Quote:
Originally Posted by jasong
I believe the "problem" with multiple contexts comes down simply to what the understanding GTD appears to be. Part of that is that GTD is supposed to get you to "think" a certain way about how you approach your tasks, and one way it gets you to do that is to make you think about which context you need to accomplish a very physical action.
|
For a paper system one context / action makes sense. On paper, you would have to duplicate the action in order to put it into multiple contexts.
This problem does not exist on the computer. Putting something into multiple contexts isn't a case of having to duplicate the activity. It's simply selecting all the contexts into which the activity needs to fall.
As far as GTD "getting you to think a certain way about how you approach your tasks", yes -- the "thinking" involved is to identify the context(s). On paper, one context. On the computer, it could be multiple contexts. If I have an activity that falls naturally into more than one context (plenty of examples already provided), then I think in the GTD way and assign them that way making use of the computer to be able to have one activity and all its contexts.
Also, as far as "getting you to think", a big part of GTD is to unload something and also to think about it less. With multiple contexts, processing of activities is much easier because I don't have to think about (or remember) how I artificially need to impose my activities into a one-context scheme.
Most other task and outline programs allow some way to assign multiple values, whether they call it tags, keywords, or multi-valued column. In the case of this program, context would be one area where people may want that capability. If they don't, then they only need to assign one context.
I think some people think of context mode as just another way to sort and view their project activities. If a person wants single contexts / activity, they can have that. On the other hand, with multiple contexts, a person can go into context mode and be sure that any context they look at contains all of the activities that naturally fall there.
I think user mcoad (sorry if I missed your name) has expressed it best.
I think, if this is to be implemented, that this might be the best time to do it -- alpha stage. Changes later would confuse users, and require much more explanation about how to code things (esp. if the user-interface changes a bit for selecting contexts). A change now would make any user interface changes while this is in alpha, and the alpha-testers would be up to speed on how multiple contexts work, and when it goes to 1.0 there would be many alpha testers who would understand and aid in forums how multiple contexts work.
I think it would be important to have a user preference setting for single or multiple contexts, and default it to single. That reduces confusion for the new user.
If single contexts are implemented, after 1.0 there will be users who would want multiple contexts. So, instead of this "problem" going away, it will persist, and discussion of it will persist.
If multiple contexts are implemented, this will still be discussed, but more in terms not of the limitation of the program (single contexts) but in how the program offers the option of powerful multiple contexts _if_ one wants or needs to use that.
Anway, I have to go shovel snow. It sucks to live in Wisconsin during the winter !!! 23 degrees here -- I'm jealous of all you southerners.