> I know etoile probably could not deliver everything needed now to > optimize these workflows, especially when most software are not > based on > GNUStep for users to integrate. But if the project can present the > public an "image", that how things gonna be, it would bring a lot of > interest from people to the project, creating community that make this > happen. Thus if we have clear ideas how we gonna do, how the > difference > users would feel if they start to use etoile (with example scenario > like > I gave above), then we gain support. How do you think? > > And can you paint a picture based on some scenario how Etoile > evolves to be?
As you mentioned, Etoile, at this point, definitely doesn't have everything needed to help people like the above users. We have infrastructure in place to start to allow for these things, but building the actual services, as well as the remaining infrastructure, will take some time. To explain the ideas a bit more, here are some of my thoughts on what Etoile will eventually look like. SERVICES --------------------------------- One way to think about the services is that developers will no longer have to build massive, monolithic applications. What the Unix philosophy advocates in building single command-line applications that do one thing well and allow information to be piped through them, Etoile hopes to do for the desktop. This is a really good model for open source development as well, in that developers don't have to get up to speed on a full code base in order to contribute to an application. They can just write what amount to "plug-ins", specifying what types of documents can be input and output, and then dealing with only the specifics of what their service does to the data in between, whether that's transforming it or displaying it in some fashion. The advantages for users are also many. Users aren't tied to a full application -- they can pick and choose the services they need, and can easily swap in better ones as they are created. In many ways, this is similar to the model used on the iPhone for its applications, particularly for things like Twitter where you have a standard API and all the apps just follow that model. In this way, the user can build a workflow around the services that they like the best from a competing arena of similar services (just as iPhone users can pick the Twitter app they like best from a whole host of similar apps). Users can have a single service for a single purpose across all types of documents. This is contrary to the current software world, where there is an excessive amount of redundancy. For example, today, many traditional applications end up recreating the "Save as..." dialog box or the Font picker or a menu of Drawing tools. In Etoile, a user will have one such service, the one that they consider the best for their particular needs, and they will have to learn only that particular service's functionality. As an additional user benefit, users will save memory by not having to hold a huge program in memory all the time. From my experience, I know that I only use a very small percentage of all the features in an app like Illustrator, Photoshop or Word, and so opening up only the services I was actually using would be a big benefit, and would probably make my machine feel much more responsive (or, alternatively, would allow Etoile to run on more limited hardware). PROJECTS --------------------------------- Outside of services, we have the concept of Projects which will assist with workflows. To use your scenario of the web designer (which I personally am, so I am intimately familiar with the process), if I wanted to update a website using the traditional desktop model, I would have to open up a variety of applications and documents: a browser (or two or three), a local server admin panel, an FTP client, a text editor, an image editor, a database client, maybe a terminal, etc, etc. And I would have to do this every time I need to make edits to the site I was working on. With the Etoile model, I would open up the documents associated with those functions just once and keep them running in the Project I've associated with the site. Then, every time I need to make a change to the site, I just reopen the Project and all my apps and documents are there exactly where I left them. Think of a Project like a combination of a folder, a virtual desktop and a virtual machine image and you have the general idea. This is a major source of the power behind the Etoile model, and will most definitely save users time in their work. OBJECT BROWSER --------------------------------- Etoile will also feature a "file" browser/manager that itself will be pluggable with new services. The documents listed in such a browser will be the actual live documents, not just an iconic representation. Because of this, any browser can instantly become a music/movie playlist and media player, a photo manager, an FTP/SCM client, a source code project browser, and more. Additionally, because the documents are "live", users can edit them in place, merely by clicking on them to activate them then scaling them to the appropriate size (see http://jesseross.com/clients/etoile/ui/flash/200902/ for an example of scaling and activating). SHELF --------------------------------- Copy and paste will no longer be a single item, destructive process. Objects can be "picked up" (copied) from anywhere and "dropped" (pasted) into a Shelf, which allows them to be stored for easy reuse in other documents. This aids in the workflow by keeping frequently used objects (like an avatar icon or snippets of text/code) in a single quickly accessible location. It may also be possible to assign key commands to items in the shelf to have them dropped somewhere when the key command is triggered. I am sure there are many more examples of how a user's workflow is eased by using Etoile. I'll let some of the other devs pipe in with their ideas if they feel the need or want to elaborate on any of my thoughts. I hope that helps -- let me know if you have any other questions. J. _______________________________________________ Etoile-discuss mailing list [email protected] https://mail.gna.org/listinfo/etoile-discuss
