Re: BeanShell alternatives (was Re: Status of projectand adding 1.5 generics)

> That said, BeanShell2 does continue to work properly because the
> enhancements have been minor.  I challenge anyone here to find a new bug
> introduced in BeanShell2, that is a bug in BeanShell2 that is not in
> BeanShell.

The change to how .java files are located ?  If someone relied on that, then 
the BeanShell2 "feature" is a "bug" to them.

>
>
>
>> I also disagree that limiting changes to bug fixes makes a project dead.
>>
>
> Can you think of a single open source project that is limited to bug fixes
> that isn't dead?

Any implementation of a specification or standard, for example ... XSLT, 
XQuery, C, C++ etc,sh, ksh,bind, TCP/IP , the gnu standard C library and 
common tools ... (when's the last time a "feature" was added to fopen 
strcpy, cat, cp, ls, ....  )  I wouldn't call any of those "dead".
Once the standard is completely and exactly implemented, anything but bug 
fixes are a detriment, not a feature.

Also "mature" open source projects such as Gallery 1, spamprobe,

But you do have a point.  Most "active" OS projects (not quite the same as 
!= "dead") like to change ... I think that's often because the developers 
love writing code. (I do !)

But I don't think that's always a feature.  Its a constant effort to keep up 
with evolving programs.

Not saying I don't like things to get better ... But in BeanShell's case in 
particular when the original author is no longer contributing, and there 
really hasn't been anyone with his vision and expertise and dedication in 
this project to take it over as well as the author (no insult intended),
feature addition is a big risk.
Languages are HARD.  Problem is they are harder to "get right" then they 
look. "its just code" right? but no.  Envisioning the consequences of 
changing languages ... even "simple"
 scripting languages like bsh takes real dedication and immersion.  Its 
really tough work even for the best of us.  Its easy to add features, but 
its amazingly hard to think through all the implications and hold to a 
coheriant vision.  Even Strustrup thinks so and he's about as smart as they 
get. I found his commentary on how they choose enhancements to C++ in 
C++2008 very enlightening ... in particular his frustration at how difficult 
it is to do a language well, and how few people understood that.
Its humbling.


-David

-----------------------------------------------------------
David A. Lee
[email protected]
http://www.calldei.com


------------------------------------------------------------------------------
SF.Net email is Sponsored by MIX09, March 18-20, 2009 in Las Vegas, Nevada.
The future of the web can't happen without you.  Join us at MIX09 to help
pave the way to the Next Web now. Learn more and register at
http://ad.doubleclick.net/clk;208669438;13503038;i?http://2009.visitmix.com/