Simplicity vs. Rigidity
Patrick Logan states the truth of it:
So JDO is a mess (original report), certainly EJB is a mess (hence JDO), but let's face it, the root cause is the language itself.
And it's not any better over there in CLR-ville either. The dead hand of rigid systems is holding the entire software development world back....
Comments
[] May 8, 2004 23:24:48.201
you know, most people actually do not care about the language trivialities that you try to inject into any discussion of smalltalk versus java. they just want to solve a problem and use the most standard solution out there so it will be easier to maintain.
Re: Simplicity vs. Rigidity
[James Robertson] May 9, 2004 0:41:58.972
Comment on Simplicity vs. Rigidity by James Robertson
First off, you must care - you responded. Second, higher levels of productivity are hardly trivial.
[] May 10, 2004 3:12:29.941
ahh..i just visit here to make fun of you and your obsession with java...i don't actually care about smalltalk.
Re: Simplicity vs. Rigidity
[James Robertson] May 10, 2004 8:32:46.207
Comment on Simplicity vs. Rigidity by James Robertson
So just out of wild curiousity, did you have a technical answer, or just FUD?
[] May 10, 2004 11:09:38.190
Instead of smirking on the sidelines looking into the various internal "fights" going on between java factions (where one side or the other shades the facts so as to get a leg up on the other), reflect for a moment on the fact that Java has so much momentum and enthusiasm going for it right now that it can actually accomodate so many different choices for developers. That's the real story, not some half-assed attempt to somehow link spec wars to deficiencies in the core language itself.
Re: Simplicity vs. Rigidity
[James Robertson] May 10, 2004 12:23:32.432
Comment on Simplicity vs. Rigidity by James Robertson
The point is, the JVM cannot support things like Smalltalk or Lisp at all well. The VM has too many assumptions about typing built in; it's not evolved since about 1996, and Sun shows no willingness to do so. This is very like the hold that Cobol took in the 60's. The JVM is old technology, and it's not going anywhere. I have some hope that MS will add useful mods to the CLR in support of dynamic languages - they at least have people on their VM staff who understand the issues. Sun is happy with old technology.
[] May 10, 2004 17:17:33.504
So, what does the JDO/EJB debate have to do with your point, which seems a valid enough one? As you may know, all these arguments between the two camps simply has to do with powerplays. For example, http://alexwinston.com/blog/2004/05/08/1084034150000.html
By criticizing everything about Java you dilute your major points.
Re: Simplicity vs. Rigidity
[James Robertson] May 10, 2004 17:50:36.875
Comment on Simplicity vs. Rigidity by James Robertson
My point is that things like EJB and JDO are way more baroque than they need to be, that's all. Nothing more
[] May 11, 2004 10:38:15.268
If EJB and JDO are way more baroque as you put it, then that is simply a by product of the large numbers of people experimenting with Java and putting their spin on what they believe the best solution is. There are many ways of doing things in Java - I think that is a strong feature, not a disadvantage. If you dislike all the problems of persistence using JDO/EJB, then just use JDBC like many developers do.
I think you are starting to confuse Java the language with Java the platform, as someone put it to you.
Re: Simplicity vs. Rigidity
[Robert Church] May 11, 2004 12:58:19.309
I think James is pretty clear on the difference between the Java language and the Java platform. I've done some work with EJB and I work with a lot of dynamic languages, and James is 100% correct. Most of EJB is a workaround for Java's type system. Look at O/R mapping tools for Smalltalk, Python, et al. The difference is hard to miss.