Display habit-forming period

I think most of us have already heard of the 21-day period (or other
similar lengths) purportedly needed to establish a new habit. Google’s Matt
Cutts has a nice and short TED talk about his 30-day challenge system[1].

Today as I looked into my graph after inputting some data, I realized that
the fact the akrasia horizon is a moving “target” (in quotes because it’s
not a target at all; in fact, the definition of a horizon implies that it
is always out of reach) makes me a little anxious and I would instead like
to see a fixed goal I could aim to (sort of a horizon in the sense used in
the term “event horizon” for black holes, i.e. something you can actually
cross). Something like a “habit threshold”.

I decided to post here asking you about it, and having heard of the 21-day
thing, I went to look for a good link to add to my post. Interestingly (put
perhaps not surprisingly), I found out that the 21-day habit-forming
concept is kind of a myth. There are some interesting articles on this
[2][3]. However, these articles mention how studies have revealed that
while habit-forming depends, naturally, on the habit and on the person, the
average period they recorded was 66 days. Also, they found that missing a
day or two in that stretch wasn’t critical and didn’t require counting to
start over.

With that said, I would suggest a new optional feature: the ability to
display the next “habit threshold” on the graph, based on the description
above, as well as the location of such thresholds crossed in the past. This
would provide a shorter-term sub-goal (depending on the goal length, of
course), sort of how checkpoints work in videogames, thus making it easier
to muster the discipline to stick through especially when close to them; I
personally find that, having set a year-long goal, I sometimes feel a
little bit anxious looking at the road and seeing it as never-ending
stairs. Plus, past “checkpoints” would give us pride and confidence that we
can do it — yes, the need for self-discipline and keeping good habits is a
life-long struggle, and it makes sense to represent it as a continually
increasing/decreasing line (depending on the goal, of course!), but a
little encouragement may go a long way in helping us stick with our
improvement commitments — in this case, by visualizing our goals closer to
the present and act on them.

What do you think?

[1] http://www.youtube.com/watch?v=JnfBXjWm7hc
[2]
http://www.guardian.co.uk/lifeandstyle/2009/oct/10/change-your-life-habit-28-day-rule
[3]
http://lifehacker.com/5926583/why-habits-arent-actually-formed-in-21-days

What about using the “tmax” option (Upper bound on the plot range for the
x-axis (latest date)") in the “Advanced” tab and setting it to 21 or 66
days from your start point? Once you reach it, you could change the date to
your next milestone, or delete it altogether. That way you get to see the
distance between your latest point and your milestone, as well as how
high/low your yellow brick road will be by then.

Essy

On Friday, July 12, 2013 11:04:14 AM UTC-4, Waldir Pimenta wrote:

I think most of us have already heard of the 21-day period (or other
similar lengths) purportedly needed to establish a new habit. Google’s Matt
Cutts has a nice and short TED talk about his 30-day challenge system[1].

Today as I looked into my graph after inputting some data, I realized that
the fact the akrasia horizon is a moving “target” (in quotes because it’s
not a target at all; in fact, the definition of a horizon implies that it
is always out of reach) makes me a little anxious and I would instead like
to see a fixed goal I could aim to (sort of a horizon in the sense used in
the term “event horizon” for black holes, i.e. something you can actually
cross). Something like a “habit threshold”.

I decided to post here asking you about it, and having heard of the 21-day
thing, I went to look for a good link to add to my post. Interestingly (put
perhaps not surprisingly), I found out that the 21-day habit-forming
concept is kind of a myth. There are some interesting articles on this
[2][3]. However, these articles mention how studies have revealed that
while habit-forming depends, naturally, on the habit and on the person, the
average period they recorded was 66 days. Also, they found that missing a
day or two in that stretch wasn’t critical and didn’t require counting to
start over.

With that said, I would suggest a new optional feature: the ability to
display the next “habit threshold” on the graph, based on the description
above, as well as the location of such thresholds crossed in the past. This
would provide a shorter-term sub-goal (depending on the goal length, of
course), sort of how checkpoints work in videogames, thus making it easier
to muster the discipline to stick through especially when close to them; I
personally find that, having set a year-long goal, I sometimes feel a
little bit anxious looking at the road and seeing it as never-ending
stairs. Plus, past “checkpoints” would give us pride and confidence that we
can do it — yes, the need for self-discipline and keeping good habits is a
life-long struggle, and it makes sense to represent it as a continually
increasing/decreasing line (depending on the goal, of course!), but a
little encouragement may go a long way in helping us stick with our
improvement commitments — in this case, by visualizing our goals closer to
the present and act on them.

What do you think?

[1] http://www.youtube.com/watch?v=JnfBXjWm7hc
[2]
This column will change your life: How long does it really take to change a habit? | Life and style | The Guardian
[3]
http://lifehacker.com/5926583/why-habits-arent-actually-formed-in-21-days

Waldir, love the way you’re thinking and especially love Essy for
providing an immediate (though perhaps partial and inconvenient)
solution. Eager to hear if you find that Essy’s solution suffices. I
guess the main thing it’s missing is displaying previously reached
checkpoints, but I’m not convinced that’s a good idea. It’s more
complexity in the graph and just showing that you’ve been on the
yellow brick road every day is arguably even better than showing that
you reached specific past checkpoints. Besides, part of the point of
Beeminder is that every single day is such a checkpoint.

On Fri, Jul 12, 2013 at 8:33 AM, Essentiae essentiae@gmail.com wrote:

What about using the “tmax” option (Upper bound on the plot range for the
x-axis (latest date)") in the “Advanced” tab and setting it to 21 or 66 days
from your start point? Once you reach it, you could change the date to your
next milestone, or delete it altogether. That way you get to see the
distance between your latest point and your milestone, as well as how
high/low your yellow brick road will be by then.

Essy

On Friday, July 12, 2013 11:04:14 AM UTC-4, Waldir Pimenta wrote:

I think most of us have already heard of the 21-day period (or other
similar lengths) purportedly needed to establish a new habit. Google’s Matt
Cutts has a nice and short TED talk about his 30-day challenge system[1].

Today as I looked into my graph after inputting some data, I realized that
the fact the akrasia horizon is a moving “target” (in quotes because it’s
not a target at all; in fact, the definition of a horizon implies that it is
always out of reach) makes me a little anxious and I would instead like to
see a fixed goal I could aim to (sort of a horizon in the sense used in the
term “event horizon” for black holes, i.e. something you can actually
cross). Something like a “habit threshold”.

I decided to post here asking you about it, and having heard of the 21-day
thing, I went to look for a good link to add to my post. Interestingly (put
perhaps not surprisingly), I found out that the 21-day habit-forming concept
is kind of a myth. There are some interesting articles on this [2][3].
However, these articles mention how studies have revealed that while
habit-forming depends, naturally, on the habit and on the person, the
average period they recorded was 66 days. Also, they found that missing a
day or two in that stretch wasn’t critical and didn’t require counting to
start over.

With that said, I would suggest a new optional feature: the ability to
display the next “habit threshold” on the graph, based on the description
above, as well as the location of such thresholds crossed in the past. This
would provide a shorter-term sub-goal (depending on the goal length, of
course), sort of how checkpoints work in videogames, thus making it easier
to muster the discipline to stick through especially when close to them; I
personally find that, having set a year-long goal, I sometimes feel a little
bit anxious looking at the road and seeing it as never-ending stairs. Plus,
past “checkpoints” would give us pride and confidence that we can do it —
yes, the need for self-discipline and keeping good habits is a life-long
struggle, and it makes sense to represent it as a continually
increasing/decreasing line (depending on the goal, of course!), but a little
encouragement may go a long way in helping us stick with our improvement
commitments — in this case, by visualizing our goals closer to the present
and act on them.

What do you think?

[1] http://www.youtube.com/watch?v=JnfBXjWm7hc
[2]
This column will change your life: How long does it really take to change a habit? | Life and style | The Guardian
[3]
http://lifehacker.com/5926583/why-habits-arent-actually-formed-in-21-days


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

I, too, remember reading lots of articles about these 66 days supposedly
supported by ‘studies’. In reality that’s one single study and if I
remember correctly even there they found significant differences between
different habits.

So I don’t think there is real scientific evidence supporting 66 days, 30
days or any other number at this point.

On Fri, Jul 12, 2013 at 8:13 PM, Daniel Reeves dreeves@beeminder.comwrote:

Waldir, love the way you’re thinking and especially love Essy for
providing an immediate (though perhaps partial and inconvenient)
solution. Eager to hear if you find that Essy’s solution suffices. I
guess the main thing it’s missing is displaying previously reached
checkpoints, but I’m not convinced that’s a good idea. It’s more
complexity in the graph and just showing that you’ve been on the
yellow brick road every day is arguably even better than showing that
you reached specific past checkpoints. Besides, part of the point of
Beeminder is that every single day is such a checkpoint.

On Fri, Jul 12, 2013 at 8:33 AM, Essentiae essentiae@gmail.com wrote:

What about using the “tmax” option (Upper bound on the plot range for the
x-axis (latest date)") in the “Advanced” tab and setting it to 21 or 66
days
from your start point? Once you reach it, you could change the date to
your
next milestone, or delete it altogether. That way you get to see the
distance between your latest point and your milestone, as well as how
high/low your yellow brick road will be by then.

Essy

On Friday, July 12, 2013 11:04:14 AM UTC-4, Waldir Pimenta wrote:

I think most of us have already heard of the 21-day period (or other
similar lengths) purportedly needed to establish a new habit. Google’s
Matt
Cutts has a nice and short TED talk about his 30-day challenge
system[1].

Today as I looked into my graph after inputting some data, I realized
that
the fact the akrasia horizon is a moving “target” (in quotes because
it’s
not a target at all; in fact, the definition of a horizon implies that
it is
always out of reach) makes me a little anxious and I would instead like
to
see a fixed goal I could aim to (sort of a horizon in the sense used in
the
term “event horizon” for black holes, i.e. something you can actually
cross). Something like a “habit threshold”.

I decided to post here asking you about it, and having heard of the
21-day
thing, I went to look for a good link to add to my post. Interestingly
(put
perhaps not surprisingly), I found out that the 21-day habit-forming
concept
is kind of a myth. There are some interesting articles on this [2][3].
However, these articles mention how studies have revealed that while
habit-forming depends, naturally, on the habit and on the person, the
average period they recorded was 66 days. Also, they found that missing
a
day or two in that stretch wasn’t critical and didn’t require counting
to
start over.

With that said, I would suggest a new optional feature: the ability to
display the next “habit threshold” on the graph, based on the
description
above, as well as the location of such thresholds crossed in the past.
This
would provide a shorter-term sub-goal (depending on the goal length, of
course), sort of how checkpoints work in videogames, thus making it
easier
to muster the discipline to stick through especially when close to
them; I
personally find that, having set a year-long goal, I sometimes feel a
little
bit anxious looking at the road and seeing it as never-ending stairs.
Plus,
past “checkpoints” would give us pride and confidence that we can do it

yes, the need for self-discipline and keeping good habits is a life-long
struggle, and it makes sense to represent it as a continually
increasing/decreasing line (depending on the goal, of course!), but a
little
encouragement may go a long way in helping us stick with our improvement
commitments — in this case, by visualizing our goals closer to the
present
and act on them.

What do you think?

[1] http://www.youtube.com/watch?v=JnfBXjWm7hc
[2]

This column will change your life: How long does it really take to change a habit? | Life and style | The Guardian

[3]

http://lifehacker.com/5926583/why-habits-arent-actually-formed-in-21-days


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


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.

Sorry, guys, for the long delay. I think some good points have been raised.

First of all, I wouldn’t consider Essy’s suggestion a solution – perhaps a
workaround, but a quite inconvenient one at that, since we’d have to
continually adjust our graphs. I have (and I believe many chronic
procrastinators have as well) been a victim of overplanning and overplaying
with my task sceduling tools in the past, and I don’t think Beeminder
should encourage that.

Second, I agree that displaying past “checkpoints” in the graph will
clutter it unnecessarily. Still, I’m sure we can find a way to have
information on past progress available without such problems. For example,
a site called boincstats.com (for the distributed computing platform) shows
stats such as most productive day, recent average progress, etc. Beeminder
could have a “longest in-the-road streak”, as well as stats similar to
those I mentioned. For all of these, it would be easy to filter the values
for time period (last week, last month, all time, etc). A simple stats
table in the settings page (say, a new “stats” tab) could provide this. It
should also be fairly easy to extend these stats from only showing maximum
values to also include averages such as the median, mean or mode. But these
are only suggestions for completeness; I’m aware that showing too much data
to the user may compromise its potential for generating insight. Options to
alleviate this could be de-emphasising the extra data, perhaps by using a
smaller font, fainter colors, or placing the values in a tooltip.

Finally, Brian’s remark combined with the fact that habit formation periods
vary wildly due to many factors, would suggest that indeed showing a
fixed-length period as a target in the graph wouldn’t be ideal. But the
measures I described above would likely be a good --dare I say, better–
substitute. So a graph could have, say, a vertical line placed at the
median length of a streak (starting from the start of the current streak,
so it doesn’t slide continuously), showing a period for which, by
definition, we are likely to be able to make a continuous streak, and a
horizontal line, intersecting it, at the median value we have accumulated
in such a period.

All of this is mostly number crunching; user-visible changes are an extra
tab in the goal settings containing a simple html table, and two extra
(optional) lines in the graph. So implementing it might be a low-effort,
high-payout effort, assuming the encouragement benefits of “achievements”
indeed as positive as I suspect. We could even go wild and show some of
those stats as badges in the userpage, as seems to be quite popular in many
sites lately :stuck_out_tongue:

So, what do you think?

Waldir

On Saturday, July 13, 2013 1:40:21 PM UTC+1, Brian F Crain wrote:

I, too, remember reading lots of articles about these 66 days supposedly
supported by ‘studies’. In reality that’s one single study and if I
remember correctly even there they found significant differences between
different habits.

So I don’t think there is real scientific evidence supporting 66 days, 30
days or any other number at this point.

On Fri, Jul 12, 2013 at 8:13 PM, Daniel Reeves <dre...@beeminder.com<javascript:>

wrote:

Waldir, love the way you’re thinking and especially love Essy for
providing an immediate (though perhaps partial and inconvenient)
solution. Eager to hear if you find that Essy’s solution suffices. I
guess the main thing it’s missing is displaying previously reached
checkpoints, but I’m not convinced that’s a good idea. It’s more
complexity in the graph and just showing that you’ve been on the
yellow brick road every day is arguably even better than showing that
you reached specific past checkpoints. Besides, part of the point of
Beeminder is that every single day is such a checkpoint.

On Fri, Jul 12, 2013 at 8:33 AM, Essentiae <esse...@gmail.com<javascript:>>
wrote:

What about using the “tmax” option (Upper bound on the plot range for
the
x-axis (latest date)") in the “Advanced” tab and setting it to 21 or 66
days
from your start point? Once you reach it, you could change the date to
your
next milestone, or delete it altogether. That way you get to see the
distance between your latest point and your milestone, as well as how
high/low your yellow brick road will be by then.

Essy

On Friday, July 12, 2013 11:04:14 AM UTC-4, Waldir Pimenta wrote:

I think most of us have already heard of the 21-day period (or other
similar lengths) purportedly needed to establish a new habit. Google’s
Matt
Cutts has a nice and short TED talk about his 30-day challenge
system[1].

Today as I looked into my graph after inputting some data, I realized
that
the fact the akrasia horizon is a moving “target” (in quotes because
it’s
not a target at all; in fact, the definition of a horizon implies that
it is
always out of reach) makes me a little anxious and I would instead
like to
see a fixed goal I could aim to (sort of a horizon in the sense used
in the
term “event horizon” for black holes, i.e. something you can actually
cross). Something like a “habit threshold”.

I decided to post here asking you about it, and having heard of the
21-day
thing, I went to look for a good link to add to my post. Interestingly
(put
perhaps not surprisingly), I found out that the 21-day habit-forming
concept
is kind of a myth. There are some interesting articles on this [2][3].
However, these articles mention how studies have revealed that while
habit-forming depends, naturally, on the habit and on the person, the
average period they recorded was 66 days. Also, they found that
missing a
day or two in that stretch wasn’t critical and didn’t require counting
to
start over.

With that said, I would suggest a new optional feature: the ability to
display the next “habit threshold” on the graph, based on the
description
above, as well as the location of such thresholds crossed in the past.
This
would provide a shorter-term sub-goal (depending on the goal length, of
course), sort of how checkpoints work in videogames, thus making it
easier
to muster the discipline to stick through especially when close to
them; I
personally find that, having set a year-long goal, I sometimes feel a
little
bit anxious looking at the road and seeing it as never-ending stairs.
Plus,
past “checkpoints” would give us pride and confidence that we can do
it —
yes, the need for self-discipline and keeping good habits is a
life-long
struggle, and it makes sense to represent it as a continually
increasing/decreasing line (depending on the goal, of course!), but a
little
encouragement may go a long way in helping us stick with our
improvement
commitments — in this case, by visualizing our goals closer to the
present
and act on them.

What do you think?

[1] http://www.youtube.com/watch?v=JnfBXjWm7hc
[2]

This column will change your life: How long does it really take to change a habit? | Life and style | The Guardian

[3]

http://lifehacker.com/5926583/why-habits-arent-actually-formed-in-21-days


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+u...@googlegroups.com <javascript:>.
For more options, visit https://groups.google.com/groups/opt_out.


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


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+u...@googlegroups.com <javascript:>.
For more options, visit https://groups.google.com/groups/opt_out.

Mulling this; thanks Waldir! I don’t want to just throw spaghetti at
the wall though (eg, “longest in-the-road streak” doesn’t strike me as
an important stat to highlight, though we do have a similar statistic
in the legit check emails) so what would be especially great is if you
can identify specific stats that you really want (like you personally)
and add them individually on uservoice.beeminder.com so people can
vote on them.

I do think we should highlight your current actual (historical
average) rate, like what you could’ve had the yellow brick road dialed
to, but we’re brewing something else for that…

On Fri, Sep 6, 2013 at 8:40 AM, Waldir Pimenta waldir.pimenta@gmail.com wrote:

Sorry, guys, for the long delay. I think some good points have been raised.

First of all, I wouldn’t consider Essy’s suggestion a solution – perhaps a
workaround, but a quite inconvenient one at that, since we’d have to
continually adjust our graphs. I have (and I believe many chronic
procrastinators have as well) been a victim of overplanning and overplaying
with my task sceduling tools in the past, and I don’t think Beeminder should
encourage that.

Second, I agree that displaying past “checkpoints” in the graph will clutter
it unnecessarily. Still, I’m sure we can find a way to have information on
past progress available without such problems. For example, a site called
boincstats.com (for the distributed computing platform) shows stats such as
most productive day, recent average progress, etc. Beeminder could have a
“longest in-the-road streak”, as well as stats similar to those I mentioned.
For all of these, it would be easy to filter the values for time period
(last week, last month, all time, etc). A simple stats table in the settings
page (say, a new “stats” tab) could provide this. It should also be fairly
easy to extend these stats from only showing maximum values to also include
averages such as the median, mean or mode. But these are only suggestions
for completeness; I’m aware that showing too much data to the user may
compromise its potential for generating insight. Options to alleviate this
could be de-emphasising the extra data, perhaps by using a smaller font,
fainter colors, or placing the values in a tooltip.

Finally, Brian’s remark combined with the fact that habit formation periods
vary wildly due to many factors, would suggest that indeed showing a
fixed-length period as a target in the graph wouldn’t be ideal. But the
measures I described above would likely be a good --dare I say, better–
substitute. So a graph could have, say, a vertical line placed at the median
length of a streak (starting from the start of the current streak, so it
doesn’t slide continuously), showing a period for which, by definition, we
are likely to be able to make a continuous streak, and a horizontal line,
intersecting it, at the median value we have accumulated in such a period.

All of this is mostly number crunching; user-visible changes are an extra
tab in the goal settings containing a simple html table, and two extra
(optional) lines in the graph. So implementing it might be a low-effort,
high-payout effort, assuming the encouragement benefits of “achievements”
indeed as positive as I suspect. We could even go wild and show some of
those stats as badges in the userpage, as seems to be quite popular in many
sites lately :stuck_out_tongue:

So, what do you think?

Waldir

On Saturday, July 13, 2013 1:40:21 PM UTC+1, Brian F Crain wrote:

I, too, remember reading lots of articles about these 66 days supposedly
supported by ‘studies’. In reality that’s one single study and if I remember
correctly even there they found significant differences between different
habits.

So I don’t think there is real scientific evidence supporting 66 days, 30
days or any other number at this point.

On Fri, Jul 12, 2013 at 8:13 PM, Daniel Reeves dre...@beeminder.com
wrote:

Waldir, love the way you’re thinking and especially love Essy for
providing an immediate (though perhaps partial and inconvenient)
solution. Eager to hear if you find that Essy’s solution suffices. I
guess the main thing it’s missing is displaying previously reached
checkpoints, but I’m not convinced that’s a good idea. It’s more
complexity in the graph and just showing that you’ve been on the
yellow brick road every day is arguably even better than showing that
you reached specific past checkpoints. Besides, part of the point of
Beeminder is that every single day is such a checkpoint.

On Fri, Jul 12, 2013 at 8:33 AM, Essentiae esse...@gmail.com wrote:

What about using the “tmax” option (Upper bound on the plot range for
the
x-axis (latest date)") in the “Advanced” tab and setting it to 21 or 66
days
from your start point? Once you reach it, you could change the date to
your
next milestone, or delete it altogether. That way you get to see the
distance between your latest point and your milestone, as well as how
high/low your yellow brick road will be by then.

Essy

On Friday, July 12, 2013 11:04:14 AM UTC-4, Waldir Pimenta wrote:

I think most of us have already heard of the 21-day period (or other
similar lengths) purportedly needed to establish a new habit. Google’s
Matt
Cutts has a nice and short TED talk about his 30-day challenge
system[1].

Today as I looked into my graph after inputting some data, I realized
that
the fact the akrasia horizon is a moving “target” (in quotes because
it’s
not a target at all; in fact, the definition of a horizon implies that
it is
always out of reach) makes me a little anxious and I would instead
like to
see a fixed goal I could aim to (sort of a horizon in the sense used
in the
term “event horizon” for black holes, i.e. something you can actually
cross). Something like a “habit threshold”.

I decided to post here asking you about it, and having heard of the
21-day
thing, I went to look for a good link to add to my post. Interestingly
(put
perhaps not surprisingly), I found out that the 21-day habit-forming
concept
is kind of a myth. There are some interesting articles on this [2][3].
However, these articles mention how studies have revealed that while
habit-forming depends, naturally, on the habit and on the person, the
average period they recorded was 66 days. Also, they found that
missing a
day or two in that stretch wasn’t critical and didn’t require counting
to
start over.

With that said, I would suggest a new optional feature: the ability to
display the next “habit threshold” on the graph, based on the
description
above, as well as the location of such thresholds crossed in the past.
This
would provide a shorter-term sub-goal (depending on the goal length,
of
course), sort of how checkpoints work in videogames, thus making it
easier
to muster the discipline to stick through especially when close to
them; I
personally find that, having set a year-long goal, I sometimes feel a
little
bit anxious looking at the road and seeing it as never-ending stairs.
Plus,
past “checkpoints” would give us pride and confidence that we can do
it —
yes, the need for self-discipline and keeping good habits is a
life-long
struggle, and it makes sense to represent it as a continually
increasing/decreasing line (depending on the goal, of course!), but a
little
encouragement may go a long way in helping us stick with our
improvement
commitments — in this case, by visualizing our goals closer to the
present
and act on them.

What do you think?

[1] http://www.youtube.com/watch?v=JnfBXjWm7hc
[2]

This column will change your life: How long does it really take to change a habit? | Life and style | The Guardian
[3]

http://lifehacker.com/5926583/why-habits-arent-actually-formed-in-21-days


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+u...@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


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+u...@googlegroups.com.

For more options, visit https://groups.google.com/groups/opt_out.


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

I’ve been thinking some more about this, and I would streamline my proposal
down to two things:

  1. In the graph, display a “potential data point” located at the value
    [previous day + median value of positive (in a “do more” goal) data
    points]. This would be aimed at encouraging immediate action, showing a
    perfectly achievable target that we can reach on that day. Additionally,
    there could be another target, above it, located at [previous day + maximum
    daily value], which is also a goal that is proven to be achievable,
    although harder. I was thinking something like this:
    http://i.imgur.com/iGM35SR.png

  2. In the “goal stats” box in the right sidebar of each goal’s page,
    display 3 badges: “best day”, “best ” (depending on the goal’s rate
    units; e.g. week), and “longest run”, defined as a continually improving
    streak (i.e., excluding flat or negative values, but allowing one or two
    such days in the middle of a streak, per the reports of the studies I
    mentioned earlier in this thread: “they found that missing a day or two in
    that stretch wasn’t critical and didn’t require counting to start over.”).
    This would be aimed at increasing confidence in our overall ability to
    improve.

Any thoughts?

On Friday, September 6, 2013 10:08:20 PM UTC+1, Daniel Reeves wrote:

Mulling this; thanks Waldir! I don’t want to just throw spaghetti at
the wall though (eg, “longest in-the-road streak” doesn’t strike me as
an important stat to highlight, though we do have a similar statistic
in the legit check emails) so what would be especially great is if you
can identify specific stats that you really want (like you personally)
and add them individually on uservoice.beeminder.com so people can
vote on them.

I do think we should highlight your current actual (historical
average) rate, like what you could’ve had the yellow brick road dialed
to, but we’re brewing something else for that…

On Fri, Sep 6, 2013 at 8:40 AM, Waldir Pimenta <waldir....@gmail.com<javascript:>>
wrote:

Sorry, guys, for the long delay. I think some good points have been
raised.

First of all, I wouldn’t consider Essy’s suggestion a solution –
perhaps a
workaround, but a quite inconvenient one at that, since we’d have to
continually adjust our graphs. I have (and I believe many chronic
procrastinators have as well) been a victim of overplanning and
overplaying
with my task sceduling tools in the past, and I don’t think Beeminder
should
encourage that.

Second, I agree that displaying past “checkpoints” in the graph will
clutter
it unnecessarily. Still, I’m sure we can find a way to have information
on
past progress available without such problems. For example, a site
called
boincstats.com (for the distributed computing platform) shows stats
such as
most productive day, recent average progress, etc. Beeminder could have
a
“longest in-the-road streak”, as well as stats similar to those I
mentioned.
For all of these, it would be easy to filter the values for time period
(last week, last month, all time, etc). A simple stats table in the
settings
page (say, a new “stats” tab) could provide this. It should also be
fairly
easy to extend these stats from only showing maximum values to also
include
averages such as the median, mean or mode. But these are only
suggestions
for completeness; I’m aware that showing too much data to the user may
compromise its potential for generating insight. Options to alleviate
this
could be de-emphasising the extra data, perhaps by using a smaller font,
fainter colors, or placing the values in a tooltip.

Finally, Brian’s remark combined with the fact that habit formation
periods
vary wildly due to many factors, would suggest that indeed showing a
fixed-length period as a target in the graph wouldn’t be ideal. But the
measures I described above would likely be a good --dare I say, better–
substitute. So a graph could have, say, a vertical line placed at the
median
length of a streak (starting from the start of the current streak, so it
doesn’t slide continuously), showing a period for which, by definition,
we
are likely to be able to make a continuous streak, and a horizontal
line,
intersecting it, at the median value we have accumulated in such a
period.

All of this is mostly number crunching; user-visible changes are an
extra
tab in the goal settings containing a simple html table, and two extra
(optional) lines in the graph. So implementing it might be a low-effort,
high-payout effort, assuming the encouragement benefits of
“achievements”
indeed as positive as I suspect. We could even go wild and show some of
those stats as badges in the userpage, as seems to be quite popular in
many
sites lately :stuck_out_tongue:

So, what do you think?

Waldir

On Saturday, July 13, 2013 1:40:21 PM UTC+1, Brian F Crain wrote:

I, too, remember reading lots of articles about these 66 days
supposedly
supported by ‘studies’. In reality that’s one single study and if I
remember
correctly even there they found significant differences between
different
habits.

So I don’t think there is real scientific evidence supporting 66 days,
30
days or any other number at this point.

On Fri, Jul 12, 2013 at 8:13 PM, Daniel Reeves dre...@beeminder.com
wrote:

Waldir, love the way you’re thinking and especially love Essy for
providing an immediate (though perhaps partial and inconvenient)
solution. Eager to hear if you find that Essy’s solution suffices. I
guess the main thing it’s missing is displaying previously reached
checkpoints, but I’m not convinced that’s a good idea. It’s more
complexity in the graph and just showing that you’ve been on the
yellow brick road every day is arguably even better than showing that
you reached specific past checkpoints. Besides, part of the point of
Beeminder is that every single day is such a checkpoint.

On Fri, Jul 12, 2013 at 8:33 AM, Essentiae esse...@gmail.com wrote:

What about using the “tmax” option (Upper bound on the plot range
for
the
x-axis (latest date)") in the “Advanced” tab and setting it to 21 or
66
days
from your start point? Once you reach it, you could change the date
to
your
next milestone, or delete it altogether. That way you get to see the
distance between your latest point and your milestone, as well as
how
high/low your yellow brick road will be by then.

Essy

On Friday, July 12, 2013 11:04:14 AM UTC-4, Waldir Pimenta wrote:

I think most of us have already heard of the 21-day period (or
other
similar lengths) purportedly needed to establish a new habit.
Google’s
Matt
Cutts has a nice and short TED talk about his 30-day challenge
system[1].

Today as I looked into my graph after inputting some data, I
realized
that
the fact the akrasia horizon is a moving “target” (in quotes
because
it’s
not a target at all; in fact, the definition of a horizon implies
that
it is
always out of reach) makes me a little anxious and I would instead
like to
see a fixed goal I could aim to (sort of a horizon in the sense
used
in the
term “event horizon” for black holes, i.e. something you can
actually
cross). Something like a “habit threshold”.

I decided to post here asking you about it, and having heard of the
21-day
thing, I went to look for a good link to add to my post.
Interestingly
(put
perhaps not surprisingly), I found out that the 21-day
habit-forming
concept
is kind of a myth. There are some interesting articles on this
[2][3].
However, these articles mention how studies have revealed that
while
habit-forming depends, naturally, on the habit and on the person,
the
average period they recorded was 66 days. Also, they found that
missing a
day or two in that stretch wasn’t critical and didn’t require
counting
to
start over.

With that said, I would suggest a new optional feature: the ability
to
display the next “habit threshold” on the graph, based on the
description
above, as well as the location of such thresholds crossed in the
past.
This
would provide a shorter-term sub-goal (depending on the goal
length,
of
course), sort of how checkpoints work in videogames, thus making it
easier
to muster the discipline to stick through especially when close to
them; I
personally find that, having set a year-long goal, I sometimes feel
a
little
bit anxious looking at the road and seeing it as never-ending
stairs.
Plus,
past “checkpoints” would give us pride and confidence that we can
do
it —
yes, the need for self-discipline and keeping good habits is a
life-long
struggle, and it makes sense to represent it as a continually
increasing/decreasing line (depending on the goal, of course!), but
a
little
encouragement may go a long way in helping us stick with our
improvement
commitments — in this case, by visualizing our goals closer to the
present
and act on them.

What do you think?

[1] http://www.youtube.com/watch?v=JnfBXjWm7hc
[2]

This column will change your life: How long does it really take to change a habit? | Life and style | The Guardian

[3]

http://lifehacker.com/5926583/why-habits-arent-actually-formed-in-21-days


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+u...@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


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+u...@googlegroups.com.

For more options, visit https://groups.google.com/groups/opt_out.


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+u...@googlegroups.com <javascript:>.
For more options, visit https://groups.google.com/groups/opt_out.


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

That looks nice! Thanks Waldir! Possibly we should save this for when
we have javascript graphs with explanatory mouseovers and whatnot. If
you have a trello account and are game, could I add you to our trello
board at Trello ?

On Mon, Sep 9, 2013 at 6:40 AM, Waldir Pimenta waldir.pimenta@gmail.com wrote:

I’ve been thinking some more about this, and I would streamline my proposal
down to two things:

  1. In the graph, display a “potential data point” located at the value
    [previous day + median value of positive (in a “do more” goal) data points].
    This would be aimed at encouraging immediate action, showing a perfectly
    achievable target that we can reach on that day. Additionally, there could
    be another target, above it, located at [previous day + maximum daily
    value], which is also a goal that is proven to be achievable, although
    harder. I was thinking something like this: http://i.imgur.com/iGM35SR.png

  2. In the “goal stats” box in the right sidebar of each goal’s page, display
    3 badges: “best day”, “best ” (depending on the goal’s rate units;
    e.g. week), and “longest run”, defined as a continually improving streak
    (i.e., excluding flat or negative values, but allowing one or two such days
    in the middle of a streak, per the reports of the studies I mentioned
    earlier in this thread: “they found that missing a day or two in that
    stretch wasn’t critical and didn’t require counting to start over.”). This
    would be aimed at increasing confidence in our overall ability to improve.

Any thoughts?

On Friday, September 6, 2013 10:08:20 PM UTC+1, Daniel Reeves wrote:

Mulling this; thanks Waldir! I don’t want to just throw spaghetti at
the wall though (eg, “longest in-the-road streak” doesn’t strike me as
an important stat to highlight, though we do have a similar statistic
in the legit check emails) so what would be especially great is if you
can identify specific stats that you really want (like you personally)
and add them individually on uservoice.beeminder.com so people can
vote on them.

I do think we should highlight your current actual (historical
average) rate, like what you could’ve had the yellow brick road dialed
to, but we’re brewing something else for that…

On Fri, Sep 6, 2013 at 8:40 AM, Waldir Pimenta waldir....@gmail.com
wrote:

Sorry, guys, for the long delay. I think some good points have been
raised.

First of all, I wouldn’t consider Essy’s suggestion a solution –
perhaps a
workaround, but a quite inconvenient one at that, since we’d have to
continually adjust our graphs. I have (and I believe many chronic
procrastinators have as well) been a victim of overplanning and
overplaying
with my task sceduling tools in the past, and I don’t think Beeminder
should
encourage that.

Second, I agree that displaying past “checkpoints” in the graph will
clutter
it unnecessarily. Still, I’m sure we can find a way to have information
on
past progress available without such problems. For example, a site
called
boincstats.com (for the distributed computing platform) shows stats such
as
most productive day, recent average progress, etc. Beeminder could have
a
“longest in-the-road streak”, as well as stats similar to those I
mentioned.
For all of these, it would be easy to filter the values for time period
(last week, last month, all time, etc). A simple stats table in the
settings
page (say, a new “stats” tab) could provide this. It should also be
fairly
easy to extend these stats from only showing maximum values to also
include
averages such as the median, mean or mode. But these are only
suggestions
for completeness; I’m aware that showing too much data to the user may
compromise its potential for generating insight. Options to alleviate
this
could be de-emphasising the extra data, perhaps by using a smaller font,
fainter colors, or placing the values in a tooltip.

Finally, Brian’s remark combined with the fact that habit formation
periods
vary wildly due to many factors, would suggest that indeed showing a
fixed-length period as a target in the graph wouldn’t be ideal. But the
measures I described above would likely be a good --dare I say, better–
substitute. So a graph could have, say, a vertical line placed at the
median
length of a streak (starting from the start of the current streak, so it
doesn’t slide continuously), showing a period for which, by definition,
we
are likely to be able to make a continuous streak, and a horizontal
line,
intersecting it, at the median value we have accumulated in such a
period.

All of this is mostly number crunching; user-visible changes are an
extra
tab in the goal settings containing a simple html table, and two extra
(optional) lines in the graph. So implementing it might be a low-effort,
high-payout effort, assuming the encouragement benefits of
“achievements”
indeed as positive as I suspect. We could even go wild and show some of
those stats as badges in the userpage, as seems to be quite popular in
many
sites lately :stuck_out_tongue:

So, what do you think?

Waldir

On Saturday, July 13, 2013 1:40:21 PM UTC+1, Brian F Crain wrote:

I, too, remember reading lots of articles about these 66 days
supposedly
supported by ‘studies’. In reality that’s one single study and if I
remember
correctly even there they found significant differences between
different
habits.

So I don’t think there is real scientific evidence supporting 66 days,
30
days or any other number at this point.

On Fri, Jul 12, 2013 at 8:13 PM, Daniel Reeves dre...@beeminder.com
wrote:

Waldir, love the way you’re thinking and especially love Essy for
providing an immediate (though perhaps partial and inconvenient)
solution. Eager to hear if you find that Essy’s solution suffices. I
guess the main thing it’s missing is displaying previously reached
checkpoints, but I’m not convinced that’s a good idea. It’s more
complexity in the graph and just showing that you’ve been on the
yellow brick road every day is arguably even better than showing that
you reached specific past checkpoints. Besides, part of the point of
Beeminder is that every single day is such a checkpoint.

On Fri, Jul 12, 2013 at 8:33 AM, Essentiae esse...@gmail.com wrote:

What about using the “tmax” option (Upper bound on the plot range
for
the
x-axis (latest date)") in the “Advanced” tab and setting it to 21 or
66
days
from your start point? Once you reach it, you could change the date
to
your
next milestone, or delete it altogether. That way you get to see the
distance between your latest point and your milestone, as well as
how
high/low your yellow brick road will be by then.

Essy

On Friday, July 12, 2013 11:04:14 AM UTC-4, Waldir Pimenta wrote:

I think most of us have already heard of the 21-day period (or
other
similar lengths) purportedly needed to establish a new habit.
Google’s
Matt
Cutts has a nice and short TED talk about his 30-day challenge
system[1].

Today as I looked into my graph after inputting some data, I
realized
that
the fact the akrasia horizon is a moving “target” (in quotes
because
it’s
not a target at all; in fact, the definition of a horizon implies
that
it is
always out of reach) makes me a little anxious and I would instead
like to
see a fixed goal I could aim to (sort of a horizon in the sense
used
in the
term “event horizon” for black holes, i.e. something you can
actually
cross). Something like a “habit threshold”.

I decided to post here asking you about it, and having heard of the
21-day
thing, I went to look for a good link to add to my post.
Interestingly
(put
perhaps not surprisingly), I found out that the 21-day
habit-forming
concept
is kind of a myth. There are some interesting articles on this
[2][3].
However, these articles mention how studies have revealed that
while
habit-forming depends, naturally, on the habit and on the person,
the
average period they recorded was 66 days. Also, they found that
missing a
day or two in that stretch wasn’t critical and didn’t require
counting
to
start over.

With that said, I would suggest a new optional feature: the ability
to
display the next “habit threshold” on the graph, based on the
description
above, as well as the location of such thresholds crossed in the
past.
This
would provide a shorter-term sub-goal (depending on the goal
length,
of
course), sort of how checkpoints work in videogames, thus making it
easier
to muster the discipline to stick through especially when close to
them; I
personally find that, having set a year-long goal, I sometimes feel
a
little
bit anxious looking at the road and seeing it as never-ending
stairs.
Plus,
past “checkpoints” would give us pride and confidence that we can
do
it —
yes, the need for self-discipline and keeping good habits is a
life-long
struggle, and it makes sense to represent it as a continually
increasing/decreasing line (depending on the goal, of course!), but
a
little
encouragement may go a long way in helping us stick with our
improvement
commitments — in this case, by visualizing our goals closer to the
present
and act on them.

What do you think?

[1] http://www.youtube.com/watch?v=JnfBXjWm7hc
[2]

This column will change your life: How long does it really take to change a habit? | Life and style | The Guardian
[3]

http://lifehacker.com/5926583/why-habits-arent-actually-formed-in-21-days


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+u...@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


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+u...@googlegroups.com.

For more options, visit https://groups.google.com/groups/opt_out.


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+u...@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


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

That’s flattering, but one of the reasons I need beeminder in the first
place is my tendency for getting involved in side projects and losing focus
on my main work. :slight_smile:
I do have the trello board bookmarked and will check it out occasionally,
but for now I can only guarantee (semi-regular) participation in this
mailing list.
If you still think it could be a good idea to add me, my account is linked
to the address waldir@email.com.

On Tuesday, September 10, 2013 2:53:26 AM UTC+1, Daniel Reeves wrote:

That looks nice! Thanks Waldir! Possibly we should save this for when
we have javascript graphs with explanatory mouseovers and whatnot. If
you have a trello account and are game, could I add you to our trello
board at Trello ?

On Mon, Sep 9, 2013 at 6:40 AM, Waldir Pimenta <waldir....@gmail.com<javascript:>>
wrote:

I’ve been thinking some more about this, and I would streamline my
proposal
down to two things:

  1. In the graph, display a “potential data point” located at the value
    [previous day + median value of positive (in a “do more” goal) data
    points].
    This would be aimed at encouraging immediate action, showing a perfectly
    achievable target that we can reach on that day. Additionally, there
    could
    be another target, above it, located at [previous day + maximum daily
    value], which is also a goal that is proven to be achievable, although
    harder. I was thinking something like this:
    http://i.imgur.com/iGM35SR.png

  2. In the “goal stats” box in the right sidebar of each goal’s page,
    display
    3 badges: “best day”, “best ” (depending on the goal’s rate
    units;
    e.g. week), and “longest run”, defined as a continually improving streak
    (i.e., excluding flat or negative values, but allowing one or two such
    days
    in the middle of a streak, per the reports of the studies I mentioned
    earlier in this thread: “they found that missing a day or two in that
    stretch wasn’t critical and didn’t require counting to start over.”).
    This
    would be aimed at increasing confidence in our overall ability to
    improve.

Any thoughts?

On Friday, September 6, 2013 10:08:20 PM UTC+1, Daniel Reeves wrote:

Mulling this; thanks Waldir! I don’t want to just throw spaghetti at
the wall though (eg, “longest in-the-road streak” doesn’t strike me as
an important stat to highlight, though we do have a similar statistic
in the legit check emails) so what would be especially great is if you
can identify specific stats that you really want (like you personally)
and add them individually on uservoice.beeminder.com so people can
vote on them.

I do think we should highlight your current actual (historical
average) rate, like what you could’ve had the yellow brick road dialed
to, but we’re brewing something else for that…

On Fri, Sep 6, 2013 at 8:40 AM, Waldir Pimenta waldir....@gmail.com
wrote:

Sorry, guys, for the long delay. I think some good points have been
raised.

First of all, I wouldn’t consider Essy’s suggestion a solution –
perhaps a
workaround, but a quite inconvenient one at that, since we’d have to
continually adjust our graphs. I have (and I believe many chronic
procrastinators have as well) been a victim of overplanning and
overplaying
with my task sceduling tools in the past, and I don’t think Beeminder
should
encourage that.

Second, I agree that displaying past “checkpoints” in the graph will
clutter
it unnecessarily. Still, I’m sure we can find a way to have
information
on
past progress available without such problems. For example, a site
called
boincstats.com (for the distributed computing platform) shows stats
such
as
most productive day, recent average progress, etc. Beeminder could
have
a
“longest in-the-road streak”, as well as stats similar to those I
mentioned.
For all of these, it would be easy to filter the values for time
period
(last week, last month, all time, etc). A simple stats table in the
settings
page (say, a new “stats” tab) could provide this. It should also be
fairly
easy to extend these stats from only showing maximum values to also
include
averages such as the median, mean or mode. But these are only
suggestions
for completeness; I’m aware that showing too much data to the user
may
compromise its potential for generating insight. Options to alleviate
this
could be de-emphasising the extra data, perhaps by using a smaller
font,
fainter colors, or placing the values in a tooltip.

Finally, Brian’s remark combined with the fact that habit formation
periods
vary wildly due to many factors, would suggest that indeed showing a
fixed-length period as a target in the graph wouldn’t be ideal. But
the
measures I described above would likely be a good --dare I say,
better–
substitute. So a graph could have, say, a vertical line placed at the
median
length of a streak (starting from the start of the current streak, so
it
doesn’t slide continuously), showing a period for which, by
definition,
we
are likely to be able to make a continuous streak, and a horizontal
line,
intersecting it, at the median value we have accumulated in such a
period.

All of this is mostly number crunching; user-visible changes are an
extra
tab in the goal settings containing a simple html table, and two
extra
(optional) lines in the graph. So implementing it might be a
low-effort,
high-payout effort, assuming the encouragement benefits of
“achievements”
indeed as positive as I suspect. We could even go wild and show some
of
those stats as badges in the userpage, as seems to be quite popular
in
many
sites lately :stuck_out_tongue:

So, what do you think?

Waldir

On Saturday, July 13, 2013 1:40:21 PM UTC+1, Brian F Crain wrote:

I, too, remember reading lots of articles about these 66 days
supposedly
supported by ‘studies’. In reality that’s one single study and if I
remember
correctly even there they found significant differences between
different
habits.

So I don’t think there is real scientific evidence supporting 66
days,
30
days or any other number at this point.

On Fri, Jul 12, 2013 at 8:13 PM, Daniel Reeves dre...@beeminder.com

wrote:

Waldir, love the way you’re thinking and especially love Essy for
providing an immediate (though perhaps partial and inconvenient)
solution. Eager to hear if you find that Essy’s solution suffices.
I
guess the main thing it’s missing is displaying previously reached
checkpoints, but I’m not convinced that’s a good idea. It’s more
complexity in the graph and just showing that you’ve been on the
yellow brick road every day is arguably even better than showing
that
you reached specific past checkpoints. Besides, part of the point
of
Beeminder is that every single day is such a checkpoint.

On Fri, Jul 12, 2013 at 8:33 AM, Essentiae esse...@gmail.com
wrote:

What about using the “tmax” option (Upper bound on the plot range
for
the
x-axis (latest date)") in the “Advanced” tab and setting it to 21
or
66
days
from your start point? Once you reach it, you could change the
date
to
your
next milestone, or delete it altogether. That way you get to see
the
distance between your latest point and your milestone, as well as
how
high/low your yellow brick road will be by then.

Essy

On Friday, July 12, 2013 11:04:14 AM UTC-4, Waldir Pimenta wrote:

I think most of us have already heard of the 21-day period (or
other
similar lengths) purportedly needed to establish a new habit.
Google’s
Matt
Cutts has a nice and short TED talk about his 30-day challenge
system[1].

Today as I looked into my graph after inputting some data, I
realized
that
the fact the akrasia horizon is a moving “target” (in quotes
because
it’s
not a target at all; in fact, the definition of a horizon
implies
that
it is
always out of reach) makes me a little anxious and I would
instead
like to
see a fixed goal I could aim to (sort of a horizon in the sense
used
in the
term “event horizon” for black holes, i.e. something you can
actually
cross). Something like a “habit threshold”.

I decided to post here asking you about it, and having heard of
the
21-day
thing, I went to look for a good link to add to my post.
Interestingly
(put
perhaps not surprisingly), I found out that the 21-day
habit-forming
concept
is kind of a myth. There are some interesting articles on this
[2][3].
However, these articles mention how studies have revealed that
while
habit-forming depends, naturally, on the habit and on the
person,
the
average period they recorded was 66 days. Also, they found that
missing a
day or two in that stretch wasn’t critical and didn’t require
counting
to
start over.

With that said, I would suggest a new optional feature: the
ability
to
display the next “habit threshold” on the graph, based on the
description
above, as well as the location of such thresholds crossed in the
past.
This
would provide a shorter-term sub-goal (depending on the goal
length,
of
course), sort of how checkpoints work in videogames, thus making
it
easier
to muster the discipline to stick through especially when close
to
them; I
personally find that, having set a year-long goal, I sometimes
feel
a
little
bit anxious looking at the road and seeing it as never-ending
stairs.
Plus,
past “checkpoints” would give us pride and confidence that we
can
do
it —
yes, the need for self-discipline and keeping good habits is a
life-long
struggle, and it makes sense to represent it as a continually
increasing/decreasing line (depending on the goal, of course!),
but
a
little
encouragement may go a long way in helping us stick with our
improvement
commitments — in this case, by visualizing our goals closer to
the
present
and act on them.

What do you think?

[1] http://www.youtube.com/watch?v=JnfBXjWm7hc
[2]

This column will change your life: How long does it really take to change a habit? | Life and style | The Guardian

[3]

http://lifehacker.com/5926583/why-habits-arent-actually-formed-in-21-days


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+u...@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


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+u...@googlegroups.com.

For more options, visit https://groups.google.com/groups/opt_out.


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+u...@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


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+u...@googlegroups.com <javascript:>.

For more options, visit https://groups.google.com/groups/opt_out.


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

I have previously used the end date in the road dial to help establish new
goals and habits. I hadn’t thought of the visible target as being a
habit-forming milestone, but that’s effectively what it was.

It’s also convenient to set a slope ‘just for the next 30 days’ and to have
a scheduled date for re-evaluating that intention. You can approximate that
with the ‘take a break’ feature (which can be used to increase slope too!),
but it lacks the hard deadline of a goal end date, or the visible sign of
the approaching target.

I have sometimes used Essy’s approach of setting tmax, particularly for
conferences and other real-world hard deadlines.

What’s less convenient is that Beeminder goals are geared towards being
infinite rather than ever coming to an end, so recommitting to those goals
is occasionally not very smooth. (I think we’ve fixed most of that now, but
if there’s weirdness, email support.)

Philip

On Wednesday, September 11, 2013 10:38:45 AM UTC+1, Waldir Pimenta wrote:

That’s flattering, but one of the reasons I need beeminder in the first
place is my tendency for getting involved in side projects and losing focus
on my main work. :slight_smile:
I do have the trello board bookmarked and will check it out occasionally,
but for now I can only guarantee (semi-regular) participation in this
mailing list.
If you still think it could be a good idea to add me, my account is linked
to the address wal...@email.com <javascript:>.

On Tuesday, September 10, 2013 2:53:26 AM UTC+1, Daniel Reeves wrote:

That looks nice! Thanks Waldir! Possibly we should save this for when
we have javascript graphs with explanatory mouseovers and whatnot. If
you have a trello account and are game, could I add you to our trello
board at Trello ?

On Mon, Sep 9, 2013 at 6:40 AM, Waldir Pimenta waldir....@gmail.com
wrote:

I’ve been thinking some more about this, and I would streamline my
proposal
down to two things:

  1. In the graph, display a “potential data point” located at the value
    [previous day + median value of positive (in a “do more” goal) data
    points].
    This would be aimed at encouraging immediate action, showing a
    perfectly
    achievable target that we can reach on that day. Additionally, there
    could
    be another target, above it, located at [previous day + maximum daily
    value], which is also a goal that is proven to be achievable, although
    harder. I was thinking something like this:
    http://i.imgur.com/iGM35SR.png

  2. In the “goal stats” box in the right sidebar of each goal’s page,
    display
    3 badges: “best day”, “best ” (depending on the goal’s rate
    units;
    e.g. week), and “longest run”, defined as a continually improving
    streak
    (i.e., excluding flat or negative values, but allowing one or two such
    days
    in the middle of a streak, per the reports of the studies I mentioned
    earlier in this thread: “they found that missing a day or two in that
    stretch wasn’t critical and didn’t require counting to start over.”).
    This
    would be aimed at increasing confidence in our overall ability to
    improve.

Any thoughts?

On Friday, September 6, 2013 10:08:20 PM UTC+1, Daniel Reeves wrote:

Mulling this; thanks Waldir! I don’t want to just throw spaghetti at
the wall though (eg, “longest in-the-road streak” doesn’t strike me as
an important stat to highlight, though we do have a similar statistic
in the legit check emails) so what would be especially great is if you
can identify specific stats that you really want (like you personally)
and add them individually on uservoice.beeminder.com so people can
vote on them.

I do think we should highlight your current actual (historical
average) rate, like what you could’ve had the yellow brick road dialed
to, but we’re brewing something else for that…

On Fri, Sep 6, 2013 at 8:40 AM, Waldir Pimenta waldir....@gmail.com
wrote:

Sorry, guys, for the long delay. I think some good points have been
raised.

First of all, I wouldn’t consider Essy’s suggestion a solution –
perhaps a
workaround, but a quite inconvenient one at that, since we’d have to
continually adjust our graphs. I have (and I believe many chronic
procrastinators have as well) been a victim of overplanning and
overplaying
with my task sceduling tools in the past, and I don’t think
Beeminder
should
encourage that.

Second, I agree that displaying past “checkpoints” in the graph will
clutter
it unnecessarily. Still, I’m sure we can find a way to have
information
on
past progress available without such problems. For example, a site
called
boincstats.com (for the distributed computing platform) shows stats
such
as
most productive day, recent average progress, etc. Beeminder could
have
a
“longest in-the-road streak”, as well as stats similar to those I
mentioned.
For all of these, it would be easy to filter the values for time
period
(last week, last month, all time, etc). A simple stats table in the
settings
page (say, a new “stats” tab) could provide this. It should also be
fairly
easy to extend these stats from only showing maximum values to also
include
averages such as the median, mean or mode. But these are only
suggestions
for completeness; I’m aware that showing too much data to the user
may
compromise its potential for generating insight. Options to
alleviate
this
could be de-emphasising the extra data, perhaps by using a smaller
font,
fainter colors, or placing the values in a tooltip.

Finally, Brian’s remark combined with the fact that habit formation
periods
vary wildly due to many factors, would suggest that indeed showing a
fixed-length period as a target in the graph wouldn’t be ideal. But
the
measures I described above would likely be a good --dare I say,
better–
substitute. So a graph could have, say, a vertical line placed at
the
median
length of a streak (starting from the start of the current streak,
so it
doesn’t slide continuously), showing a period for which, by
definition,
we
are likely to be able to make a continuous streak, and a horizontal
line,
intersecting it, at the median value we have accumulated in such a
period.

All of this is mostly number crunching; user-visible changes are an
extra
tab in the goal settings containing a simple html table, and two
extra
(optional) lines in the graph. So implementing it might be a
low-effort,
high-payout effort, assuming the encouragement benefits of
“achievements”
indeed as positive as I suspect. We could even go wild and show some
of
those stats as badges in the userpage, as seems to be quite popular
in
many
sites lately :stuck_out_tongue:

So, what do you think?

Waldir

On Saturday, July 13, 2013 1:40:21 PM UTC+1, Brian F Crain wrote:

I, too, remember reading lots of articles about these 66 days
supposedly
supported by ‘studies’. In reality that’s one single study and if I
remember
correctly even there they found significant differences between
different
habits.

So I don’t think there is real scientific evidence supporting 66
days,
30
days or any other number at this point.

On Fri, Jul 12, 2013 at 8:13 PM, Daniel Reeves <
dre...@beeminder.com>
wrote:

Waldir, love the way you’re thinking and especially love Essy for
providing an immediate (though perhaps partial and inconvenient)
solution. Eager to hear if you find that Essy’s solution suffices.
I
guess the main thing it’s missing is displaying previously reached
checkpoints, but I’m not convinced that’s a good idea. It’s more
complexity in the graph and just showing that you’ve been on the
yellow brick road every day is arguably even better than showing
that
you reached specific past checkpoints. Besides, part of the point
of
Beeminder is that every single day is such a checkpoint.

On Fri, Jul 12, 2013 at 8:33 AM, Essentiae esse...@gmail.com
wrote:

What about using the “tmax” option (Upper bound on the plot
range
for
the
x-axis (latest date)") in the “Advanced” tab and setting it to
21 or
66
days
from your start point? Once you reach it, you could change the
date
to
your
next milestone, or delete it altogether. That way you get to see
the
distance between your latest point and your milestone, as well
as
how
high/low your yellow brick road will be by then.

Essy

On Friday, July 12, 2013 11:04:14 AM UTC-4, Waldir Pimenta
wrote:

I think most of us have already heard of the 21-day period (or
other
similar lengths) purportedly needed to establish a new habit.
Google’s
Matt
Cutts has a nice and short TED talk about his 30-day challenge
system[1].

Today as I looked into my graph after inputting some data, I
realized
that
the fact the akrasia horizon is a moving “target” (in quotes
because
it’s
not a target at all; in fact, the definition of a horizon
implies
that
it is
always out of reach) makes me a little anxious and I would
instead
like to
see a fixed goal I could aim to (sort of a horizon in the sense
used
in the
term “event horizon” for black holes, i.e. something you can
actually
cross). Something like a “habit threshold”.

I decided to post here asking you about it, and having heard of
the
21-day
thing, I went to look for a good link to add to my post.
Interestingly
(put
perhaps not surprisingly), I found out that the 21-day
habit-forming
concept
is kind of a myth. There are some interesting articles on this
[2][3].
However, these articles mention how studies have revealed that
while
habit-forming depends, naturally, on the habit and on the
person,
the
average period they recorded was 66 days. Also, they found that
missing a
day or two in that stretch wasn’t critical and didn’t require
counting
to
start over.

With that said, I would suggest a new optional feature: the
ability
to
display the next “habit threshold” on the graph, based on the
description
above, as well as the location of such thresholds crossed in
the
past.
This
would provide a shorter-term sub-goal (depending on the goal
length,
of
course), sort of how checkpoints work in videogames, thus
making it
easier
to muster the discipline to stick through especially when close
to
them; I
personally find that, having set a year-long goal, I sometimes
feel
a
little
bit anxious looking at the road and seeing it as never-ending
stairs.
Plus,
past “checkpoints” would give us pride and confidence that we
can
do
it —
yes, the need for self-discipline and keeping good habits is a
life-long
struggle, and it makes sense to represent it as a continually
increasing/decreasing line (depending on the goal, of course!),
but
a
little
encouragement may go a long way in helping us stick with our
improvement
commitments — in this case, by visualizing our goals closer to
the
present
and act on them.

What do you think?

[1] http://www.youtube.com/watch?v=JnfBXjWm7hc
[2]

This column will change your life: How long does it really take to change a habit? | Life and style | The Guardian

[3]

http://lifehacker.com/5926583/why-habits-arent-actually-formed-in-21-days


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+u...@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


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+u...@googlegroups.com.

For more options, visit https://groups.google.com/groups/opt_out.


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+u...@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


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+u...@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


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/d/optout.