[CL] 24707:2007 Corrigendum

[Ed Barkmeyer]

John Sowa wrote:

>>  2. Will the corrigendum be a completely updated copy of
>>     the IS or will it be a collection of changed pages?

Harry responded:

> Unfortunately, no. The corrigendum is a simple list of single  
> paragraphs, each of which looks something like this:
> 
> In 5.1.2, paragraph 2, replace “a dummy argument, a function result,  
> or an automatic array of a
> procedure” by “declared only in a subprogram or interface body”.

The ISO practice is still steeped in the tradition of paper copy.  They 
don't want to schedule another publication run, particularly if they 
have a shelf full of unsold copies from the first printing.  If you have 
already bought a copy of the standard, they will send you the paper 
corrigendum for free and you can patch your copy however you will. 
(John's suggestion matches the practice of IBM, Digital and others in 
updating manuals in that time -- the manuals are presented in notebook 
form, and they ship you replacement *pages*.  But the ISO practice is 
cheaper.)  If ISO goes to a second or third print run for a standard, 
they will often incorporate the Corrigenda-to-date in the later print 
runs (because that is cheaper than reprinting the Corrigenda).

The interesting thing here is that we are asking for public availability 
in the same time frame as the creation of the Corrigendum.  Depending on 
the relative speed of the different divine mills, ITTF may suggest that 
we make a *corrected* version of the PDF the publicly available one.

> There's no rationale, justification, etc. These are all taken care of  
> in the defect report which is handled by the editing group (us!).

Well, we do have the option in the "Introduction" to the Corrigendum to 
summarize and perhaps provide rationale for the changes.  This is 
particularly useful when a conceptually simple correction causes text 
changes in a dozen different places.  And where the correction itself 
requires clarification, a part of the correction itself can be adding a 
Note to the standard.

But, just like any editing committee that resolves ballot comments, the 
committee "progresses" only the changes to the specification, and files 
the issue resolutions, issue by issue, in a separate report that is just 
an SC32 N-numbered document.  That is standard ISO practice.  It 
captures the rationale for whatever changes are made, and it captures 
the rationale for dismissing some defect reports with no change.

-Ed

-- 
Edward J. Barkmeyer                        Email: edbark at nist.gov
National Institute of Standards & Technology
Manufacturing Systems Integration Division
100 Bureau Drive, Stop 8263                Tel: +1 301-975-3528
Gaithersburg, MD 20899-8263                FAX: +1 301-975-4694