gtg-contributors team mailing list archive

[Bertrand Rousseau <bertrand.rousseau@xxxxxxxxx>]

Hmm, so we're doing etymology too now? ;-)

Regarding subtasks. How we define subtask has been described by ploum
in one of his mail, posted about 2 weeks ago on gtg-contributors [1]:

"In essence, a subtask B of A is a task that you consider as a
prerequisite before working on A."

So, it's really important for us that a subtask must considered as a
prerequisite of its parent task. It thus corresponds to a "break down"
of the parent task in more detail (the aim being to refine a parent
task in subtasks until those can be performed directly), and not an
extension.

Now, the trouble here comes from the fact that this concept is not
clearly explained or suggested in GTG. Which leads to what ploum
defined as "abuse" (cf. [1]). IMHO, "abuse" term is a bit judgmental,
but the main point is: users don't seem to understand this concept as
us, which leads to misunderstanding with the behaviour of other
functionalities.

This is the case with the workview, which is often misunderstood.
Indeed, if subtasks are prerequisite tasks for their parent task, you
can't, by definition, perform the parent task before achieving all its
subtasks (they're exclusive). This is why the workview, which only
displays the "tasks you can actually do", hides all parent tasks (and
why those tasks disappear when you add subtasks while in the workview
mode).

Well, this is the rationale behind all this. I think it's sound.
However, I agree that how we implemented it now lead to confusions, so
it should be rethought a bit.

Bertrand

[1] https://lists.launchpad.net/gtg-contributors/msg00769.html

2012/4/3 meg ford <meg387@xxxxxxxxx>:
> It's from Latin, yeah. Subtasks are inherently subordinate, so
> changing them does not change meaning/date of the main task.
>
> Removing today might help, but it won't solve your issue, most likely.
> However, it's up to you :)
>
> Meg
>
> 2012/4/3 Izidor Matušov <izidor.matusov@xxxxxxxxx>:
>> Am 03.04.2012 16:51, schrieb meg ford:
>>
>>> I said on the wiki that I think it should be evident that it is an xor
>>> choice- either you schedule something for today, or you move it to a
>>> different view.  These should be handles differently, not as two items
>>> on a list of several.  Also, you may have a bug for the subtask case,
>>> but this is the very confusing. Logically a subtask is an extension of
>>> a task, that is just the way that the term is used. It is possible to
>>> schedule an extension of a task to another day while needing to
>>> perform your main task that day.
>>>
>>> Meg
>>
>>
>> I am sorry, but I don't understand. You are right that set start date in
>> Workview mode shouldn't have item "today" because it doesn't make sense.
>> Would removing "today" solve the problem?
>>
>> When you decompose word subtask you get Latin prefix sub- and word task.
>>
>> sub-
>> A prefix that means "underneath or lower" (as in subsoil), "a subordinate or
>> secondary part of something else" (as in subphylum.), or "less than
>> completely" (as in subtropical.)
>>
>> How I understand you that you say a subtask was an extension to main task
>> and could be done after completing the main task. But it doesn't comply with
>> definition above. (and I am confused what do you want to say).



-- 
Bertrand Rousseau