PLEASE someone make a Notational Velocity equivalent for GTasks. I would pay for that.
Which is probably why I love TaskPaper for GTD. Plain text but with smart data-detection.
31–40 of 41 posts
PLEASE someone make a Notational Velocity equivalent for GTasks. I would pay for that.
Which is probably why I love TaskPaper for GTD. Plain text but with smart data-detection.
I've created a new list and added 4 tasks a, b, c and d, and later added a2 after a.
Here's what they are doing:
Position:
a 01111111111111111111111111111111
b 10111111111111111111111111111111
c 11011111111111111111111111111111
d 11100111111111111111111111111111
a2 10011111111111111111111111111111
So it seems a clever way of allocating the position numbers without having to change them all everytime a task is moved, but I'd love to read more about it.They're using JSON to describe sets of possible JSON instances, using conventions of "..." to represent other items in a list; and italics to represent variables (though not always: example text such as "My Task" is used for some variables).
Perhaps a shell-style convention, like $taskListID and ${My Task} would be clearer? It's going outside the JSON syntax, but so is "..." - and obviously extra-linguistic constructs help alert the reader that it isn't part of the instance syntax (like italics).
I've looked at some "JSON schema" proposals, but they take the same meta-approach as XML Schema, which somehow seems even worse applied to JSON.
Regarding the actual format, they the kind attribute (kind="tasks#taskList") like a class. Since it defines the type of the message, maybe it would be clearer if it were distinguished in some way (like XML tags: - invalid, I know).
disclaimer: I don't mean to rag on these guys - their focus is Google Tasks, not a schema for JSON. It's just that it's very well documented, and so it's a clear example of using JSON as a schema. This issue will become more important over time.
Anyone knows what's the name of the algorithm / strategy that they use to allocate tasks position numbers? I've created a new list and added 4 tasks a, b, c and d, and later added a2 after a. Here's what they are doing: Position: a 01111111111111111111111111111111 b 10111111111111111111111111111111 c 11011111111111111111111111111111 d 11100111111111111111111111111111 a2 10011111111111111111111111111111 So it seems a…
http://en.wikipedia.org/wiki/Self-balancing_binary_search_tr...
Anyone knows what's the name of the algorithm / strategy that they use to allocate tasks position numbers? I've created a new list and added 4 tasks a, b, c and d, and later added a2 after a. Here's what they are doing: Position: a 01111111111111111111111111111111 b 10111111111111111111111111111111 c 11011111111111111111111111111111 d 11100111111111111111111111111111 a2 10011111111111111111111111111111 So it seems a…
But why number them at all - why not just use the order in the list as the ordering? I haven't studied their API, so there may be a good reason.
Anyone knows what's the name of the algorithm / strategy that they use to allocate tasks position numbers? I've created a new list and added 4 tasks a, b, c and d, and later added a2 after a. Here's what they are doing: Position: a 01111111111111111111111111111111 b 10111111111111111111111111111111 c 11011111111111111111111111111111 d 11100111111111111111111111111111 a2 10011111111111111111111111111111 So it seems a…
That is clever, it's like a binary version of the old BASIC convention of numbering lines 10, 20, 30 etc, to leave space in case you want to add lines in between. But why number them at all - why not just use the order in the list as the ordering? I haven't studied their API, so there may be a good reason.
This is not specific to Google tasks. If you tried to display a list of items that your users could reorder, you'd run into the same problem.
Earlier quoted context omitted.
That is clever, it's like a binary version of the old BASIC convention of numbering lines 10, 20, 30 etc, to leave space in case you want to add lines in between. But why number them at all - why not just use the order in the list as the ordering? I haven't studied their API, so there may be a good reason.
Because the 'order in the list' usually is the creation order of the items. If you want to change the tasks order, you need to save the position numbers in the database. This is not specific to Google tasks. If you tried to display a list of items that your users could reorder, you'd run into the same problem.
But I guess for RDB (and other DB?), the "physical" order isn't meant to be used as a "logical" order, and it might not even be reliable - so you need some other ordering mechanism, like position numbers. Also, not good if anything in the system depends on that physical ordering.
\not a DB programmer
Why are so many people using Google tasks ? Give alternatives like Workflowy a chance? Evernote works much better than GTasks even if it's not designed for it.
It is simple and minimalist. Keyboard shortcut in Gmail to create a task out of an email. Tasks with dates show up on Google Calendar.
Earlier quoted context omitted.
The general problem of sync is just too large for a comment, a blog post, or even a research paper. One who wants to pursue this route should buckle up and start here: http://www.ysaito.com/survey.pdf If you tell me your specific problem I might be able to help. Shoot me an email via support page on geetasks.com If your problem is specific to Google Tasks API I will not help you. Sorry, business is business. :)
Great response, and I think you're right. But I wonder... lets say you show how to solve some problems with Google Tasks, you get a higher page rank, and people with a problem with Google Tasks arrive at your page... where you incidentally note problems that your solution solves better than Google Tasks (which will be true, since they are mass market and general, and you can be happy with a smaller, more specific nic…