> I was thinking a bit about the project based environment, and here are few > questions:
I've been thinking about a lot of these same questions. Here's what I've come to: > 1. why is it better than current environment? It's basically the idea of putting all the files that relate to one another in the same folder, but it provides two things you don't get through folders alone: a visual overview of how the contents of a folder relate to one another, and meta data on the folder. For example, by viewing the folder as a desktop, we can annotate that desktop and store those notes as meta data which could be searched on. We can also move files around on that desktop, building relationships through proximity, and perhaps even annotating those relationships (using the desktop as one big spatial folder). This view would (presumably) be updated in real time, providing thumbnails of the project when looking at all projects at once. It uses technologies which already exist (at least os OS X, and perhaps through Luminocity/Gnome soon) -- those technologies are just being combined in a different way. > 2. I am going to be more productive with the project-oriented environment? if yes, then how? If you already use virtual desktops, then Projects provide _meaning_ to the desktop -- a desktop is a space for related items, not all the random contents you happen to put there. Through annotations, you can mark up that space and build a wealth of meta data for searching and referring to that project and it's contained documents. If you already organize your work into folders that represent certain clients, projects, classes or tasks, then this better conforms to the mental model of how you already work. If you don't care where things go, then you just put everything in one Project/desktop and you have the same advantages as the environment you have right now. > 3. what I can do in the project oriented environment that i can not do > now? You probably can't get a broad overview of how the contents of a folder are interrelated, and be able to visualize those relationships. If you can currently do this (via some clever use of spatial file windows and something like Desktastic < http://www.panic.com/desktastic/ >), can you visualize the contents of many such folders in one quick glance? > 4. can i use existing applications from current environments in the > project oriented environment? if yes, how? Yeah -- just think of them like virtual desktops. Whether we'll have "true" applications in Etoile remains to be seen, but there would be nothing in the Project-based Desktop which would prevent an application from running normally. > 5. i would like to write an article. why do i have to create two things > [project and article document] instead of one? If you're the organized type, you would have already created a folder to store your article in anyways. If you're not organized, you'll just add the article to whatever Project you currently have open. The organized people gain some additional functionality. The disorganized don't lose any of the functionality they have in current environments. > 6. how can I share files between projects? That's still to be determined -- maybe we have a shared shelf concept, maybe we use the ovelapping projects like David suggested earlier. > 7. when to create a copy of a file and when to share it? and how? Hmm... not sure what you're asking. I assume if you wanted a copy of a file then you would have two different file identifiers and both would be edited independently. A "shared" file would be like an alias in OS X, I guess. > 8. how are "small objects" (addresses, bookmarks, post-it notes) managed > across projects? I would suggest some type of space that is persistent or "sticky" where items could be dropped to be accessed across projects. Perhaps an individual window could also be made sticky if need be. This needs to be addressed more. > 9. should we have something like 'default' project? if yes, how it should > be handled? What do you mean? A project that all others would be based on, or a project where miscellaneous stuff goes? I could see either or both... but they'd both be user configured -- if you wanted a miscellaneous project, just make a project and put all your miscellaneous stuff there. If you want all of your projects to start with a certain configuration, make a project "template", then make a copy of it and use that as your new project. > The question #4 is very important. Many projects with great ideas are not > widely used because they can not cooperate with what already exists. Users are > not going to leave their preferred tools. They will if there is a demonstratable advantage to new tools, but that's beside the point. While Etoile will likely have either brand new applications which mimic functionality found in other apps or will take existing applications and reconfigure them so they conform to the Etoile interface, any app that can currently run in a GNUstep environment should be able to run in a Project-based Desktop. Perhaps you're confusing this idea with the whole issue of whether we will be using "true" applications or a component/palette based architecture? That discussion still needs to be decided, but the outcome of that does not in any way impact this method of visualizing the desktop. > Perhaps more questions comming soon... Good :) We need to really think through this and make sure it would be a good solution. J.
