commons-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mark R. Diggory" <mdigg...@latte.harvard.edu>
Subject Re: [all] Math needs a "user" & a "dev" email list.
Date Sun, 15 Aug 2004 14:33:31 GMT


Phil Steitz wrote:

> Kim van der Linde wrote:
> 
>> Hi Mark,
>>
>> I agree with you. And I have played advocate of the
>> devil here, as I noticed from the beginning on that
>> this is a hard-core developers environment. But in
>> daily life, I work in an environment were people are
>> the real users of the products that are developed
>> here. They are not programmers by nature, they are not
>> involved in area's of science that includes mandatory
>> programming skills like astronomy. 
> 
> 
> If they are using commons components, they are by definition java 
> developers.

Thats probably the most limiting definition of a java developer that 
I've ever heard. If you read a law book, does that make you a Lawyer? If 
you boil water on a stove, does that make you a chef?

> 
> For the people
> 
>> around me, programming is a TOOL, a tool to do comples
>> mathematical stuff that they otherwise could not do.
>> When it becomes to complex, they come to me. When I
>> put people like them on a track of an e-mail group
>> like this, they quit within days. And that is not
>> because they are incapable to understand it.
>>
>> Just to highlight this, and example that I encountered
>> myself. Untill now, I have done a lot of programming
>> of statistics, and the like, and I do statistics all
>> the time. But untill now, I never encountered the idea
>> of 'moments'. Maybe wierd, but still true.
> 
> 
> Moments are used in [math] mainly as a tool to compute more commonly 
> used statistics.  In fact, our "moments" are not really moments in the 
> normal definition (as explained in the javadoc).
> 

The method applied is based on the following paper:
Spicer, C. C. 1972. Algorithm AS 52: Calculation of power sums of 
deviations about the mean. Applied Statistics, 21: 226–227.

It is also used in SPSS:
http://www.uni-giessen.de/hrz/software/spss/algorithms/descriptives.pdf

>> I neither
>> had it at highschool, nor at the university at any
>> time. Maybe it is a wierd thing related to the Dutch
>> school system. But I think it is exemplatory for how a
>> lot of open source projects fail to recognize that the
>> main user group are not the hard-core developers they
>> are themselves. (And I am much more a hard-core
>> programmer than a lot of collegues that might benefit
>> from a package like commons-math(s)). 
> 
> 
> Well, once again, the users of [math] and any other Jakarta Commons 
> components must be Java developers. It is our job, as API designers and 
> documentation developers to make it as easy as possible for them; but we 
> are not developing end user products here.  Any specific comments that 
> you have on the useability of the [math] API or the quality of the 
> javadoc and User Guide would be much appreciated.
> 

Perhaps this is the strongest reason to establish a Math TLP independent 
of Commons Math, to provide room for those users who may not necessarily 
be java developers per se and to provide room for those math projects 
which are not necessarily commons worthy.

I do not suspect there is enough support to continue to promote the idea 
of separate users lists in the commons. I will drop the subject for the 
time being.

-Mark

-- 
Mark Diggory
Software Developer
Harvard MIT Data Center
http://www.hmdc.harvard.edu

---------------------------------------------------------------------
To unsubscribe, e-mail: commons-dev-unsubscribe@jakarta.apache.org
For additional commands, e-mail: commons-dev-help@jakarta.apache.org


Mime
View raw message