Projects with unknown total time / end dates

1

I’m wondering what the best way is to set up this kind of thing. I have a
few projects on which I want to spend x hours or minutes per day until
they’re finished, but I don’t know when they’ll be finished. Guestimating
could lead me to have hours left to work, according to my yellow brick
road, even when the project is complete. Shooting for fewer hours to be
safe will likely just result in me doing fewer hours. Is this a job for
Fine Print + Support when the project is complete? Or is it better served
by trying very hard to keep an eye on my buffer so that I can have a 7-day
buffer and can end the project on the day it’s actually complete and then
just wait for my buffer to run out?

Thoughts?
Thanks!

2

Is there some metric you can track other than time, for which you DO know
in advance what value corresponds to being done? For example, for writing-
or reading-related goals I often track page count instead of time spent.
Page count does not always correlate well with time spent—some pages take
much longer than others to write, or to read—but it helps ensure I am
really spending my time productively. Your comment about “shooting for
fewer hours will just result in doing fewer hours” makes me think this
might be an issue for you too. An ideal metric should actually measure
your progress on a goal (so you know how much you have left to do) and
should incentivize the right sort of behavior. It sounds like time is not
the right metric for your projects.

Of course, page count is a particularly obvious metric; I do realize that
for other sorts of tasks there may not be something as obvious, but it’s
worth thinking about.

-Brent

On Thu, Apr 25, 2013 at 3:41 AM, Essentiae essentiae@gmail.com wrote:

I’m wondering what the best way is to set up this kind of thing. I have a
few projects on which I want to spend x hours or minutes per day until
they’re finished, but I don’t know when they’ll be finished. Guestimating
could lead me to have hours left to work, according to my yellow brick
road, even when the project is complete. Shooting for fewer hours to be
safe will likely just result in me doing fewer hours. Is this a job for
Fine Print + Support when the project is complete? Or is it better served
by trying very hard to keep an eye on my buffer so that I can have a 7-day
buffer and can end the project on the day it’s actually complete and then
just wait for my buffer to run out?

Thoughts?
Thanks!

–
You received this message because you are subscribed to the Google Groups
“Akratics Anonymous” group.
To unsubscribe from this group and stop receiving emails from it, send an
email to akratics+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

3

One fun solution is this: If you get to the end of your project and find
that Beeminder still expects you to do X hours, first adjust Beeminder so
that the goal stops in seven days, then add a new “task” to the project:
“Spend X hours feeling pleased with myself for finishing!” If X is a very
small number, you could just keep the feeling of satisfaction near the top
of your mind for that many hour(s), then add the data point to Beeminder
and it’s all done. If it’s a larger number, and you really don’t want to
“cheat” in Beeminder, then set a calendar reminder once or twice a day to
spend a few/several minutes feeling pleased with yourself.

Or you could add a reward activity to the end of the project: “30 minute
relaxing bath because I’ve earned it”; “3 hours chatting with friends”

Keep in mind, Beeminder is an excellent tool for keeping yourself on track,
but when it’s served its purpose and you legitimately don’t need it any
more for a particular project, you can probably get away with being less
strict with yourself about updating it accurately.

On 25 April 2013 17:41, Essentiae essentiae@gmail.com wrote:

I’m wondering what the best way is to set up this kind of thing. I have a
few projects on which I want to spend x hours or minutes per day until
they’re finished, but I don’t know when they’ll be finished. Guestimating
could lead me to have hours left to work, according to my yellow brick
road, even when the project is complete. Shooting for fewer hours to be
safe will likely just result in me doing fewer hours. Is this a job for
Fine Print + Support when the project is complete? Or is it better served
by trying very hard to keep an eye on my buffer so that I can have a 7-day
buffer and can end the project on the day it’s actually complete and then
just wait for my buffer to run out?

Thoughts?
Thanks!

–
You received this message because you are subscribed to the Google Groups
“Akratics Anonymous” group.
To unsubscribe from this group and stop receiving emails from it, send an
email to akratics+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

4

Brent: I think you’re probably right that with most project, hours wouldn’t
be the right things to measure. For those with a second factor that can be
measured, I add that graph too and just track both. What I’m trying to
track, though, is more of a schedule system than a project goal (I rotate
my projects and want to track that), so hours really are the main focus. At
this point it’s really AIC (ass in chair) that I want to re-enforce (and to
prevent myself from rotating around any projects that I start avoiding).

Alys: Your “Spend X hours feeling pleased with myself for finishing!”
idea made me chuckle out loud, but I think I might go with something that.
Since it would be the last datapoint, it probably wouldn’t matter, and I
could note the actual total in the comment section of that last entry. It
makes me feel a little twitchy to think that I’ll be leaving fake data of a
kind, but I’ll just have to get over that!

On Thursday, April 25, 2013 3:41:16 AM UTC-4, Essentiae wrote:

I’m wondering what the best way is to set up this kind of thing. I have a
few projects on which I want to spend x hours or minutes per day until
they’re finished, but I don’t know when they’ll be finished. Guestimating
could lead me to have hours left to work, according to my yellow brick
road, even when the project is complete. Shooting for fewer hours to be
safe will likely just result in me doing fewer hours. Is this a job for
Fine Print + Support when the project is complete? Or is it better served
by trying very hard to keep an eye on my buffer so that I can have a 7-day
buffer and can end the project on the day it’s actually complete and then
just wait for my buffer to run out?

Thoughts?
Thanks!

5

Hey everyone, love this conversation! We’re actually so skeeved out by
fake data [1] that I might prefer a 3rd solution: invoke the SOS clause for
finishing early. Like literally let it derail and then respond to the legit
check right away that the goal date should be adjusted to be equal to the
derail date, thus undoing the derailment. In theory that kind of thing
invites even more abuse than fake data but in practice it’s quite the
opposite. Abusing the bot is all too easy – abusing us humans is hard.
Y’all seem to be super nice like that. :slight_smile:

But Alys’s solution is actually pretty awesome, and I have a proposal for a
variant that’s maybe the best of all worlds: generalize the project
slightly. Instead of “write this paper” it could be “write this paper
and/or promote it” or “write this paper and/or compile a list of potential
objections and my responses” or “write this paper and/or do related
research for a possible followup” or even just use the disjunction of a
couple different projects – working on either of them counts toward the
goal.

I should add that Brent’s solution is best of all, whenever there’s a way
to make the whole question moot with a better metric. Just that sometimes
there isn’t.

Speaking of making it moot, if you manage to keep 7 days of safety buffer
then you’re immune to the akrasia horizon. Finish the project and set the
goal to end in a week. With more than 7 days safety buffer it’s not an
issue that you stop working immediately; the goal will complete
successfully. (I know, I know, if you had the discipline to keep 7 days of
safety buffer then you wouldn’t need Beeminder. It’s good to keep in mind
though! Maybe a carrot to strive for. Being able to end a goal with no
notice is a nice perk. And it degrades gracefully: if you can manage 5 days
of safety buffer then that’s just 2 days you have to worry about staying on
a road that you no longer really need.)

[1] It’s a slippery slope to the unraveling of all Beeminder’s power! Also
the fact that Beeminder is a Quantified Self tool and fake data ruins that
aspect of it.

On Thu, Apr 25, 2013 at 5:05 AM, Essentiae essentiae@gmail.com wrote:

Brent: I think you’re probably right that with most project, hours
wouldn’t be the right things to measure. For those with a second factor
that can be measured, I add that graph too and just track both. What I’m
trying to track, though, is more of a schedule system than a project goal
(I rotate my projects and want to track that), so hours really are the
main focus. At this point it’s really AIC (ass in chair) that I want to
re-enforce (and to prevent myself from rotating around any projects that I
start avoiding).

Alys: Your “Spend X hours feeling pleased with myself for finishing!”
idea made me chuckle out loud, but I think I might go with something that.
Since it would be the last datapoint, it probably wouldn’t matter, and I
could note the actual total in the comment section of that last entry. It
makes me feel a little twitchy to think that I’ll be leaving fake data of a
kind, but I’ll just have to get over that!

On Thursday, April 25, 2013 3:41:16 AM UTC-4, Essentiae wrote:

I’m wondering what the best way is to set up this kind of thing. I have a
few projects on which I want to spend x hours or minutes per day until
they’re finished, but I don’t know when they’ll be finished. Guestimating
could lead me to have hours left to work, according to my yellow brick
road, even when the project is complete. Shooting for fewer hours to be
safe will likely just result in me doing fewer hours. Is this a job
for Fine Print + Support when the project is complete? Or is it better
served by trying very hard to keep an eye on my buffer so that I can have a
7-day buffer and can end the project on the day it’s actually complete and
then just wait for my buffer to run out?

Thoughts?
Thanks!

–
You received this message because you are subscribed to the Google Groups
“Akratics Anonymous” group.
To unsubscribe from this group and stop receiving emails from it, send an
email to akratics+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

–
http://dreev.es – search://“Daniel Reeves”
Goal tracking + Commitment contracts == http://beeminder.com

6

Ulysses wants to work 30 minutes per day on a project which he thinks might
take as few as 20 and possibly more than 60 hours. Ulysses binds himself to
a contract to work .5 hours daily with an end date two months in the
future, and specifies a road width of 3.5. Because Ulysses is an akratic
adult but not a fool, Ulysses stays on the right side of the road. When the
task is complete, Ulysses modifies the contract by dialing in the end date
as seven days from the real completion and the hours as the actual time.
Ulysses books no additional time. If Ulysses was on the right side of the
road when Ulysses finished, seven days later the contract will end with
Goal Accomplished. If Ulysses was on the wrong side of the road when
Ulysses finished, sometime in the next seven days Ulysses’ failure will
become manifest and Beeminder will enjoy a revenue event.

On Thursday, April 25, 2013 3:41:16 AM UTC-4, Essentiae wrote:

I’m wondering what the best way is to set up this kind of thing. I have a
few projects on which I want to spend x hours or minutes per day until
they’re finished, but I don’t know when they’ll be finished. Guestimating
could lead me to have hours left to work, according to my yellow brick
road, even when the project is complete. Shooting for fewer hours to be
safe will likely just result in me doing fewer hours. Is this a job for
Fine Print + Support when the project is complete? Or is it better served
by trying very hard to keep an eye on my buffer so that I can have a 7-day
buffer and can end the project on the day it’s actually complete and then
just wait for my buffer to run out?

Thoughts?
Thanks!

7

I guess I must be a fool then. I don’t know about anyone else, but there
is absolutely no way I could stay on the right side of a road that takes
a whole week to fall off!

On Mon, Apr 29, 2013 at 9:38 AM, Picky Ulysses bdb8zbs48n@snkmail.comwrote:

Ulysses wants to work 30 minutes per day on a project which he thinks
might take as few as 20 and possibly more than 60 hours. Ulysses binds
himself to a contract to work .5 hours daily with an end date two months in
the future, and specifies a road width of 3.5. Because Ulysses is an
akratic adult but not a fool, Ulysses stays on the right side of the road.
When the task is complete, Ulysses modifies the contract by dialing in the
end date as seven days from the real completion and the hours as the actual
time. Ulysses books no additional time. If Ulysses was on the right side of
the road when Ulysses finished, seven days later the contract will end with
Goal Accomplished. If Ulysses was on the wrong side of the road when
Ulysses finished, sometime in the next seven days Ulysses’ failure will
become manifest and Beeminder will enjoy a revenue event.

On Thursday, April 25, 2013 3:41:16 AM UTC-4, Essentiae wrote:

I’m wondering what the best way is to set up this kind of thing. I have a
few projects on which I want to spend x hours or minutes per day until
they’re finished, but I don’t know when they’ll be finished. Guestimating
could lead me to have hours left to work, according to my yellow brick
road, even when the project is complete. Shooting for fewer hours to be
safe will likely just result in me doing fewer hours. Is this a job
for Fine Print + Support when the project is complete? Or is it better
served by trying very hard to keep an eye on my buffer so that I can have a
7-day buffer and can end the project on the day it’s actually complete and
then just wait for my buffer to run out?

Thoughts?
Thanks!

–
You received this message because you are subscribed to the Google Groups
“Akratics Anonymous” group.
To unsubscribe from this group and stop receiving emails from it, send an
email to akratics+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

8

Picky Ulysses predicts that paying forfeits to Beeminder three times will be sufficient to educate you out of this particular foolishness. His model of akratic mechanics says It will change your motivation from “Do not fall off” to “Stay on the right side.”

This is a testable prediction.

9

It’s a prediction I’ve tested in the past. Here are my findings: 975
academic papers started later than I should have. 975 instances of “Ow; I
should not do that again” (instances that caused considerably more
post-screw-up pain than a $10 levy). 976th academic paper? …Started later
than I should have.

  1. For some akratics, 976 times is not enough to learn this particular
    lesson.
  2. 3 < 976
    C) For some akratics, 3 times is not enough to learn this particular
    lesson. :wink:

I ended up abandoning these goals as they required really frequent road
changes (as project priorities changed) and so on. (And since I track the
pomodoro time I put in in a day, I was a double-tracking.) I’m going to go
with Bent’s suggestion to track a different metric. Sometime in the next
month I’m going to create a graph that just tracks whether I met today’s
minimum (planned last night) for each of my projects with just a yes/no
(1/0). Much cleaner and I can get closer to getting the right targets that
way. If I overestimate how much is left on the last day of a project, I can
use Alys and Daniel’s ideas and do something like a post-mortem on the
project or brainstorm ways to link the project into something new, etc.

On Monday, April 29, 2013 10:58:45 AM UTC-4, Picky Ulysses wrote:

Picky Ulysses predicts that paying forfeits to Beeminder three times will
be sufficient to educate you out of this particular foolishness. His model
of akratic mechanics says It will change your motivation from “Do not fall
off” to “Stay on the right side.”

This is a testable prediction.

10

This cracked me up. I’m totally with Essentiae on this!

Also, I absolutely adore (well, it’s love/hate but it’s really good
for me) my mustdo – d – beeminder goal. I think it will solve Essentiae’s
“finishing early” problem fine. You could even always have your daily
mustdo have an implicit “or finish the project early”.

I’d still like to try something like mustdo but a bit richer, like
Paul Fenwick’s or Katja Grace’s points system for their task lists.

On Wed, May 15, 2013 at 6:09 AM, Essentiae essentiae@gmail.com wrote:

It’s a prediction I’ve tested in the past. Here are my findings: 975
academic papers started later than I should have. 975 instances of “Ow; I
should not do that again” (instances that caused considerably more
post-screw-up pain than a $10 levy). 976th academic paper? …Started later
than I should have.

  1. For some akratics, 976 times is not enough to learn this particular
    lesson.
  2. 3 < 976
    C) For some akratics, 3 times is not enough to learn this particular lesson.
    :wink:

I ended up abandoning these goals as they required really frequent road
changes (as project priorities changed) and so on. (And since I track the
pomodoro time I put in in a day, I was a double-tracking.) I’m going to go
with Bent’s suggestion to track a different metric. Sometime in the next
month I’m going to create a graph that just tracks whether I met today’s
minimum (planned last night) for each of my projects with just a yes/no
(1/0). Much cleaner and I can get closer to getting the right targets that
way. If I overestimate how much is left on the last day of a project, I can
use Alys and Daniel’s ideas and do something like a post-mortem on the
project or brainstorm ways to link the project into something new, etc.

On Monday, April 29, 2013 10:58:45 AM UTC-4, Picky Ulysses wrote:

Picky Ulysses predicts that paying forfeits to Beeminder three times will
be sufficient to educate you out of this particular foolishness. His model
of akratic mechanics says It will change your motivation from “Do not fall
off” to “Stay on the right side.”

This is a testable prediction.

–
You received this message because you are subscribed to the Google Groups
“Akratics Anonymous” group.
To unsubscribe from this group and stop receiving emails from it, send an
email to akratics+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

–
http://dreev.es – search://“Daniel Reeves”
Goal tracking + Commitment contracts == http://beeminder.com