in reply to Re^5: Perl regex speed
in thread Perl regex speed

stop posting links to pretty pictures

Would you prefer pictures that are not pretty???

Replies are listed 'Best First'.
Re^7: Perl regex speed
by eyepopslikeamosquito (Archbishop) on Oct 28, 2022 at 23:24 UTC

    Would you prefer pictures that are not pretty???

    Sorry, I should have been clearer. I was not asking for pictures at all. I smelled an Apple Fanboy and the supporting pretty pictures looked more marketing than science to me.

    Given the OP claims to be a University professor, I was frankly astonished to see these pretty pictures being presented as evidence. I expected to see peer-reviewed academic papers, or at the very least, links than contain source code along with details of how the benchmarks were run, so they can be independently reviewed and verified by others.

    Update 28 Nov 2022: Noticed this Quora story in my inbox today asking: Why is the M1 chip in Mac far overrated over the core i9 11th generation?

    One random quote that caught my eye:

    There’s a war of words going on between Apple and Intel fanboys here. The raw performance of some Intel chips exceeds the performance of the M1 and M1 Pro. On performance per watt Apple wins hands down.

      Sorry, I should have been clearer

      No - I was just being pedantic...

      I have to agree that I too was doubtful of the claims of being "for teaching". I see there are lots more comments from trusted sources (read: respected Monks) that I shall go and read through. It could be good for a giggle!

Re^7: Perl regex speed (Form follows function)
by LanX (Saint) on Oct 28, 2022 at 23:48 UTC
    "Pretty <xxx>" is an sarcastic term attacking the trend of "Form over Function" °

    This is one of the fundamental conflicts software designers have, if they are asked by amateurs (aka bosses) to dumb down carefully designed interfaces into something prettier (sic).

    In this case: Charts look better with less text and appealing colors, but loose crucial information.

    E.g: I'm afraid the new metacpan interface is an example for sacrificing function for form.

    Cheers Rolf
    (addicted to the 𐍀𐌴𐍂𐌻 Programming Language :)
    Wikisyntax for the Monastery

    °) Apparently this conflict is much older than IT and originates in architecture ... Makes sense :)

    see Form follows function

      I'm afraid the new metacpan interface is an example for sacrificing function for form

      Assuming you mean metacpan.org, yes, I see your point. The screen looks a bit too vacant for my tastes.

      Though not a Web Site designer, this interface looks to me like it's optimized for novices, rather than experts ... such as the LanX. :).

      From the "GUI Design Checklist" section at On Interfaces and APIs the two most relevant points are perhaps:

      • Define personas; design to satisfy their goals
      • Cater for both novice and expert

      I wonder if they consulted and tested with expert CPAN users when creating this new design.

        > Assuming you mean http://metacpan.org,

        yes,

        > I see your point. The screen looks a bit too vacant for my tastes.

        It's not all bad but for me it only feels like a first draft.

        And the "design jumps" are irritating while navigating.

        I don't see myself as a web "designer", so I'm lacking the right vocabulary to express my unease.

        It's like a simplified look and feel was "borrowed" from another simpler site and retrofitted over a more complex interface.

        I wished they had kept a switch to the old design as an option ... or at least as reference.

        >

      • Define personas; design to satisfy their goals
      • Cater for both novice and expert

        Funny, I listed very similar things when I gave a talk about principles which should be kept in mind when trying to evolve Perl.

        Cheers Rolf
        (addicted to the 𐍀𐌴𐍂𐌻 Programming Language :)
        Wikisyntax for the Monastery

        I am slightly surprised that 9 people have downvoted On Interfaces and APIs

        Either they are extremely good at interface design or they fail to see the wisdom of the content. I strongly suspect the latter as I see very few examples of excellent (web) interface design. It's something I look for as I am writing a book on the subject to help business owners find, brief and manage web designers so that the result is a website that serves the business needs more than the designer's ego.

      This is one of the fundamental conflicts software designers have

      All too often I see this in website design. It manifests itself as websites that are so beautifully formed that they are either impossible to use or impossible to edit or both. Often I see web design people suggest a complete rebuild to cure the problems when (usually) iterative improvements would be better. It would certainly be easier to test what is working. But there is good money in producing beautiful websites...money that is banked well before the poor conversion performance becomes apparent.

      In their book Making Websites Win, Karl Blanks and Ben Jesson (two people who truly understand website design and two truly lovely gentlemen) liken aesthetic design to asking Mo Farah to run carrying an egg and spoon!