Uses for Perl