In any case, since I am unhappy with TIOBE being taken for granted, without close reconsideration, I decided to do my part in educating people against it using my newly created anti-TIOBE page (which is just another page in my section of pages against bad software). Since this was inspired by this post and its publicity, I guess I should thank you for it.
]]>@neilb: Thanks, i'll have to publicize that in a future post.
@shlomi: The fact that you think an armsrace can take place here means you've missed some facts about TIOBE. Please do poke me on IRC if you'd like me to elaborate. Also, all your links only explain why TIOBE is bad, which is correct because it's faulty, but miss that TIOBE matters despite being bad. Again something i'll happily explain to you on IRC.
]]>However, if this is a pet project to just build software that might already exist then that is your choice.
]]>About “knowledges”. There are many definitions of “knowledge”, also depending on where on Earth you are living. Your definition is certainly an interesting one.
A propos moodle: moodle is rather courseware than learning platform.
Probably have a look at https://www.khanacademy.org/
]]>that example is not relevant. map has to _return_ 100M elements so it need memory.
that does not mean memory taken for 1..100_000_000 range
perl -e'map{print if $_ %1000==0; ()} 1..100_000_000'
which of course will return the empty list. The problem is that for map
Perl will first build a list of the numbers 1 to 100 million as SVs, but for ($x..$y)
though has been optimized in the Perl compiler.
Perl itself doesn't have special iterator support. But you still can write something as an iterator, but you have to do it on your own.
For example you create a function that returns a function that uses an closure. Or you just use OOP for keep tracking of the current state and so on.
]]>you are so right. I liked very much using PDL instead of matlab, using perl and binding sci calculations is fun and also productive for a Perl programmer!
]]>