[
https://issues.apache.org/jira/browse/MATH-674?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13113515#comment-13113515
]
Gilles commented on MATH-674:
-----------------------------
bq. How about moving it to the header (interface comment) of
UnivariateRealFunction?
+1
bq. those classes should not be final?
I don't see any drawback. Are there obvious advantages?
bq. the singleton pattern should not be used for these very basic functions?
Why?
> Too much documentation in "UnivariateRealFunction.java"?
> --------------------------------------------------------
>
> Key: MATH-674
> URL: https://issues.apache.org/jira/browse/MATH-674
> Project: Commons Math
> Issue Type: Improvement
> Reporter: Gilles
> Priority: Trivial
> Labels: documentation
> Fix For: 3.0
>
>
> I noticed that because they inherit the Javadoc for "UnivariateRealFunction",
> all functions in package "o.a.c.m.analysis.function" have their "value"
> method contain a fairly large amount of documentation that is related to a
> marginal issue (advice about why and when a user would need to implement his
> own exceptions for his own functions) _that does not apply to the function
> being documented_.
> It might be confusing to have that documentation appear there (i.e. in a
> place where none of the functions are defined by the user). It should
> probably be moved elsewhere (user guide).
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira
