[clue-tech] KDE4 Anyone?

Jed S. Baer cluemail at jbaer.cotse.net
Wed Jul 30 22:20:55 MDT 2008


I haven't tried it yet, as I'm completely happy with 3.5, and some of
what I've been reading doesn't have me exactly thrilled. Of course, I
realize that I shouldn't judge KDE4 based on SJVN's bitching, but when I
follow links from his blog posts over to places such as:

  http://nepomuk.semanticdesktop.org/

I cringe a bit.

Maybe I'm just behind the curve. I haven't "bought in" to the whole
"semantic web" hype, though I do find the use of things such as keywords
and tagging to be useful. But that sort of stuff isn't exactly bleeding
edge, and knowledge taxonomies aren't anything new. Yeah, I know (or
suspect?) that what's supposed to make this all so powerful is using
whatever the latest web-based P2P, autodiscovery, SOAPY, RESTful,
muti-linked, ultra-xreffed whatever tools and stuff to tie related things
together in some way that theoretically makes following chains of
semantically related data easy. Except that with over a trillion pages on
the WWW, you just get a whole pile of stuff you aren't interested in that
people tagged with terms that they think apply. And I'm sure that
knowledge taxonomies are useful when applied within a controlled scope.

I just don't see the benefit for my window manager.

Granted, I'm not one of those people who has a root window full of icons
representing directories, or links to stuff I use a lot. A lot of the
time, I launch apps from the command line. I usually have 4 or more
terminal sessions going at once, and I use the command line for a lot of
the tasks that more GUI-minded folks would do in Konqueror, Dolphin,
Nautilus, or whatever.

But I do recognize the utility of such applications, and do find them
useful on some occasions. But do they really need to recapitulate
phylogeny in order to do their jobs? (Sorry, dumb reference there to
recapitulation theory there, which I couldn't resist.)

Don't get me wrong. If people want to write software that tries to do all
this sort of stuff, hey, it's their time and they can do what they like.
And nobody's forcing me to use it either. It just seems like overkill to
me. I can fire up my various favorite applications and edit the files
they operate upon just fine without needing an ontological / taxonomic
unified semantic pile of managed knowledge. If, at the command line, I
type 'gimp filename', The GIMP will either open the file successfully, or
not.

And is anyone else kinda tired of icons (or other representations) that
look like blobs, or are in some other way borrowing from Apple's Aqua
theme?

Anyways, just curious what everyone else thinks about all this stuff. I'm
trying to decide what the best way is for me to have a look at it in
actual use. I fear my old computer will be too slow for it, but maybe
running it in a VM would work. I just don't want to mess up my main
machine with it (and maybe it's possible to do a clean uninstall of it
after trying it out, but I really like my existing environment, and I'm
loathe to do anything that might make my tweaks go away).

jed


More information about the clue-tech mailing list