Report customization nightmare

[Mike Leone]

On 6/29/2010 3:41 PM, Derek Atkins had this to say:
> Mike Leone<turgon at mike-leone.com>  writes:
>
>>> The problem with scheme is of course unfamiliarity working with
>>> the"functional" languages as most of us learned "procedural" languages
>>> first and it's a different way of thinking. Not that some of our more
>>> common languages can't be used both ways. While learning C I was
>>> challenged to rewrite one of my simple programs so that it would be pure
>>> "function evaluation" instead of "procedure" and yes it can be done.
>>
>> The most important point is that requiring a "normal" end-user to
>> learn an arcane programming language in order to get a customized
>> report is really user-hostile, in my opinion. No, I don't have an
>> alternate solution, but I do know that any of non-geek users who I
>> might recommend GnuCash to, won't know how, nor will bother to learn,
>> programming for just one application.
>
> Opinions are in high supply around here.  Solutions, not so much.  We
> have what we have, for better or for worse.  Until someone comes up with
> a new solution, or offers to provide the effort to create a new
> solution, we're kinda stuck with what we have.  We ARE making
> incremental improvements (e-guile, webkit/css) but a full-on
> report-system replacement is a large piece of work.  Are YOU going to do
> it?

Not being a programmer ... no, I'm not. :-) As I said, I don't have a 
better solution. But that does not invalidate my (and just about every 
one else's, apparently) claim that what we have is not all that useful, 
for a (large?) segment of current users.

> The point of GnuCash is to provide an 85% solution..  It's not going to
> supply 100% of everything that 100% of the users need.  Indeed, no
> application can.  However it can certainly supply most of the needs to
> most of the users.

I wouldn't think that using guile to make custom reports is anywhere 
near an 85% solution. No offense.

> In terms of reports, I would ask why all these so many users want to
> write custom reports?  What's wrong with the existing reports that
> already exist?

Since many (most?) want to change them, then they must not be meeting 
the needs of many (most?) users. And without a solution that requires 
(possibly non-technical) users to learn a programming language to pull 
out something like just a specific subset of accounts, you won't really 
grow the user base beyond those technical types of users, who don't mind 
learning a programming language in order to run a summary report the way 
they want (that's just an example).

Which may be fine, if that's where you (a generic you, not you 
personally) want GnuCash to stay.

I will say that I could learn to program guile; I've programmed in other 
languages before. I won't, because I don't want to go through that much 
trouble for what is - to me - a non-work-related application. I don't 
need the info that much, to go through that much trouble on my own time. 
And so I don't use GnuCash to it's fullest potential, because it's too 
much effort for me to do so, for an application that I use for personal 
reasons.

Just my opinion.