http://java.sys-con.com/node/965189
I am certainly going back to Java, just as soon as I figure out how to configure Tomcat.
Until then, anyone care to take on this logic? Has anyone here used Java the way they say they should?
Phlip wrote:
http://java.sys-con.com/node/965189
I am certainly going back to Java, just as soon as I figure out how to configure Tomcat.
Until then, anyone care to take on this logic? Has anyone here used Java the way they say they should?
I like what he showed of the model-driven nature of OpenXava -- that seems extremely cool (if it actually works). It's too bad that he chose to write the article as a dis to Rails (which he clearly doesn't know a damn thing about) rather than as a plug for OpenXava.
He should have remembered: when writing articles, never eschew the opportunity to fail to neglect to use a *positive* viewpoint.
-- Phlip
Best,
Phlip wrote:
it says right here!
well if it said it on the interwebs, then it *must* be true.
lol
in all seriousness, take a look at first few paragraphs. there are an awful lot of qualifiers. ie, "this article *tries* to demonstrate"... "I prefer" comes up a lot...
At the end of the day, this is going to depend on your style, preferences, and what you're used to.
But I can make my judgement simply by *looking* at the page. It's poorly constructed. When I loaded it, it sent me to the very bottom (to the "Top Stories" portion). The design is cluttered and certainly not streamlined. Portions of the page are cut off.
finally, there's personal experience. I can't believe that anyone who has designed and implemented webapps in Java and Ruby on Rails could possibly argue that Java is faster/easier/more straightforward (that is to say -- I have designed and built webapps in both. Rails was waaaaay faster for me). Furthermore I find that the Ruby language is more concise and intuitive (in general -- there are one or two exceptions).
Steve, you don't need to go too far, just take the subtitle and the very first sentence:
"This article demonstrates that Java is more productive than Ruby" "This article tries to demonstrate that Java can be more productive than Ruby."
Some of the comments on rails are a bit out of date (ie the bit where you have to write sql by hand rather than write a migration and having to restart the server to pick up code changes.). What I can't tell from there is how easy it is once you want to change the default appearance, eg I can get a 0 code generation admin interface with activescaffold but once you start to want significant changes to that interface it can get pretty tortuous. I'd also get pretty tired of writing those accessors.
Fred
Perhaps he wanted to piggy back off the fame of ruby & rails
Frederick Cheung wrote:
Some of the comments on rails are a bit out of date (ie the bit where you have to write sql by hand rather than write a migration and having to restart the server to pick up code changes.).
You... you mean showing off you don't know enough about Rails to work with it is not a valid way to compare?? Gosh!
So this guy’s saying that using one library that someone has put a TON of effort in for scaffolding is better than using another library’s scaffolding system?
How is this Java vs Rails (… god what is with people and HORRIBLE comparisons. Language vs Library?!)?
Java is Fail, it leads to people thinking like the poor sod who wrote the post.
Jason
Phlip wrote:
Until then, anyone care to take on this logic?
Actually I found the following statement about sums it up:
"The productivity in Java world is a cultural problem, not a technical one. That is this is not a Java fault, it's our fault, we, the Java developers, need to design very beautiful architectures..."
I have nothing against Java however I look at the example Recipe.java and it reminds me of one of the reasons I switched to using Rails in the first place...would I prefer a 50+ line model full of getters & setters or a succinct 2 line one in Rails?
really want to see & code is the business logic for your app.
Luke
I tried learn Java language. Five years ago I went to a Java course at a well known university here and it was so boring. In fact I enjoyed it cero minutes learning something about Java language, at point that I enjoyed much more to learn something about object techiques and that was the most exiting in that course, but zero enjoying from the Java side.
Then I worked in state offices where there was a 'team' of javaboys, and the results I think were not at all good. To build an simple application, they were an entire year with a lot of troubles and without generate something elegant or at least reliable. Well, I know that maybe it was fault of programmers.
I prefer go back to Php code or learn Django framework and learn a good language as Python before go with Java.
Cesar
PS: Of course, all that I wrote is from my point of view and without offenses
This article seems to have very dated information (released May 16, 2009? Is the writer using Internet Archive for his browsing?).
Furthermore, the approach of (seemingly) only having control over the model seems ridiculous to me. The presented framework seems useful only for "throw-away" applications that don't really do anything interesting.
That all said, I'm obviously making the same mistake the writer of the article made; I haven't done much extensive research into the capabilities of OX, and can't speak to it's benefits or detriments. But I also haven't written an article titled "RoR is better than Java LOLOLOL", so there you go.
I saw that article a while back... the system described (OpenXava) looks great, as long as you want your site to look just like all the others created with that framework. Maybe my Java-acronym-parsing skills are rusty, but I couldn't find anything on the OpenXava wiki about theming, skinning, etc. A truly determined response to that article would show how easy it is to change the Rails scaffold code, and how *not* easy OpenXava makes it.
Personally, I count the number of Java refugees at my local Ruby
Brigade meeting, and that's all I need to know...
--Matt Jones
Altair wrote:
This article seems to have very dated information (released May 16, 2009? Is the writer using Internet Archive for his browsing?).
I don't know enough Java to read the article, but at a guess the author is bragging about features that Rails inspired in a Java library... then inexplicably claiming Rails does not have those same features.
Also, I could not find the word "test" in the article...
Phlip wrote:
http://java.sys-con.com/node/965189
I am certainly going back to Java, just as soon as I figure out how to configure Tomcat.
Until then, anyone care to take on this logic? Has anyone here used Java the way they say they should?
Here is my biggest problem with the entire article:
"Model Driven Framework"
My day job is as a Java developer, and I can testify that this is the line of thinking of most Java developers I know. In my opinion "Model Driven" design is exactly upside-down. End users don't care one bit about how beautiful, or elegant, your model code is. They don't want to see a user interface that exposes the design of the model.
What users care about is the experience. The interface that allows them to accomplish whatever task is at hand. If that means pulling data from 20 different model objects and mashing them together on a single page, then so be it. The tools the developers use must be able to support whatever presentation makes sense for the end user, not for themselves to make development life easy.
This is a line of thinking that Ruby on Rails confirmed for me as a developer.
So he compares OpenXava's database schema ability to Rails'......... hand-coded SQL?
What?
I especially like this part:
@recipe.date = Time.now
The equivalent in OX is adding the @DefaultValueCalculator annotation in model:
@DefaultValueCalculator(CurrentDateCalculator.class) private Date date;
You obtain the same effect in a more declarative way.
I'd like to meet the RoR dev that creates a date column and manually uses that instead of the magic created_at/updated_at timestamps.
Agreed. That's what Behaviour Driven Development (a la Cucumber) is all about. I can remember a recent example where I was starting with the model and ending up with attributes I didn't need. When I did it the BDD way, the models only contained what they needed to support the views (which the user cares about).
I've learned the hard way (i.e. by maintaining my own crap!) not to overcomplicate things with factory this and generic-abstraction that. It takes some experience to know the difference between a simple solution and a hack/quickfix. I'm still not there yet.
Robert
I couldn't have said it better than Robert.