tag:blogger.com,1999:blog-6975983.post8447152714379265643..comments2023-09-18T01:22:11.655-07:00Comments on Build and Break: Ruby versus ScalaRichard Taylorhttp://www.blogger.com/profile/14036876973506495788noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-6975983.post-27425097738159793892009-05-02T18:20:00.000-07:002009-05-02T18:20:00.000-07:00I am only giving a summary of the argument. The Tw...I am only giving a summary of the argument. The Twitter crew said "Ruby, like many scripting languages, has trouble being an environment for long lived processes". One thread, I think from Slashdot, was that CRuby is well known to leak memory, and that they should have used JRuby instead. Others countered that JRuby lacked needed features. <br /><br />My experience with server code is that memory is always a problem. We have just spent a lot more time than it should have taken with what appears to be a memory leak in JNI code.Richard Taylorhttps://www.blogger.com/profile/14036876973506495788noreply@blogger.comtag:blogger.com,1999:blog-6975983.post-31759076501870290502009-05-02T16:21:00.000-07:002009-05-02T16:21:00.000-07:00thanks for this article. I'm not sure I follow th...thanks for this article. I'm not sure I follow the third paragraph, though. it seems to imply that dynamic languages are more likely to leak memory and that statically typed languages provide more control over memory. I think dynamic versus static typing is an orthogonal issue. I think the absence or presence of automatic garbage collection is more important. But there are different kinds of memory leakage and as you mentioned even with a garbage collector (as in Java) memory can still leak. The issues there have more to do with whether objects end up getting collected when they aren't really being used anymore, I think. Does static typing give you more control over leakage (e.g., in Java)?Paul O'Rorkehttp://ororke.com/paul/blog/noreply@blogger.com