[Jython/Java] Which books, which IDEs, which other tools?

[charles]

"Alex Martelli" <aleax at aleax.it> wrote in message news:<a2ri74$r9t$1 at serv1.iunet.it>...
> <pirx at mail.com> wrote in message news:a2psar$slq$1 at si05.rsvl.unisys.com...
>     ...
> > wait and see which way it will go. But overall it seems to me that in a
>  few
> > years we will have most IDEs based on one of three standards: VS/.NET,
> > NetBeans, Eclipse - from economic/engineering/complexity necessity. Python
> 
> Interesting hypothesis.  Why wouldn't Komodo survive, for example?  Or,
> say, PythonWorks?
> 
I look at it from the point of view of a potential added-value
developer/team.

1. Both NetBeans/Eclipse are supported by large sponsors with vested
interest - they use that codebase as a base for their commercial
offerings (Forte/WebSphere development). That provides extras like
professional documentation, coordination of efforts of individual
contributors, longevity, long-term focus etc. Also standardization of
the baseline architecture, interfaces etc. Stability.

2. They are both Java-based, which brings a wealth of Java platform to
any added value work there. One could argue that (free) XML parsers
are available on LINUX - but try to find a free metadata repository
there. Rule-based engines etc. All of that makes for less work if one
wants to add a new language or interface  to NetBeans/Eclipse. Just
consider the problem of adding, say, UML diagramming to your favourite
language and consider what's available already under Netbeans and
Java.

3. All of that adds up to enormous savings. One place I know spends
total 50 person-years/year for their IDE, which is only a part of
their product. Of that at least 40py/y is spend on stuff that is basic
and handled much better by NetBeans/Eclipse. I strongly suspect that
similar costs are borne by many other  producers. (I hear that IBM's
VisualAge for Java costed about 300person-years/year). Even a small
IDE will cost 10py/y to modify/maintain etc. Of which 80% will be
spend on 'obvious' stuff. Producers which will build on existing base
will quickly win, as they will be able to deliver more functionality
at less cost. This is why all other IDEs are in long-term trouble -
unless they have a powerful backer. This is getting too long-winded,
sorry.



> > will probably be gently pushed that way anyway.
> 
> If Python can be developed with the same IDEs also used to develop
> other parts/components in other languages, that surely does sound
> like a good thing.
> 
> 
> Alex