Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Rails and Django comparison whitepaper

vaporbase.com

Post reply on HN
Print view View original

11–15 of 15 posts

Re: Rails and Django comparison whitepaper

#11

Post by reitzensteinm »

hsiung wrote: post #8
wmf wrote: post #7

Earlier quoted context omitted.

Haven't you heard? Performance doesn't matter; it's all about scalability these days. After all, what's the difference between 100 servers and 200 servers?

Let me get this straight. So you would choose the framework with slightly better plugin support and larger community even if it meant it was three times slower and consumed twice as much memory? If you have deep pockets and have no adversion to unnecessarily managing huge server farms, then go ahead. I've presided over two rails websites in the past and topping out at around 15req/s on average was not fun. Rewriting…

Sarcasm meter broken this morning?

Re: Rails and Django comparison whitepaper

#12

Post by chaostheory »

AF wrote: post #10
chaostheory wrote: post #5

Overall I feel that the paper is good and accurate... There is one thing that annoyed me though (having used both Python and Ruby extensively): "Ruby and Python are really quite similar. For example, Both are strong object-oriented languages." That is NOT True. OO was probably not part of Python's design from the start. It feels really tacked on when you have to call 'self' everywhere, and try doing inheritance past…

The article is correct. In this interview ( http://twit.tv/floss11 ) Guido states that Python has been OO from the outset (1990). The thing you have to understand is Ruby and Python do OO differently. To Python everything is a hash table. This isn't the case for Ruby and their differences mainly stem from that. It adds interesting elements like functions are very simple to pass around in Python because they are not b…

No it isn't. "Both are strong object-oriented languages": the key word (that is wrong) is 'strong'. Based on my examples Python may implement OO better than say Perl but it is NOT a strong implementation of OO compared to other languages.

It's great to know how and why Python is the way it is, but at the end of the day you can give a million reasons, but the kettle will still be black.

What you posted is like me explaining that (current) Ruby is not slow (which is a joke), it just implements threading differently...

Re: Rails and Django comparison whitepaper

#13

Post by AF »

chaostheory wrote: post #12
AF wrote: post #10

Earlier quoted context omitted.

The article is correct. In this interview ( http://twit.tv/floss11 ) Guido states that Python has been OO from the outset (1990). The thing you have to understand is Ruby and Python do OO differently. To Python everything is a hash table. This isn't the case for Ruby and their differences mainly stem from that. It adds interesting elements like functions are very simple to pass around in Python because they are not b…

No it isn't. "Both are strong object-oriented languages": the key word (that is wrong) is 'strong'. Based on my examples Python may implement OO better than say Perl but it is NOT a strong implementation of OO compared to other languages. It's great to know how and why Python is the way it is, but at the end of the day you can give a million reasons, but the kettle will still be black. What you posted is like me expl…

You can directly compare the speed of an application or a language.

It is not subjective to say 'C is faster than Perl' or 'Perl is faster than Ruby'. That is true, in almost every instance, C IS faster than the equivalent Perl code and the equivalent Perl code is faster than Ruby (whether that holds true for Ruby 1.9 we shall see).

On the other hand, what constitutes a 'strong' implementation of OO is purely subjective. Would you consider Common Lisp's implementation of OO 'strong'? It is completely different than how Ruby works. But it isn't any more or less OO, it is a different way of doing it. Now if we really want to get into subjectiveness, Smalltalkers would probably say that Ruby's OO is not 'strong' because Ruby resorts to not making conditionals actual objects, unlike Smalltalk, in which literally just about everything is an object.

You might even say that Python is more OO than Ruby because functions are objects in Python but methods are not similar first-class constructs in Ruby (note: I wouldn't actually argue this, but it is just an example of how the languages do something differently).

Re: Rails and Django comparison whitepaper

#14

Post by chaostheory »

AF wrote: post #13
chaostheory wrote: post #12

Earlier quoted context omitted.

No it isn't. "Both are strong object-oriented languages": the key word (that is wrong) is 'strong'. Based on my examples Python may implement OO better than say Perl but it is NOT a strong implementation of OO compared to other languages. It's great to know how and why Python is the way it is, but at the end of the day you can give a million reasons, but the kettle will still be black. What you posted is like me expl…

You can directly compare the speed of an application or a language. It is not subjective to say 'C is faster than Perl' or 'Perl is faster than Ruby'. That is true, in almost every instance, C IS faster than the equivalent Perl code and the equivalent Perl code is faster than Ruby (whether that holds true for Ruby 1.9 we shall see). On the other hand, what constitutes a 'strong' implementation of OO is purely subject…

I still feel your 1st post was weak, but you make good points in this post - u win; i'm not keen on religious flame wars

Re: Rails and Django comparison whitepaper

#15

Post by jamesbritt »

hsiung wrote: post #8
wmf wrote: post #7

Earlier quoted context omitted.

Haven't you heard? Performance doesn't matter; it's all about scalability these days. After all, what's the difference between 100 servers and 200 servers?

Let me get this straight. So you would choose the framework with slightly better plugin support and larger community even if it meant it was three times slower and consumed twice as much memory? If you have deep pockets and have no adversion to unnecessarily managing huge server farms, then go ahead. I've presided over two rails websites in the past and topping out at around 15req/s on average was not fun. Rewriting…

> Rewriting to Django and getting 45req/s and less memory usage was definitely worth it.

I'm curious: did you consider switching to another Ruby Web framework?

Post reply on HN