[Usability] Re: making the conceptual model more concrete



Daniel Borgmann writes:
And when I listen to radio? Later you say that with documents, you mean
objects rather, so in this case I would see "the player" as the object
to work with (make it play funky tunes), not the music itself, which can
be everything from a document (audio file) to a CD or radio station. Am
I wrong? So if the actual media player is the object, would I go New ->
Funky Media Player to get it?

No, you're still looking at this wrong. You don't create a new media player, there is no need to have the same interface to play iradio as the one to play mp3's. In rb you have this idea of sources (which seth and I both aren't all that crazy about). Instead in the object world you would have a folder or search folder that contains links to all you iradio stations. Perhaps this folder would have a special view that is presented similar to rb' iradio player. The New Menu would probably be something like New->Music-> Internet Radio Station from which you could create a link to a new station. Also for container views (like the iradio folder) you might have file->new radio station. I'm just conjecturing here to give you a view of what it might look like, not what it has to look like though.
A game could even fit it into this model nicely. Assuming you have a "New" menu it could have an option for Game and popup a dialog allowing you to choose which game (or possibly a submenu). You could than save that game to a folder, and the next time you want to continue the game you simply open the existing game from the folder you have saved it to. Alternatively you could start a new game from the New menu of course.

Ok, that makes sense. In a very twisted way though. ;)
I wonder if it's really easier to find saved games on your desktop than
to start the game of your choice and then load the state of your choice.

Well having a folder (or search folder) labeled games where you save your existing games creates an interface that puts the user is far more control. If you want to delete an existing game you simply drag it to the trash for instance.

> I believe that this would resemble reality much more closely, both "real
> life" reality (we both work with documents and use devices in our lifes)
> and computing reality (what people practically use their computers for).

Objects sound good to me. Because it would be pretty close to what I'm
thinking. If the calender can be an object, why not the media player
(radio?) or web browser? I find this a much more natural way to think

because the calendar is the object you are manipulting. You don't manipulate a browser[1] or a media player, you manipulate web pages and mp3s/oggs (which you just happen to be presented as a collection instead of individual files because it makes manipulation easier).
The next question would be, how would those "non-document objects" be
handled? Should they constantly run and stay in memory or only if you
click them? Should they be represented by files while they are not
running? How could this all work and if they are permanent objects on my
desktop, how could I "put them aside" when I don't need them at the
moment and how would my computer deal with all the stuff on my desktop
when I can't "quit" them to free the memory? I really like the general
idea though.

This is where the magic of processes, swap space etc. all come into play. Notice that it is impossible to "quit" epiphany to free memory. You can only close individual windows, and it just happens that when you close all the browser windows epiphany is removed from memory (but this is not something the user need be aware of at all). In fact a user should never have to be concerned with quiting an application to free memory or other system resources. Try explaining processes, threads and memory usage to ordinary users, they'll look at you like you're a genius and have absolutely no clue what you are talking about :) The operating system is responsible for making sure that memory is properly allocated, and that underused processes are removed from memory and swapped to disk etc. Users shouldn't have to worry about this at all. dave [1] Although I do admit the web does create some interesting problems here and I've been thinking alot about how to solve them.




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]