Comments on: Ruby versus Smalltalk, a useless comparison http://blog.amber.org/2005/07/15/ruby-versus-smalltalk/ Thoughts of a minor lunatic Wed, 21 Oct 2009 01:55:36 +0000 http://wordpress.org/?v=2.9.2 hourly 1 By: Omer Raviv http://blog.amber.org/2005/07/15/ruby-versus-smalltalk/comment-page-1/#comment-12864 Omer Raviv Fri, 30 Jun 2006 15:44:11 +0000 http://blog.amber.org/2005/07/15/ruby-versus-smalltalk/#comment-12864 The solution to the REXML performance problem, by the way, is simple. Use one of the libxml2 wrappers under Linux. Or, if you're running under Windows and don't care about interoperability, use MSXML via the 'winole32' library (included in the Ruby distribution). There's even a pre-made wrapper for these COM objects with the Ruby 1.8.4 Windows installer - look up 'xml.rb' in your Ruby directory. The solution to the REXML performance problem, by the way, is simple. Use one of the libxml2 wrappers under Linux. Or, if you’re running under Windows and don’t care about interoperability, use MSXML via the ‘winole32’ library (included in the Ruby distribution). There’s even a pre-made wrapper for these COM objects with the Ruby 1.8.4 Windows installer – look up ‘xml.rb’ in your Ruby directory.

]]>
By: Madhu http://blog.amber.org/2005/07/15/ruby-versus-smalltalk/comment-page-1/#comment-5578 Madhu Sat, 11 Mar 2006 20:00:28 +0000 http://blog.amber.org/2005/07/15/ruby-versus-smalltalk/#comment-5578 Your comment that high performance VM development is hard work is an important one. The ability to easily leverage someone elses hard work seems to gloss over much of the magic that is going on. The result is a tendency of some to trivialize high performance and even argue that it is unecessary. It's true that a lot of code can be written in any language with any level of performance. Often, these are trivial examples, but the backbone of the information age requires some really fast stuff built by some really smart people. Unfortunately, these people are often forgotten because they don't get a lot of publicity. Your comment that high performance VM development is hard work is an important one. The ability to easily leverage someone elses hard work seems to gloss over much of the magic that is going on. The result is a tendency of some to trivialize high performance and even argue that it is unecessary.

It’s true that a lot of code can be written in any language with any level of performance. Often, these are trivial examples, but the backbone of the information age requires some really fast stuff built by some really smart people. Unfortunately, these people are often forgotten because they don’t get a lot of publicity.

]]>
By: petrilli http://blog.amber.org/2005/07/15/ruby-versus-smalltalk/comment-page-1/#comment-2336 petrilli Sat, 16 Jul 2005 15:45:21 +0000 http://blog.amber.org/2005/07/15/ruby-versus-smalltalk/#comment-2336 I spoke too soon. It looks like Ruby runs directly off the parse tree. Wow. I spoke too soon. It looks like Ruby runs directly off the parse tree. Wow.

]]>
By: petrilli http://blog.amber.org/2005/07/15/ruby-versus-smalltalk/comment-page-1/#comment-2335 petrilli Sat, 16 Jul 2005 15:41:14 +0000 http://blog.amber.org/2005/07/15/ruby-versus-smalltalk/#comment-2335 As far as I know, Ruby does what Python does, compiles down to byte-codes and then executes them on a "virtual machine". Perl does this, most do. I could be totally wrong, and if so, woops, but I'm not sure how else you could do it without ending up with some massively ugly code. As far as I know, Ruby does what Python does, compiles down to byte-codes and then executes them on a “virtual machine”. Perl does this, most do. I could be totally wrong, and if so, woops, but I’m not sure how else you could do it without ending up with some massively ugly code.

]]>
By: Tom http://blog.amber.org/2005/07/15/ruby-versus-smalltalk/comment-page-1/#comment-2334 Tom Sat, 16 Jul 2005 09:31:06 +0000 http://blog.amber.org/2005/07/15/ruby-versus-smalltalk/#comment-2334 I wasn't aware that Ruby (in its current released versions) even had a VM to be optimised! I wasn’t aware that Ruby (in its current released versions) even had a VM to be optimised!

]]>
By: petrilli http://blog.amber.org/2005/07/15/ruby-versus-smalltalk/comment-page-1/#comment-2333 petrilli Sat, 16 Jul 2005 05:02:07 +0000 http://blog.amber.org/2005/07/15/ruby-versus-smalltalk/#comment-2333 Well, as someone pointed out to me on the Smalltalk IRC channel, the VM underlying VisualWorks (my Smalltalk of choice) has had 20 years of development, if not more, by the smartest people in the world. It's sad to say that these sort of things are truly _hard work_, and so I don't hold out a lot of hope. This is not to say Ruby is "slow" or "too slow", just simply that it's different. People should always work in what they feel productive because in most cases, developer time is the most precious commodity. Well, as someone pointed out to me on the Smalltalk IRC channel, the VM underlying VisualWorks (my Smalltalk of choice) has had 20 years of development, if not more, by the smartest people in the world. It’s sad to say that these sort of things are truly hard work, and so I don’t hold out a lot of hope.

This is not to say Ruby is “slow” or “too slow”, just simply that it’s different. People should always work in what they feel productive because in most cases, developer time is the most precious commodity.

]]>
By: Vincent Foley http://blog.amber.org/2005/07/15/ruby-versus-smalltalk/comment-page-1/#comment-2332 Vincent Foley Sat, 16 Jul 2005 03:49:25 +0000 http://blog.amber.org/2005/07/15/ruby-versus-smalltalk/#comment-2332 It will be interesting to see if Ruby's performance improve with Ruby 2.0 and the new VM (YARV). It will be interesting to see if Ruby’s performance improve with Ruby 2.0 and the new VM (YARV).

]]>