-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 You're correct I'd not yet read your second posting, busy week with work so catching up again.
You've made it clearer to me now and I think it's a great idea and I wish you luck with it but I don't even know where you're going to start. You've described a few situations where to do something effectively you had to use a Script to chain linux tools together. That's pretty much programming. It's pretty easy on the command line to find all files that end in .jpg and execute the following command on them. That's always been the power of the command line and it's why loads of linux people still use the command line. You want to give that power to the masses. One might be inclined to say traitor ;-) Some of the components could be plugins in the local desktop or some might be out in the internet, all that cloudy stuff people are talking about. You put together a pipeline of different coloured building blocks and have it perform a task you regularly do like all the scripts you keep in your ~/bin folder. I think it's brilliant but at the end of the day your boss is still going to ask you to put the different bits together because he's either too stupid or too lazy to do it the same as shell scripts now ;-) I'm thinking of a solution here but I can't think of anything that would enable the stupid and the lazy. Plumbing is a question of taking standard pipes and standard pipe fittings and putting them together but we'd all still hire a plumber to do the job. I'll move on but keep us informed on how you get on. It's certainly an interesting project. John Arlo White wrote: > In the first email I think I made the mistake of describing what I think > this desktop idea will become in 10 or 20 years: a computer system like > you read about in sci-fi books, that can understand everything and help > you manipulate it intuitively. To start I just want to explore the idea > of organizing your files in different ways and associating things > together through contexts, projects, or custom tags you create. I > really want to integrate email and im because it lets the system follow > your communications and interactions with people. Once you do this, you > can tell your computer: "Show me files that Bob has sent me", or "Show > me pictures of Bob" or "Show me pictures taken by Bob". That's the > first big step. > > More comments inline... > > Arigead wrote: > Arlo White wrote: > >>>> I've been following the Enlightenment project for years, always >>>> impressed by the strength of vision and dedication of the developers. >>>> Every once in a while I take an inventory of the graphical toolkits out >>>> there and am always disappointed by the fact that the EFL is the most >>>> progressive desktop gui system out there and yet hasn't really broken >>>> into the mainstream. All the other GUIs (QT, GTK, Windows) are built >>>> around boring components (boxes, pull-downs, radio, etc.) The concepts >>>> behind these mainstream toolkits are decades old. >>>> >>>> > very interesting email, which I've cut to save me sending it all back at > you. It got me to thinking but I finally decided that it wasn't a good > idea. That's only my opinion and it is based on my own personal > interface with a computer. I'm running a Linux Distribution and probably > at least 50% of the people on this list are as well. Every one of us has > probably got a very different "Setup". We could start a religious war on > any number of subjects, best editor, best web browser, best Distro, best > email client... > >> This desktop I'm imagining would be customizable as well. You could >> plugin different modules to change behavior, adjust preferences, etc. >> And it's going to integrate and work with applications, so if you don't >> use all of its features it doesn't really matter. I'm not sure if you >> read my second email with the GNU tools/shell analogy, which I think >> represents my idea better. This desktop will be exactly like working >> with GNU tools. There are a lot of little tools that you can connect >> together to do useful things. A lot of the time you just use less to >> view files, but when you need a real editor, you fire up vi/emacs (or >> whatever you use). The same thing will happen here. In this system you >> can rotate or add captions to pictures, but when you want to do some >> real editing, you fire of the GIMP. > >> I just think that certain concepts, such as email are integrated in our >> lives so deeply that they should have the same significance on your >> desktop. Email is more of a communication point than a productive >> application. I think being able to associate communications with the >> project and files you're working on will be very powerful. So certain >> applications like email clients may need to be replaced, or at least >> augmented with plugins so they integrate with the desktop. > You're idea is based on the fact that everybody is going to use their > computer in exactly the same way. When I think about that given that > there are a number of people in the world who buy a PC with Windows > pre-installed and they never change a thing, except the wallpaper. Maybe > you're correct. > >> I do have the more typical non-technical user in mind for a lot of >> this. Computers should be much more intuitive and simpler to use. I >> look back at some of my work experience and wonder why computers can't >> just do those things. For example, at one job we needed to batch >> convert a lot of tiffs to pngs, I wrote a little script to do this. But >> really, you should be able to just search for and select the files you >> want to convert, and then be able to convert them to whatever format you >> want. In another situation my boss wanted a listing of some folders in >> the file system, again I wrote a little script to spit out the names to >> a file. But why can't you do this with the computer naturally. It's >> these little tasks that people can't do currently without knowing how to >> program that need to be available to users in a more intuitive visual way. > I'm still thinking of your idea though and trying to imagine it. If I > selected a file then your computer has to know what it is and what that > "is" means. If I select a picture there are any number of things that I > can do with that picture, similarly with a music file. Perhaps I want to > play it, perhaps I want to edit it, perhaps I want to email it to a > friend, perhaps I want to add it to a play list. I select what I want to > do with the file by selecting the Application that's relevant, to what I > want to do. I can "invent" a brand new action to perform on a file by > creating a brand new application. I could write an app that plays a > music file backwards. > > I might be wrong but in your system it's a huge monolithic system if I > select a file I'll get 100's of options to then select. Is there a > fundamental different to right clicking on a music file and getting a > list "Edit, Play, Add to Playlist, Email to Friend, Delete,,," and the > current system of selecting the App I want. It seems to me to be a > difference only in the selection mechanism, and who controls the code. > >> These are good points and I've been thinking about these issues. As you >> add more and more tools that can be used on some file how do you expose >> them without overwhelming the user? I've thought about having >> categories for tools and if there are lots of tools associated with a >> file they get broken up into categories. You could also have a >> "More..." button or just search for a tool by name. It is functionally >> the same as a context-menu, but in the end it will probably look and >> behave very differently. A picture, or in this case, an EDJE prototype >> is probably worth a thousand words...so I'll try to work on that soon. > I personally like the old unix ideal of having a smaller app that does > something well, then having everything and the kitchen sink all built > in together. Sounds like you'd be almost combining the functionality of > the Linux kernel, the X Windows System, All the Apps that have ever been > invented into one build that spits out uImage. You'll need a big build > server ;-) My mom don't use too much functionality that people on this > list would use. I'm afraid a more profound Enlightenment would not > interest her but I'll ask the next time I see her ;-) > >> I think here is where I may have misrepresented my idea. It isn't a >> single code base. It's a set of modular core components that supply the >> main user-interface, and a collection of tool or mini-applications that >> have some standard metadata. When you install a new tool it will be >> added to an internal database where the computer can ask "what tools can >> this file be manipulated with?". It's not replacing the kernel or X, >> it's built on top of libraries that are on top of these layers, just >> like other Linux applications. I agree that a monolithic single >> code-base wouldn't work. > >> The key is that those core tools will have to be very clever. The >> context menu will have to adjust to any number of tools being associated >> with what you clicked on. The desktop will have to adjust to holding >> any number of files. I think this can be done, I guess my first step is >> to start writing some of these core tools to prove that. > Thanks for the email though got me thinking. > > John > -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.9 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iEYEARECAAYFAknqW+YACgkQXlbjSJ5n4BD35wCgk1o8Q7ZxgYpl53wG6F+6U6rx UzAAoKO3JggCdFris0tBfH04Gout1wB4 =PLR/ -----END PGP SIGNATURE----- ------------------------------------------------------------------------------ Stay on top of everything new and different, both inside and around Java (TM) technology - register by April 22, and save $200 on the JavaOne (SM) conference, June 2-5, 2009, San Francisco. 300 plus technical and hands-on sessions. Register today. Use priority code J9JMT32. http://p.sf.net/sfu/p _______________________________________________ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
