[uf-discuss] Scraping or parsing?

[Mike Schinkel]

Karl Dubost wrote:
> > I'll give you those, but there is something
> > fundamentally different about them, i.e. they are for
> > visual presentation not logic and data encoding. And
> > there is SVG. Still, I have to ponder why tools have
> > worked there but not elsewhere.  It could be simply
> > because their level of complexity in text would be far
> > beyond what a human could comprehend.
> > 
> and *pdf* (given in the list) I could have added, vcard,
> vcalendar, vectorial illustrator. All of those, I do NOT
> want edit by hands, even if I had the possibility ;)

It's not about wanting to or not wanting to, it's about being able to. And
just because some people want to use tools not all want to, so being able to
is very important.

vcard is easy to edit by hand, if need be. Same with vcalendar. What's more,
both of those are generallyfgenerator or exported as opposed to authored.

Also, being able to comprehend when reading and also being able to author by
hand makes it is much easier for people to write code that generates the
format. The easier that is, the more conforming implementations there will
be.

So I come back to my central thesis; design for humans first (even if those
humans are programmers), build tools second.

> > But wouldn't you agree, people tend to use the promise
> > of a tool as a crutch when they should instead strive
> > to make things in the raw grokable by humans first?
> > 
> That is a different issue :)  Human is too broad to be
> meaningful.

Is it a different issue? At least not closely interlinked?

> Human is too broad to be meaningful.

I think Wikipedia does a pretty good job [1]. ;-)

> 	The goal is really to make a technology which is easy
> to use depending on the 	ecosystem.
> then using the argument that: 	1. complexity of the
> technology is NOT important because there are/ will be
> tools. 	2. simplicity of the technology is a MUST because
> of "hand authoring". are both flawed, IMHO.

But you are twisting my words. I said: 

	B.) Don't even think about tools until it can be done 
	and easily understood by a human. Only then should tools 
	be created.
	
So I didn't say 'a MUST because of "hand authoring"'; I said it needs to be
understandable by humans. :)  Of course I did imply handauthoring be
accessible, and stand behind that. But unless you are just picking nits, I
don't think that is flawed.

> I'm really happy penballs exist even if I could use ink
> with a feather. I'm really happy to have light
> measurement on my camera, even if I could use my own
> lightmeter (which I do on a 6x6) 

Different domains. Apples and oranges. I'm talking in the domain where
instructions for computer processing are given.

> I'm really I have not to teach HTML to 
> my parents, and just give them a wysiwyg
> editor ;)

Frankly, I'd rather teach a man to fish instead of having to fish for them
all the time. I *did* teach my dad HTML so he'd quit asking me why the damn
WYSIWYG editor kept screwing up his preferred layout and wouldn't let him
fix it. But maybe that's just the former programming instructor in me...

And 15 years later, I've yet to see an HTML editor that will allow me the
same level of control of *output* (I'm not even talking control of markup)
that I can get with a simple notepad (Even MS Word is infuriating in that
respect albeit only someone ignorant of its excesses would use it to
generate HTML for web publishing.) Yes, I'd rather have the WYSIWYG, but
only if it comes w/o the frustrating limitations. Which is why I still do
most of my editing on Notepad (well, Notepad2 that is.) 

>  but yes I think we agree. :)

You do enjoy being pedantic, don't you? ;-)
 
-- 
-Mike Schinkel
http://www.mikeschinkel.com/blogs/
http://www.welldesignedurls.org
http://atlanta-web.org - http://t.oolicio.us
"It never ceases to amaze how many people will proactively debate away
attempts to improve the web..."


[1] http://en.wikipedia.org/wiki/Human