[ILUG] Inefficiency of graphical file browsers

[Justin]

Yeah I've been thinking about this too. Moving files in windows is a 
pita without using cygwin or  WSH.
The file windows often freeze if you have more than a few 10,000 file 
in a directory.

And try using the gui to zip up all files of type x into and archive 
called themonththeyweremodified_in+mm_dd_yyyy.zip format .... etc..



I do think something like OsX tigers automator is kinda of cool.
http://www.apple.com/macosx/tiger/automator.htl

I don't think bash and chaining commands really has this licked either. 
It takes a lot of knowledge and experience to know how to chain  those 
commands together and/or even which commands you need. And then there 
are the things like too many commands etc...



Justin

On Sep 28, 2004, at 5:49 PM, P at draigBrady.com wrote:

> Tanney, Austin wrote:
> > Thanks everyone.. this one seems to be doing it. Taking its time
> > but its a hell of a lot of data to move (about 10 gb). The funny
> > thing is that its data thats stored on a windoze server but there
> > is basically no percievable way to actually do this with windows
> > so I had to mount the share on the Linux box and do it through this!
>
> I was just thinking about this last week.
>
> Graphical file browsers are not very useful
> for working on a tree IMHO, which is very common.
> Even for files in one directory it's hard to perform
> a common operation on each file.
>
> Lately this has become much more of a problem due
> to web sites and picture archives for e.g.
> In essence since we're digitizing stuff in increasing
> volumes, the current interfaces to that data are showing
> their limitations. There is a lot of fluff lately about
> digital workflows which usually boil down to a mess of
> proprietry apps that are limited in their scope:
> http://www.robgalbraith.com/bins/multi_page.asp?cid=7-6453-6821
>
> Graphical browsers are useful for navigating to a certain file
> in a tree or selecting disjoint sets of files. The former isn't
> scalable really as one needs to search the tree as it gets large.
> The later is usually only required for (re)filing in the
> hierarchy. Note, for filing I find a two pane view of source
> and destination directory is optimal and mc is my util
> of choice for this. One could have a "two pane" view by having
> multiple instances of the graphical browser and using drag and drop.
> However I've seen very few people use them like this.
>
> In summary, I think interacting with data through
> an "explorer" type interface is extremely inefficient.
> Also unix (Linux/OS X/...) should become more popular due to its
> elegance in dealing with (large amounts) of data.
>
> -- 
> Pádraig Brady - http://www.pixelbeat.org
> --- Following generated by rotagator ---
>
> Users
>
> su            #change to root user
> su -l         #same but get correct root $PATH etc. (login shell)
> su -l user    #change to user
> su -c "cmd"   #just run cmd as root and exit
> sudo cmd      #just run cmd as root without password and exit
> visudo        #setup sudo (as root). `man sudoers` for more info
> --
> -- 
> Irish Linux Users' Group
> http://www.linux.ie/mailman/listinfo/ilug/
>
>