How to make a "custom" expense report [was Re: Report customization nightmare]

[Mike or Penny Novack]

>
>True, programmers do. I speak of end users, who (usually) are not
>programmers. In the example discussed, the only way to remove the
>sections of the report that I wish to customize would be for me to
>re-program a copy of the report, using a programming language (guile,
>I presume). And my contention, this entire time and multiple threads,
>is that expecting an *ordinary end user* (read: non-computer
>technically oriented or skilled)  to learn a programming language, to
>remove an account from a report *that you didn't choose to be on the
>report in the first place*, is unrealistic, user antagonist, and
>really should be addressed.
>  
>
Michael, clear misunderstanding.

"......read: non-computer technically oriented or skilled) to learn a 
programming language, to remove an account from a report ..."

Absolutely not! I would not expect the end user to be programming 
anything to be removing a line, etc. from the report. I would be 
expecting them to be using their favorite editor application to do that. 
The desired END RESULT is a report on paper or in some PORTABLE machine 
readable format (where it can be safely stored and read in the future 
whether or not GnuCash exists). Since the "report" (as produced by 
GnuCash is going to end up OUTSIDE of GnuCash in any case it makes no 
difference whether this editing is done before or after leaving GnuCash. 
You are (rightly) complaining that to do so before means learning a 
programming language but wrong to think that the editing requires 
learning a programming language because it doesn't have to be done before.

>And that was just an example. If I had a need for other reports, I
>might find other examples of design and implementation that could use
>improvement.
>  
>
And THAT is the crux of the problem (form the programmer/analysts point 
of view). Why we see this as a MUCH bigger task than you do. There are 
countless possible report modifications that all users collectively 
would want and so what would be needed is a GENERALIZED "report editor". 
A lot of work to do this "reinventing of the wheel" AND since this would 
be a brand new editor, end users would have to learn how to use it.

Michael

PS

>
>  
>
>>There isn't an obvious length of
>>time into the future that the reports would have to be readable and no
>>assurance GnuCash would still be around 5-10 years from now. So I wouldn't
>>be leaving them just stored as GnuCash reports.
>>    
>>
>
>Then don't. But why shouldn't I, if I choose to?
>  
>
THIS is very relevant.  It's relevant to other user questions like "why 
no password protection?". Were I an analyst responsible for this project 
(I'm not, I am indeed an experienced analyst but for GnuCash just a 
user) one of my responsibilities would include keeping the application 
unfriendly to things lazy users might want to do but were extremely 
dangerous to their interests. In other words, follow the rule "don't be 
an enabler for foolish user choices".

You are in effect saying something like "why not make it easy for me use 
GnuCash in such a way as all my financial data could be lost tomorrow? 
It's my choice." Sorry, but any application I designed would make it 
hard for users to JUST store their important data in some proprietary 
format. They would at least have to see a warning message every time, 
one that they could not disable.