[jdom-interest] Stirring up Trouble
Jools Enticknap
jools at jools.org
Wed Jul 19 07:41:14 PDT 2000
>
> So what does that mean? That I am finding that intuitive is a tough
> word. It means "does what the developer expects, as long as the
> developer's expectations are correct." Sort of tricky, huh? In light of
> that, I find that JDOM should /not/ be intuitive in the case where the
> developer is wrong, as that could encourage really bad concepts and
> practices.
Intuative is such a subjective word, and I would not want to describe many
things as such.
<point-in-case>
I drive a Saab 900, and the ignition keys are in the center console
between the driver and the passenger. Most people who get in my car say
"Thats a weird place to have the ignition".
But when you consider that Saab make aircraft too, and light aircraft
often have the ignition key in the center console between the pilot and
co-pilot next to the throttle.
So for a light aircraft pilot it would seem 'intuative' to find the
ignition in the center console.
<point-in-case>
I would prefer to describe jdom as consistant in it's approach, if
nothing else.
>
> So, here's the crux of (at least) this email:
>
> getContent(boolean preserveWhitespace);
>
> is perfectly intuitive.
>
> But what does
>
> getContent()
>
> do? I'll be the first to admit that I argued and argued that this should
> not return whitespace, because that wasn't intuitive to the user. But if
> the user things XML content ignores whitespace around textual content,
> they would be WRONG. And I would be WRONG to encourage that belief. So
> I'm now advocating that getContent() simply call getContent(true) and
> return text with whitespace around it. I'm so convinced I'm willing to
> make the /very/ painful changes in my book and defend the decision to
> roughly 25000 readers (already!). So let the talks begin, I'm wanting
> this resolved ASAP.
>
When this particular thorny issue was originally discussed I proffered the
thought that a user, experienced or otherise __would__ want the
whitespace.
My reasoning was based on the fact that if you do nothing then you can't
be wrong, but if you _do_ change the original data in the default case
then you could only be right 50% of the time.
And I still feel the same.
--Jools
More information about the jdom-interest
mailing list