WEBVTT
00:00.000 --> 00:13.920
And we are live. Hello again, everyone. It's been a while. I've been on break for the last
00:13.920 --> 00:19.000
one hour. It feels like I've been really far. I'm with David. Hi, David. How are you doing?
00:19.000 --> 00:23.360
Hi, Leo. I'm good. Thank you for having me on.
00:23.360 --> 00:26.480
So David, you do have the pad open in front of you.
00:26.480 --> 00:27.480
That's right.
00:27.480 --> 00:32.800
Yes. I think Sasha is putting up the pad for us so that they don't have to look at my very
00:32.800 --> 00:38.520
tiny face on BBB. I'm very small. I'm trying to push the boundaries. David, if you want
00:38.520 --> 00:44.320
to take some questions, I'm sorry, I've got hells whispering in my ears at the same time.
00:44.320 --> 00:49.800
Give me just a second. David, can you maybe read the first question and try to answer
00:49.800 --> 00:50.800
them?
00:50.800 --> 00:58.640
Sure. Question one. Excuse me. Do you use this just for yourself or do you use this
00:58.640 --> 01:04.160
to discuss or show with doctors or health professionals? The answer is it is useful
01:04.160 --> 01:10.880
for me and it's useful for health care providers. So it's not I'm not just able to see connections
01:10.880 --> 01:15.440
myself, even if we're not seeing a connection. If I'm talking to someone, if I'm talking
01:15.440 --> 01:20.640
to a clinician, I have brought this in on a tablet. If it's an in-person thing or on
01:20.640 --> 01:26.040
the phone, if it's a video appointment with someone, just send it in an email. It's very
01:26.040 --> 01:33.160
convenient to just send the graph around as a screenshot. So it's definitely been really
01:33.160 --> 01:40.920
useful to because one thing that I touch on in the talk is that let's say that we're talking
01:40.920 --> 01:45.880
about depression, for example. Let's say, oh, well, how have you been sleeping this
01:45.880 --> 01:49.440
week? Well, if you had objective information, that's perfect. Go ahead.
01:49.440 --> 01:53.840
Oh, sorry. I was mispressing my button for production. You're good.
01:53.840 --> 01:59.960
Oh, it just caught a little clip of sound. I see. Okay. Sorry. Yes. So where was I? I
01:59.960 --> 02:05.960
would say, okay, so on the one hand, having objective information is good not only for
02:05.960 --> 02:09.360
you, but when it's time to speak to someone and give them a picture of what's going on,
02:09.360 --> 02:14.800
you don't have to generalize. Even if you're having a good day, generalizing can be really
02:14.800 --> 02:19.300
like, oh, well, how are you sleeping? Oh, well, this day I woke up that, but which way
02:19.300 --> 02:22.720
was that Wednesday or Thursday? And then I remember this friend came over, so it must
02:22.720 --> 02:29.000
have been Friday. That's not I've discovered that actually just first thing in the morning
02:29.000 --> 02:33.720
waking up recording. How many pieces of gum did I have yesterday? How much did I sleep?
02:33.720 --> 02:40.000
Okay. So someone else, this segues right into the next question. How do you input the health
02:40.000 --> 02:46.120
data? Semi-automated with Org Mode capture templates, copy paste, automated with a smart
02:46.120 --> 02:51.080
watch, and if this, then that. Tasker, Org Mode document to automatically add stuff like
02:51.080 --> 02:57.040
sleeping data. Which parts are and are not automated? Okay. So it turns out I'm using
02:57.040 --> 03:03.320
an Org Mode capture template to automate the insertion of all the little category tags
03:03.320 --> 03:10.600
in the property drawer, but not the values. I'm not using a smart watch or anything whatsoever.
03:10.600 --> 03:17.560
I am just, I look at the clock when I go to bed and I look at the bed when I wake, look
03:17.560 --> 03:21.480
at the clock when I wake up. The very first thing I do is turn on my computer and turn
03:21.480 --> 03:30.160
on Emacs and enter that data. So, you know, it would be really cool if we could get some
03:30.160 --> 03:35.120
kind of data, you know, because if it's all, you know, if it's be able to be massaged into
03:35.120 --> 03:40.760
Org Mode format, it can then go into Org GNU plot. Or maybe you could cut Org Mode out
03:40.760 --> 03:45.600
of the equation if you find a way to go directly from that data to either CSV, which Org Mode,
03:45.600 --> 03:50.760
I'm sorry, which GNU plot can read, or, you know, whether you have to wear a little script
03:50.760 --> 03:56.560
in between. And the next question, how do you track the various health statistics that
03:56.560 --> 04:02.720
you're gathering? The ones, you know, for example, nicotine is an estimate. I'm pretty
04:02.720 --> 04:08.460
sure it was six. It was pretty sure it was seven. You know what I mean? It's pretty close
04:08.460 --> 04:12.820
because I do count and I think I could at most be off by one or two. So it's not fatal
04:12.820 --> 04:16.980
to the enterprise of collecting data. The same thing with sleep. I look at the clock,
04:16.980 --> 04:23.240
but I only count half hours. I say, oh, I slept seven hours, you know, seven and a half.
04:23.240 --> 04:28.280
It's very much a back of the napkin thing, but it's precise enough overall that as you
04:28.280 --> 04:33.240
can see from the data in the talk, you can see that my sleep declined pretty steadily
04:33.240 --> 04:38.160
over the course of that. So even if some of the values are off a little bit, you're still
04:38.160 --> 04:44.060
seeing trends and you can still see connections. So it's not perfect. I would love to have
04:44.060 --> 04:49.680
some type of more automated data, but all right. So I'm going to move on to the next
04:49.680 --> 04:56.280
question. It's possible to download data from Apple Watch's health app. Is it easy enough
04:56.280 --> 05:07.000
to incorporate those.csv files into your implementation of gnuplot? Okay. So I think
05:07.000 --> 05:10.680
what it would involve if you wanted to use my template generator, it would involve making
05:10.680 --> 05:17.880
your template, but then modifying the gnuplot script to read from that file instead of from
05:17.880 --> 05:27.200
an org mode table. So I think that what you would want to do is select the, yeah, I mean,
05:27.200 --> 05:33.680
and if you want, you know, I'll be around in the IRC after and I might even be able
05:33.680 --> 05:38.200
to look at their documentation or something, just even just give a peek at it to give a
05:38.200 --> 05:39.640
better answer to this question.
05:39.640 --> 05:45.640
Cool. And just to remind everyone, so, sorry, let me try to put up, okay. I did crush Dev
05:45.640 --> 05:49.400
earlier when trying to switch window. Oh, okay. I managed to do it. Yes, I still have
05:49.400 --> 05:58.240
it in me. So let me, slightly later, we will also be opening the discussion. So this BBB
05:58.240 --> 06:01.520
instance in which we are currently, we will be opening it so that people can come in with
06:01.520 --> 06:06.160
questions and maybe you're considering the personal nature of this talk, you know, maybe
06:06.160 --> 06:11.520
try to be a little wary of sharing personal information, especially if the stream is live,
06:11.520 --> 06:17.000
but otherwise you can have a chat with David, provide to David you're still available in,
06:17.000 --> 06:21.880
you know, 10, 20 minutes and feel free to come in and ask questions because really that's
06:21.880 --> 06:26.480
what Emacs Conf is about. It's about getting in touch with the speakers and this is the
06:26.480 --> 06:30.200
opportunity that you have to have a one-on-one with them rather than an asynchronous one
06:30.200 --> 06:34.120
with a pad. Sorry for the interruption. I just felt I remember. We're probably going
06:34.120 --> 06:38.080
to open the chat in about five to 10 minutes. We still have a lot of questions and I know
06:38.080 --> 06:42.280
I'd prefer if David focused on those questions. So please go ahead, David.
06:42.280 --> 06:48.800
Thank you so much, Leo. And yes, I am available for after, so I'll stick around. Okay. So
06:48.800 --> 06:54.120
regarding the medication tracking, you only have the option to record missed or not. If
06:54.120 --> 06:58.280
one needs to take multiple medications throughout the day, how would you propose to track that
06:58.280 --> 07:05.240
within GNU plot or separate? Okay. Yeah, as it is, I don't record which medication because
07:05.240 --> 07:12.600
it's the one I miss is typically just the same one. So the deal is, is that you would
07:12.600 --> 07:19.660
probably want to, you could count it as either doses or milligrams. The problem with tracking
07:19.660 --> 07:27.660
as milligrams, and yes, you would want to use separate variables for those. And probably
07:27.660 --> 07:36.080
what would make sense is either to record some lines as doses, meaning that one, whatever
07:36.080 --> 07:43.040
that ends up being, if it's a tablet or if it's the proper dose. And then that way, if
07:43.040 --> 07:47.100
one medication is late and another is not, or whatever you need to record, they're on
07:47.100 --> 07:54.440
separate lines, separate columns of the CSV file or I'm sorry, not CSV file of your org
07:54.440 --> 08:04.540
mode, separate property drawer entries in your capture template. I'm trying to think.
08:04.540 --> 08:08.520
It would be a little bit more work, but you could track that. The problem with tracking
08:08.520 --> 08:13.760
milligrams is that some medications, you would have a large range on the chart because some
08:13.760 --> 08:18.320
medications are a few milligrams and some medications are lots and lots for a dose.
08:18.320 --> 08:27.720
So tracking it as single doses might make sense. All right. So how's the workflow? This
08:27.720 --> 08:32.120
is the next question here. How's the workflow when working on the canoe plot code? Can you
08:32.120 --> 08:38.560
control C, control C and the SVG output on the right is updated automatically? The answer
08:38.560 --> 08:46.200
is yes. You can use something called auto revert tail mode. So you hit meta X to get
08:46.200 --> 08:53.040
the command and then auto revert tail mode. And that will cause the SVG file to automatically
08:53.040 --> 08:58.840
update every time you hit control C, control C on the canoe plot file, provided that you're
08:58.840 --> 09:07.680
viewing the same file that the canoe plot is going to overwrite. Let's see. Question.
09:07.680 --> 09:13.000
How much time does it take to process the amount of data that you add inside canoe Emacs?
09:13.000 --> 09:21.360
So at this point I have three or four months of data and it takes to update the whole thing,
09:21.360 --> 09:27.360
like, you know, get capture all the entries. It takes like a second or two. And this is
09:27.360 --> 09:31.200
a pretty decent PC. So we're not talking about something that takes a long time each day
09:31.200 --> 09:39.840
I updated. It's just, it's done pretty quick. All right. Next question. Will indent guide
09:39.840 --> 09:45.120
behave well with YAML files for Helm? I don't know the answer to that. Indent guide is really
09:45.120 --> 09:52.720
nice and it seems to behave pretty reasonably, but I wouldn't be able to speak to that.
09:52.720 --> 09:59.280
Okay. Just a quick interruption. So we have opened the chats now for people to join us
09:59.280 --> 10:04.360
on BBB. So you can go to the talk page on Emacs. You go to David's talk, which is called
10:04.360 --> 10:08.760
health and you should be able to join the BBB. So if you have questions that you'd like
10:08.760 --> 10:13.800
to ask with your voice to David, feel free to join us. And David, in the meantime, you
10:13.800 --> 10:16.960
can reply to the last question you have.
10:16.960 --> 10:22.880
Okay. Thank you. Great. The last question is, have you noticed your behavior changing
10:22.880 --> 10:29.000
as a result of tracking your data? And the answer is yes. The accountability has given,
10:29.000 --> 10:36.160
it draws habits, good or bad habits. It illustrates them in sharp relief. All of a sudden it's
10:36.160 --> 10:44.440
very clear what's going on, what's not going on. And it illuminates a sense of fog because
10:44.440 --> 10:49.080
it's hard to generalize when your awareness is changing over time. It's hard to generalize
10:49.080 --> 10:53.560
about what's going on each day and having the objective information. Yes, I'd say it
10:53.560 --> 10:57.080
has changed my behavior in a good way.
10:57.080 --> 11:04.240
I mean, that's a very good thing, I suppose. The whole point of tracking so much stuff,
11:04.240 --> 11:07.680
tracking people, it's an interesting angle that you have because you're creating data
11:07.680 --> 11:12.320
on a lot of health aspects. And people usually when they talk about using Emacs to better
11:12.320 --> 11:16.000
their life, they talk about all, they talk about getting things done. They talk about
11:16.000 --> 11:22.560
all the usual suspects when it comes to task organization. But plotting over time is something
11:22.560 --> 11:27.120
that is technically possible and that people talk about, but I feel like you've gone way
11:27.120 --> 11:32.720
far into this particular topic. So well done you. And you feel as a result you have a particular
11:32.720 --> 11:40.800
insight into seeing the curves evolve basically over time. And I think it provides very interesting
11:40.800 --> 11:46.760
data, something that is more easily understandable than just a list of to-dos in an agenda over
11:46.760 --> 11:47.760
time.
11:47.760 --> 11:54.320
Yeah. And what's nice is that I still can have to-dos and notes in that same file that
11:54.320 --> 12:00.600
produces my chart and that has all the data in it. And I keep that in version control,
12:00.600 --> 12:05.040
which stays in a private repo. And what's nice about that is everything's backed up.
12:05.040 --> 12:11.920
You know what I mean? Pardon me. I'm going to have a sip of water here.
12:11.920 --> 12:19.680
Please do. I mean, I'm actually down to termite. It's the good floral for thermos. And I've
12:19.680 --> 12:23.120
down two since we started EmacsCon today.
12:23.120 --> 12:30.680
Oh, wow. Yeah. And so I lost my train of thought.
12:30.680 --> 12:36.840
Take your time to find it. We still have about eight minutes of Q&A. By the way, I'll remind
12:36.840 --> 12:41.200
people we have opened the BBB room. So if you want to join us and ask live questions
12:41.200 --> 12:56.120
to David, feel free to do so. I'm sorry. Could you repeat that? I was asking you if you found
12:56.120 --> 12:59.000
the train of thought again.
12:59.000 --> 13:08.480
Hold on. Hold on. The last question, have I noticed behavior changing? Yeah. Then we
13:08.480 --> 13:12.400
talked about, oh, I talked about the fact that you can have to do the notes and appointments
13:12.400 --> 13:18.080
and all that kind of stuff. Like, for example, one of my next. Well, oh, now I remember the
13:18.080 --> 13:22.560
point that I was going to get into. The point is that once I have this routine of getting
13:22.560 --> 13:28.600
up every single morning and first thing, putting in my information. That habit, I feel like
13:28.600 --> 13:33.200
build on. For example, I added a new variable the other day. You know, I can add I thought
13:33.200 --> 13:37.920
of something. I'm like, hey, I should track that in my data. It reminds me I say, hey,
13:37.920 --> 13:41.080
there's not enough green triangles. I need to get some exercise this week. You know,
13:41.080 --> 13:48.520
the weather, I haven't wanted to get outside because the weather's been so dismal. So it's
13:48.520 --> 13:58.240
absolutely been a game changer. It really has been. And it doesn't have to be. Oh, go
13:58.240 --> 14:04.880
ahead. Go ahead. No, no, no. I was going to say that it's really good to hear that software
14:04.880 --> 14:10.080
can have such a potent impact, just the ability to track your data over time. Because the
14:10.080 --> 14:14.480
thing is, I think whatever you're feeling, you know, whenever you have mental ailments
14:14.480 --> 14:20.320
or whatnot, it's really hard to to have a healthy relationship to the chronology of
14:20.320 --> 14:26.360
feeling better. And it's easy to get lost into the immediate sensation. But I feel like
14:26.360 --> 14:30.120
being able to track over time that, you know, this daily checking in that you do in the
14:30.120 --> 14:34.760
morning of keeping track of your health data. I think it's the first step that shows you
14:34.760 --> 14:42.800
over time that, oh, yes, I am actually feeling better. And so as a result, being able to
14:42.800 --> 14:50.440
visually see, oh, yes, I am actually doing better. It must be reassuring, I assume.
14:50.440 --> 14:55.060
It is. I think that's a really interesting turn of phrase you had there earlier when
14:55.060 --> 15:02.840
you said it's harder to have a healthy relationship to the chronology of getting better. I think
15:02.840 --> 15:10.160
those were your exact words there. You know, the idea that, you know, if you've got to
15:10.160 --> 15:16.080
manage something, if you've got to roll with the punches, then seeing what you've accomplished,
15:16.080 --> 15:21.440
seeing yourself bounce back, you know, and then being able to see a trend, you might
15:21.440 --> 15:26.920
not notice unless you're tracking per the clock that sleep has started to decline and
15:26.920 --> 15:33.880
that it's time to start keeping your eye out. And it is reassuring, and you're right about
15:33.880 --> 15:37.920
that, you know. It's not always easy to be in the mindset of saying, hey, these lines
15:37.920 --> 15:39.960
move and they'll move again.
15:39.960 --> 15:46.560
Yeah, and I feel also it's kind of like having a scientific take on your well-being because
15:46.560 --> 15:53.960
you are parameter, okay, really tough word to say for the Frenchman that I am, but parameterizing
15:53.960 --> 15:58.720
the elements of your life, such as sleep, and we already have sleep is already more
15:58.720 --> 16:03.960
or less of a parameter in our lives. You know, we know, oh, you should sleep at least eight
16:03.960 --> 16:08.820
hours. You should have one hour, 30 blocks of sleep all the time. So this is a well parameterized
16:08.820 --> 16:13.960
element of our lives. But other elements that you're doing or the conflation of different
16:13.960 --> 16:19.360
curves feels like it's pointing out, it's providing more data and being able to see
16:19.360 --> 16:23.920
the trend, which is a key word in your presentation in what you do. I think that's really good.
16:23.920 --> 16:28.800
Chronology and trends and being able to see what works. I just have this example crossing
16:28.800 --> 16:33.360
my mind. Some people are plagued with canker sores, which are really nasty stuff that you
16:33.360 --> 16:38.040
get in mouth. But more often than not, it's really hard to track why you're getting them.
16:38.040 --> 16:42.600
A lot of people say it's due to stress or to poor sleep or to deficiency in iron or
16:42.600 --> 16:47.780
whatnot, or you're toothpaste. You might have understood that I actually suffer from them.
16:47.780 --> 16:52.240
So I actually know a lot of the reasons. But being able to track health data like this
16:52.240 --> 16:57.560
feels like it would be able to correlate the appearance of canker sores with maybe poor
16:57.560 --> 17:03.240
sleep or maybe changing medication or maybe stress at work or stuff like this. I really
17:03.240 --> 17:06.240
like this stance and I think you're really onto something here.
17:06.240 --> 17:12.960
Yeah. And if people wanted to chat about that more in terms of putting people's heads together,
17:12.960 --> 17:19.120
because let's say someone who has a more sophisticated medication tracking need, maybe they come
17:19.120 --> 17:23.760
up with a solution and they share their config. There's one thing I didn't talk about in the
17:23.760 --> 17:31.580
talk, which is I had one line where I have a formula instead, because Canoe can plot
17:31.580 --> 17:36.760
all kinds of formulas. So I have it plot an overall trouble score, which I did not show
17:36.760 --> 17:45.520
in the video, but I had it assigned three points for every hour above or below the goal
17:45.520 --> 17:49.360
of eight hours. So too much would give me lots of points on the trouble and too little
17:49.360 --> 17:54.880
will give me lots of points. So the relation to the goal or whatever, if stress is high,
17:54.880 --> 18:00.920
then the overall, if the anxiety line is high, then the overall trouble line should get high.
18:00.920 --> 18:06.800
And what I found is that that trouble line really did track with my recollection of
18:06.800 --> 18:12.560
how the last few weeks had gone. And so that trouble line is, you can see when it's trending
18:12.560 --> 18:21.600
even if it's not super obvious from the other lines. And maybe I'll share that too.
18:21.600 --> 18:28.400
All right, David, we are about at the end of time for the Q&A. Thank you so much for
18:28.400 --> 18:33.000
answering all the questions. I think it was very valuable to not only have your long presentation,
18:33.000 --> 18:37.080
which was very detailed, but also have this Q&A where you got the chance to answer one,
18:37.080 --> 18:41.240
two, three, four, five, more than seven questions, which is great.
18:41.240 --> 18:46.960
That's fantastic. And thank you so much for joining in and for being here. And thank you
18:46.960 --> 18:51.720
for having me at the conference. It's really amazing to be part of the community. And I'm
18:51.720 --> 18:56.720
happy to stick around. Do you want me to drop to IRC or just keep answering questions in
18:56.720 --> 19:01.500
voice? I don't think they're going to go on the pad, are they?
19:01.500 --> 19:05.760
So actually what we can do, we can leave the BBB room standing for now. If people want
19:05.760 --> 19:10.240
to join, they can do so. Maybe Corwin will come back to close the room afterwards. But
19:10.240 --> 19:13.140
for now, you can stay in the room. And if people want to unmute themselves after we
19:13.140 --> 19:15.800
go off stream, that would be good.
19:15.800 --> 19:18.720
So we are about to go into the next talk in about 30 seconds. Let me just talk to production
19:18.720 --> 19:28.360
real quick. So, yes, we are going to go into the next talk pretty soon. You should still
19:28.360 --> 19:35.000
be able to hear me, I think. Nope, they cannot. Okay. Well, David, I leave you in the room.
19:35.000 --> 19:41.360
I'll be in touch within five minutes to see if people are not showing up.
19:41.360 --> 19:42.360
Thank you, Leo.
19:42.360 --> 19:48.160
Yes, sorry. I'm dealing with many things at the same time. So I'm going to stop the recording.
19:48.160 --> 19:52.680
No, I'm going to leave the recording on. Is there anyone in the room that want to unmute?
19:52.680 --> 19:56.760
We are not on the stream anymore now. Does anyone want to say something or talk with
19:56.760 --> 20:04.260
David? Or do we want to close the room?
20:04.260 --> 20:09.440
Even if it's just by line. Okay, cool. So what are we going to do then? David, thank
20:09.440 --> 20:14.080
you so much for your time. We're actually going to close the room right now. I'm going
20:14.080 --> 20:35.200
to stop the recording.
