Hi,
I see some product that use "java code statistics analysis" in order to my JR programmer made program as SR programmer or as AP.
I think that is better for my apps this tools.
I think that this tool could make "robots" and could make that SR programmer to cash more money.
I like see a Open Source tools.
Please, what is your opinion ?
Are you stoned?
>I see some product that use "java code statistics analysis" in order to my JR programmer made program as SR programmer or as AP.
Form and content are not the same thing.
>I think that is better for my apps this tools.
Using some form-based metrics can help flag up code that needs to be refactored
>I think that this tool could make "robots" and could make that SR programmer to cash more money.
Do you mean it turns junior programmers into robots by removing some of their individual quirks, or something else?
>I like see a Open Source tools.
http://checkstyle.sourceforge.net/ checkstyle does have a few metrics as well as configurable coding standards.
>Please, what is your opinion ?
You are either stoned or need to improve your English, as most of your questions are so badly formed as to be almost meaningless.
Sorry,
I need improve my english. I do not write english.
My boss want that I see some "java code statistics analysis" in order to
use JR programmer as SR programmer.
He see a tool that have some check in order to improve code. He think that with this tool, I make the same project but a low cost.
the tool that I see named JKing, have some rules in order to optimize the code.
Thanks ,
(About "bad English": statistically English is decreasing.)
Improving code quality is a good idea.
Javadoc usage is a pre. Say what is intended, what the requirements are,
what is problematic.
Software metrics give information on the entire architecture. They are useful,
as they provide insight in the architecture.
The problem is the interpretation and treatment afterwards.
So is connectivity in all its forms not something to be easily resolved, but important.
And individual source evaluation can give an "error probability" which you can doctor from red to green again - without quality improvement. Indeed productivity may often reduce "quality" a bit.
JR programmers can be usefull. The best to my experience is putting one or two JRs with a SR. This is called peer-to-peer management, and insures that code is checked afterwards.
Code-reviews are the same thing.
Designing software before coding, maybe though with prototypes, would be ideal, as one can talk about the code. The two dangers there are:
- fright to perform in public;
- discussing without sound basis: too abstract, and too long on side issues,
not recognizying the crucial points.
Try to minimalise the number of classes.
Design separation, independent classes in independent package hierarchies. Avoid inheritance.
UML is another way to talk about code.
It is nice to have, but should not be given more weight than it has.
Design patterns are a SR thing, but things like singletons, dependency injection, etcetera should be communicated as concepts, so another
programmer can benefit from reading.
I do not know how it is in your country; in Germany where I live, I have
found, that especially the inexperienced have an almost arrogant attitude,
the experienced are professionally friendly.
Good communication and both freedom for creativity and reviewing/redesigning are a must.
P.S.
The commercial MyEclipse has some of the things you want.
Open source solutions should also exist.
Hi,
In spain, the boss minimized the the experienced programmer. A lot of consultant-buyer and low prices programmer.
Prototypes make product finish.
In spain people get OpenSource code, make revision and put as new product. They do not apport some code to open source.
I think that I could work in Germany or another country