RE: [xml-dev] XML-DEV list - prior art

[sterling <sstouden@thelinks.com>]

I want to make clear that I am not arguing with you, 
but I do think the following might provide some food 
for thought?

I think you might find that "concepts and techniques
used" follow a hierarchy of papers or prior works, 
such that the few most recent papers on a given 
"implimentation of a prior art" would be all that 
is needed to provide the references to footnote 
the prior art used in the various lines of code.  

What would be interesting is to pick a product 
or even a portion of a product and to do the 
prior art research on that product to find good 
references to the prior art that was used to 
produce the current code and to develop a 
system to footnote the embedded prior art used 
in the code to the references. 

I think after the first few big projects are 
footnoted to locally appended references, that 
the appended reference length will become very 
manageable. 

Lets develop a senario! 
Our developer has 200 prior arts appearing in
his 6000 lines of code that need to be footnoted
and indexed to a reference, and those footnotes
would, of course, appear through out the 6000 
lines of code maybe something like the following:

    breakdown of prior arts   unique prior art
    used in the 6000 lines    occurences of
     of code                  in need of a 
     n times.                 appended ref
         
 A.  19 appear only once           19
 b. 120 appear twice               60
 c.  21 appear in triplicate        7
 d.  40 appear 4x                  10

   Max appended references needed =     96
  ( for current paper to Footnote to)


This means the number of references needed 
to footnote and index 200 prior arts used 
in the body of the 6000 lines of code would 
not be greater than 96, but it does not mean 
that 96 separate appended references would 
be needed to footnote the 6000 lines of code.

Unique              only 9 REFERENCES NEEDED 
references          to FN all prior art
needed             A, B, C, D, E, F, G, H, I
 19(1)=19          -, 3, 1, 6, 2, 1, 1, 1, 5
120(2)=60         40, 1, 2, 2, 2, -, -, 3, 1
 21(3)= 7          0, 2, 1, 1, -, -, 3, -, -
 40(4)= 4          1, -, 2, -, -, -, 1, -, -
                      
After the developer looks at his references 
he or she might discover that references 
already cite many of his prior art occurences, 
so in the above senario only 9 references might 
actually be needed to footnote all of the prior
art used in the 6000 line program. 

The unique number of different prior arts used
in a project many not be that great, since many 
of the arts used, whether novel to the current 
project or adopted from the prior art will be 
used over and over and over.  

sterling

On Fri, 29 Sep 2006, Michael Kay wrote:

> > In short every software project would carry a cumulative 
> > history of known prior art.  
> 
> I'm wondering what such a list would look like for Saxon. If anyone had time
> to produce it, I'm sure it would be bigger than the code itself.
> 
> Michael Kay
> http://www.saxonica.com/
> 
> 
> _______________________________________________________________________
> 
> XML-DEV is a publicly archived, unmoderated list hosted by OASIS
> to support XML implementation and development. To minimize
> spam in the archives, you must subscribe before posting.
> 
> [Un]Subscribe/change address: http://www.oasis-open.org/mlmanage/
> Or unsubscribe: xml-dev-unsubscribe@lists.xml.org
> subscribe: xml-dev-subscribe@lists.xml.org
> List archive: http://lists.xml.org/archives/xml-dev/
> List Guidelines: http://www.oasis-open.org/maillists/guidelines.php
>