View mode: basic / threaded / horizontal-split · Log in · Help
December 18, 2011
Bitmapped vector tries vs. arrays
This is bound to turn interesting:

http://www.reddit.com/r/programming/comments/nddlj/extreme_cleverness_functional_data_structures_in/


Andrei
December 18, 2011
Re: Bitmapped vector tries vs. arrays
Le 18/12/2011 09:22, Andrei Alexandrescu a écrit :
> This is bound to turn interesting:
>
> http://www.reddit.com/r/programming/comments/nddlj/extreme_cleverness_functional_data_structures_in/
>
>
>
> Andrei

Interesting speach, but thoses datastructures are kind of trying dealing 
with a nail using a screwdriver. Problems solved here are not problem 
that nicely solve in functionnal, so why use functionnal ?

I did always think it is stupid to enforce a paradigm because it is 
supposed to be supperior, blablabla . . .

Java did it with OOP, and we end up with many stupid thing, static main 
within a class. OOP is working for some stuff, and not for others. Just 
as functionnal.

Secondly, I think this conf is going out of sanity during the butmap 
vector trie presentation (about 40 mins). Many claims not backed with 
facts, and messy explaination (the cache stuff doesn't exlain in any way 
how this datastruct can be faster than ArrayList, and I bet it isn't 
expect on a flawed benchmark). No idea why polymorphism is revealant 
here either - but this is most likely an error and not the word the 
speaker wanted to use.

Some argument are not convincing. OK, log32(n) is close to O(1) in many 
cases but it isn't O(1). Matter of fact.

It look like the author try to convince us that oranges can taste like 
banana, and show us many techniques to go in that direction, but this is 
useless as logn as we have banana available.

Thoses datastruct are interesting anyway on a research perspective.
December 18, 2011
Re: Bitmapped vector tries vs. arrays
Andrei Alexandrescu:

> http://www.reddit.com/r/programming/comments/nddlj/extreme_cleverness_functional_data_structures_in/

You need to be registered to download the PDF slides :-(

Scala language seems rather compact and powerful to define such data structures. The function balance() at 24.21 to balance red-black trees using just pattern matching is quite nice, despite being a bit long.

But I don't know what those operators like  +A   >:   <:   mean.

Bye,
bearophile
December 18, 2011
Re: Bitmapped vector tries vs. arrays
On 12/18/2011 04:48 PM, bearophile wrote:
> Andrei Alexandrescu:
>
>> http://www.reddit.com/r/programming/comments/nddlj/extreme_cleverness_functional_data_structures_in/
>
> You need to be registered to download the PDF slides :-(
>
> Scala language seems rather compact and powerful to define such data structures. The function balance() at 24.21 to balance red-black trees using just pattern matching is quite nice, despite being a bit long.
>
> But I don't know what those operators like  +A>:<:   mean.
>
> Bye,
> bearophile

+A (-A) means A only occurs in covariant (contravariant) positions, the 
other two are used for constraints, where >: means supertype-of and <: 
means subtype-of.
December 20, 2011
Re: Bitmapped vector tries vs. arrays
On Sun, 18 Dec 2011 15:23:10 +0100, deadalnix wrote:

> 
> Some argument are not convincing. OK, log32(n) is close to O(1) in many
> cases but it isn't O(1). Matter of fact.
> 

Well, he talks about it, and we must admit that for the memory you *can* 
address, log32(n) is really, really fast. Matter of fact. :)
December 20, 2011
Re: Bitmapped vector tries vs. arrays
On 12/20/11 4:44 AM, Dejan Lekic wrote:
> On Sun, 18 Dec 2011 15:23:10 +0100, deadalnix wrote:
>
>>
>> Some argument are not convincing. OK, log32(n) is close to O(1) in many
>> cases but it isn't O(1). Matter of fact.
>>
>
> Well, he talks about it, and we must admit that for the memory you *can*
> address, log32(n) is really, really fast. Matter of fact. :)

I agree. Well, anywhere between 1 and 7. I only have a bit of a problem 
with making a number between 1 and 7 equal to 1.

Andrei
January 11, 2012
Re: Bitmapped vector tries vs. arrays
== Extrait de l'article de « bearophile (bearophileHUGS@lycos.com) »
> Andrei Alexandrescu:
> >
http://www.reddit.com/r/programming/comments/nddlj/extreme_cleverness_functional_data_structures_in/
> You need to be registered to download the PDF slides :-(

I don't think from githup you can have them by clicking on the raw link:

https://github.com/strangeloop/2011-slides/blob/master/Spiewak-FunctionalData.pdf

https://github.com/strangeloop/2011-slides/

Regards,
renoX
January 11, 2012
Re: Bitmapped vector tries vs. arrays
renoX:

> https://github.com/strangeloop/2011-slides/

Oh, very good, thank you.

Bye,
bearophile
Top | Discussion index | About this forum | D home