[Thunar-dev] Spatial or not-spatial?
Jari Rahkonen
jari.rahkonen at pp1.inet.fi
Fri Mar 4 17:13:11 CET 2005
Benedikt Meurer wrote:
> Jari Rahkonen wrote:
>
>>> Ok, this needs clearance. The analogy is correct, maybe I made a
>>> mistake while translating it to english. The whole point in the
>>> analogy is the ever increasing amount of data. With the WWW people
>>> have already realized that a query-interface is needed (tho, these
>>> interfaces aren't perfect yet, *IMHO*), but with filesystems, people
>>> keep on fighting for their treeviews.
>>>
>> I'm sorry to butt in, but I'm a bit confused. You're right in that
>> the amount of data is increasing in both the web and modern
>> filesystems, but that seems to me to be the only major similarity
>> between the two.
>
>
> Its the only similarity we are interested in. You can compare other
> aspects of WWW and local file systems, but that won't probably help in
> discussion of Thunar.
>
And why is this one aspect the most relevant? That's what I didn't get. This
makes me wonder; what do you people use file managers for? I know I don't
use them to find stuff, but to manage (Move, Copy etc.) and open files (with
relevant software). I'm very sorry if I'm missing the whole point here. I
just think there's more to it than a difference in scale. They don't call
browsers web managers, right?
>> I search the web to find data. I search my file system to locate
>> a certain file or to find certain types of files.
>
>
> I dunno about your files, but usually files contain data and thats
> their whole purpose *IMHO*, so it doesn't matter if data is contained
> in a local file or a webpage, its just the data that you are
> interested in not the container.
>
I understand that the point is to separate the process of file managing
from the underlying architecture to make it more intuitive. Files are just
containers for data, and the data is really the only thing important here.
You're right there. So the filesystem is just a big database just like the
internet, and a database cannot be used efficiently without a query
interface. In the end you shouldn't need to worry about directories, files
and whatnot at all because the data content is all that is important.
But wouldn't this kind of difference in file/data management implicate
large changes in the whole operating system UI? How would a filemanager
that does things differently (intuitive or not) from all the other
software (and I don't mean other file managers) make things easier for the
user?
>> I don't search
>> my filesystem very often because I usually know where I've stored
>> my files, or at least where in my filesystem I should look for
>> them. When it comes to the web this is rarely true.
>
>
> You don't always use google on the WWW either, but sometimes enter an
> URL directly because you know the location.
>
>> To me the internet seems more like a whopping large directory with
>> a myriad of everchanging files than a hierarchical filesystem tree,
>> so I can't help but feel that this analogy of yours seems quite
>> artificial.
>
>
> Lets drop that analogy then, and keep the point that the amount of
> data on the user's file systems is ever increasing.
>
Yes, but keep in mind that the amount of data on a user's filesystem is
never going to be even a fraction of the size of the WWW, and that the
amount of files isn't probably growing nearly as fast as the amount of
data. We've got bigger files but organizing and managing them isn't
really that much harder. I wouldn't really say that software has
completely failed to keep up here, even though there's still need for a
good file manager.
Ok, I think I've done enough damage here... There's probably an actual
point hiding somewhere in all that random rambling, but I'm sure even a
query interface couldn't help you find it. I think I'll just shut up
and be happy with thunar whenever it's ready for release.
> [ SNIP ]
>
> Benedikt
- Jari
More information about the Thunar-dev
mailing list