0:00

Welcome to Under the Radar a

0:02

show about independent iOS app development.

0:04

I'm Marco Arment. And I'm David

0:06

Smith. Under the Radar is usually

0:08

not longer than 30 minutes, so

0:10

let's get started. So recently, as

0:12

listeners of ATP might know, my

0:15

family purchased our favorite restaurants to

0:17

save it from possible worse fates.

0:19

And so we have been spending

0:21

a lot of time basically getting

0:23

that transferred over and getting it

0:25

ready and we are not. day-to-day

0:28

managers of it, fortunately the staff

0:30

will do that, but there has

0:32

been a lot of work for

0:34

us in kind of transition and

0:36

setup. And so as a result of

0:38

that, I have been doing almost no work

0:41

on overcast for maybe two months or

0:43

so. That has put me in kind

0:45

of an interesting position of

0:47

first of all, like kind of

0:49

every every day I kind of

0:51

just crossed my fingers and hoping

0:54

like, I hope nothing breaks today.

0:56

And I think there's also been

0:58

a little bit of kind of,

1:00

you know, guilt that I should

1:02

be working on it, but, you

1:04

know, that's maybe that's a topic

1:06

for a different episode. And don't

1:09

worry, everybody, I'm coming back to

1:11

it. It's just a question of

1:13

getting things, you know, finishing up

1:15

all the transfers and setup

1:17

processes so the staff can do

1:20

their jobs. self-sufficiency

1:22

in our products and our apps and

1:25

our services so that you know right

1:27

now the reason I've been able to

1:29

mostly not work on overcast for a

1:31

couple of months is the result of

1:34

having built a lot of

1:36

headroom into things having designed

1:38

things to not need frequent

1:40

or constant daily interaction and

1:42

for the very few things

1:44

that do need you know

1:46

regular interaction. making those things

1:48

as few and simple and

1:50

fast as possible. I have spanned

1:52

in my career lots of

1:55

different roles and product roles

1:57

from needing constant attention.

1:59

Like, you know, like at the peak

2:01

of Tumblr, like, well, most of my

2:03

time at Tumblr, I was needed to

2:06

be constantly on call because the growth

2:08

of the service was so strong and

2:10

we had so relatively limited resources and

2:12

staff that, like, I was constantly on

2:15

call to deal with servers that were

2:17

exceeding capacities and breaking and falling over

2:19

and failing and errors and things like

2:21

that. of that you know kind of

2:23

stress and and you know high stakes

2:26

environment and then into paper after that

2:28

was like a little bit less because

2:30

it was just way less traffic the

2:32

servers were still a little bit needy

2:35

but you know way less traffic and

2:37

then you know I had to often

2:39

deal with with the paper like the

2:41

article parser is now broken on some

2:44

popular website and I have to go

2:46

make a you know make a definition

2:48

or rule change to fix that on

2:50

some other website. So there was constant

2:52

need there and Instipaper I had a

2:55

lot of support needs as well. And

2:57

then Overcast, having learned from Tumblr and

2:59

then Instipaper, I designed a lot of

3:01

Overcast to not need nearly as much

3:04

manual interaction from me. And so there's

3:06

been different design decisions in place there,

3:08

different feature designs, different expectations, and just

3:10

a whole different kind of app. And

3:12

I kind of landed at a place

3:15

now where... I'm still running servers and

3:17

a decent amount of them. I think

3:19

I have 25 servers. I'm still running

3:21

servers, but things are much more stable

3:24

now than they used to be. So

3:26

I'm wondering, Dave, how have you found

3:28

your journey through that and what have

3:30

you found to be helpful there? Yeah,

3:32

because I think this is certainly something

3:35

that I have learned the hard way.

3:37

I think in some ways the same

3:39

as you where you, if you don't

3:41

think about... this kind of self-sufficiency or

3:44

the degree to which you are putting

3:46

yourself or some person specifically into an

3:48

essential role of your app, eventually that

3:50

is going to come back and bite

3:52

you because there will inevitably, and in

3:55

some ways it makes me think of

3:57

how like it is so easy in

3:59

the early days of designing and building

4:01

something to not think about the downsides

4:04

of success where it is, you know,

4:06

you start to think about like, oh,

4:08

well, I just wanted to succeed or

4:10

exist in the first place. I don't

4:12

think about, well, what if it exists

4:15

and exists 10 years into the future?

4:17

and is going to be something that

4:19

is going to, you know, is what

4:21

I'm doing sustainable? Like, maybe it will

4:24

be in the short term, but it

4:26

won't be in the long term. Like,

4:28

I think the most, the biggest place

4:30

I ran into this was when I

4:33

launched Feed Wrangler, gosh, that was 10,

4:35

15 years ago, a long time, which

4:37

was an RSS sinking service, and it's

4:39

like, I launched it and didn't necessarily,

4:41

Like it was initially built just in

4:44

a very basic simple way and it

4:46

was more successful than I thought it

4:48

would be, which was a great problem

4:50

to have, but it meant that I

4:53

had architected a lot of the back

4:55

end of that system to not, you

4:57

know, to be running close to its

4:59

thresholds and to be using things in

5:01

such a way that eventually, like as

5:04

you screw, the whole thing would start

5:06

to fall apart. And I was just

5:08

constantly in there getting notifications, you know,

5:10

this service is down or this aspect

5:13

of something. is above the CPU load

5:15

that I had an alert for. It

5:17

was just constantly, I was involved, I

5:19

was the linchpin, I was the thing

5:21

that was holding, like I was the

5:24

duct tape holding the whole system together.

5:26

And that was a terrible place to

5:28

be. That was not a great situation

5:30

because it was not sustainable because if

5:33

I. If it meant that I took

5:35

a couple of days away from the

5:37

system and something went wrong, very bad

5:39

things happened. And in the case of

5:41

Redrangle, there's sort of, I guess, now

5:44

a famous story where I went, you

5:46

know, spent a couple days in a

5:48

cabin in the woods and, you know,

5:50

10 minutes after I left cell phone

5:53

reception, it fell over. And for two

5:55

or three days, it was completely down.

5:57

and it started off people were grumpy

5:59

and then people were worried about my

6:01

safety and like it was a whole

6:04

situation right because I had built this

6:06

thing that was it didn't have margin

6:08

didn't have had room didn't have this

6:10

was able to be sustained by my

6:13

effort and that put me in a

6:15

bad position. And in the case of

6:17

Feed Wrangler, eventually I just solved it

6:19

with money where I went and went

6:21

from shared hosting to dedicated hosting that

6:24

was substantially more expensive but gave me

6:26

so much headroom that I didn't have

6:28

to be, you know, it's like my,

6:30

the servers could deal with my lousy

6:33

code in that way that made it

6:35

sustainable. But I think in the same

6:37

time on... iOS and on the apps

6:39

that I've built, there's an element where

6:42

I intentionally now try my best to

6:44

build and structure my apps so that

6:46

they don't rely on me for anything.

6:48

And this has been very, it was

6:50

very, you know, there's situations where like,

6:53

you know, when Widget Smith had its

6:55

moment and massively sort of had this

6:57

massive explosive growth in users, the, I

6:59

was incredibly glad that no part of

7:02

that. app had a server component at

7:04

that point. There was nothing it did

7:06

that required something other than locally on

7:08

the user's device. And if you can

7:10

push and maintain the user's device as

7:13

the extent of largely what you're doing.

7:15

there's a lot less things that can

7:17

go wrong. There's a lot of fewer

7:19

things that are going to come back

7:22

to bite you. But subsequently, as I've

7:24

been building out features or building things,

7:26

you inevitably find yourself in this place

7:28

where you're like, ooh, I could do

7:30

that, but that would require me to

7:33

build a server or to become responsible

7:35

for something or to deal with user

7:37

data in a way that is different.

7:39

And that is a tension. that now

7:42

I have this feeling of like if

7:44

it you know I'm pushing as much

7:46

logic as much thoughtfulness I can into

7:48

the client into the iPhone app itself

7:50

do as little as I can otherwise

7:53

because otherwise I do become essential. I

7:55

do have a role to play or

7:57

I have a responsibility to play. And

7:59

that is a dangerous thing to do

8:02

because life is going to change. Life

8:04

is going to come and you'll be

8:06

seasonal. Like right now you're in a

8:08

season where you are running a restaurant

8:10

and setting that up. And as you

8:13

say, that's not going to be probably

8:15

forever, but it's going to be for

8:17

some amount of time. I've definitely had

8:19

periods for health or personal or family

8:22

reasons where I have to step back

8:24

from work for extended periods for weeks

8:26

or a couple months at the longest

8:28

I think I've run into. And in

8:31

those periods, it's really awkward if you

8:33

can't go and leave because you're an

8:35

indie. Like, you don't have this big

8:37

staff or team who could just step

8:39

in and, you know, pull up the

8:42

slack. In theory, maybe you could hire

8:44

someone to be you, but like, that's

8:46

really awkward and complicated and difficult to

8:48

do, especially if the reason you're taking

8:51

this break is unplanned and awkward and

8:53

you're not in a great, you know,

8:55

physical and health, emotional, whatever state, like

8:57

there's reasons why. Planning for this ahead

8:59

of time can be really helpful. And

9:02

so yeah, it's like it is the

9:04

self-sufficiency aspect is something that I am

9:06

so grateful that past to me has

9:08

transitioned into this mindset that I want

9:11

to not be the linchpin or have

9:13

any person really be involved in the

9:15

day-to-day core functions of an app and

9:17

as a result, when things come up,

9:19

when I have to take step back,

9:22

it's usually... Totally fine. And there's nothing

9:24

that I need to worry about. I

9:26

can go off, you know, go to

9:28

the cabin in the woods for a

9:31

couple of days now and not be

9:33

worried that something is going to horribly

9:35

have, you know, blown up in the

9:37

meantime because that's just fundamentally how things

9:39

are engineered. And that took intentional choice

9:42

as a result to get to that

9:44

place. We are brought to you this

9:46

episode by Things. If you want to

9:48

achieve a goal, you've got to have

9:51

a plan. When it comes to making

9:53

a plan, there is no better tool

9:55

than the award-winning to do app Things.

9:57

The idea behind things is simple. You

9:59

create a project for each of your

10:02

goals, add the steps to reach those

10:04

goals, then schedule when you want to.

10:06

to work on them. Then each morning

10:08

when you wake up things is already

10:11

prepared your list of to-does for the

10:13

day. Just spend a few minutes reviewing

10:15

the list. put your to-do's in the

10:17

order you plan to do them and

10:19

then get on with your day. It

10:22

doesn't matter which device you're on, your

10:24

to-do's sink through the cloud, so they're

10:26

always with you, on your computer, in

10:28

your pocket, even on your wrist. You

10:31

can also connect your calendars to see

10:33

your events, scheduled to do so that

10:35

automatically repeat, you can write your notes

10:37

and things with Markdown on each item,

10:40

and so much more. of app made

10:42

to a very high standard of quality

10:44

and it's a two-time winner of Apple

10:46

Design Awards. So if you haven't tried

10:48

things you gotta check out the latest

10:51

version and I gotta say I fully

10:53

agree things is by far the nicest

10:55

to do app I've seen on Apple's

10:57

platforms. It's not even close it's by

11:00

far the best. So check out the

11:02

latest version of things if you haven't

11:04

tried it or haven't tried it recently.

11:06

You can download a free trial for

11:08

your Mac just to go to the

11:11

website at Things. That's T-H-I-N-G-S, dot app.

11:13

Exactly what you think. Things dot app.

11:15

You can also find on the app

11:17

store, just search for things. Whatever it

11:20

is you want to accomplish in life,

11:22

things can help you get there. Try

11:24

things today at things.app, or at the

11:26

link in our show notes, you will

11:28

not regret it. Our thanks to things

11:31

for their support of this show and

11:33

all of relay. So I think number

11:35

one. thing that you want to avoid

11:37

if you want your app to be

11:40

reasonably self-sufficient is editorial roles for yourself.

11:42

So this and this obviously varies based

11:44

on the type of app. So like

11:46

in my extreme case when I was

11:48

running the magazine, the entire product was

11:51

the editorial content that had to be

11:53

put out on a fixed interval. So

11:55

there was no way for me to

11:57

run that myself and have it be

12:00

any kind of self-sufficient and have it

12:02

tolerate any any temporary losses of my

12:04

time. and resources because the whole thing

12:06

was the actual content that had to

12:08

be put out every two weeks. You

12:11

know, in Overcast, what that could look

12:13

like would be things like featured podcasts

12:15

in the directory or any kind of

12:17

like, you know, this week we're featuring

12:20

this or here's a news post that

12:22

shows up in the app every two

12:24

weeks or whatever. anything like that I

12:26

have intentionally not implemented human editorial content

12:29

in overcast because if I did that

12:31

would require that I go and make

12:33

that or do that or process that

12:35

and that isn't just typing some things

12:37

into a text field necessarily like if

12:40

it's if it's podcast featuring, for instance,

12:42

for the editorial directory content, that could

12:44

require things like not only me finding

12:46

the right podcast to feature, but also

12:49

maybe like writing up a quick blurb

12:51

or getting a certain size image or

12:53

something like that, and all those things

12:55

would add up, and it's better off

12:57

if I don't rely on that kind

13:00

of thing, it would really have myself

13:02

for that kind of thing, because then

13:04

I would never be able to take

13:06

a week off from that. That's something

13:09

I'd have to have to do all

13:11

the time. Kind of an offshoot to

13:13

editorial is anything that is content in

13:15

your app that is written by humans

13:17

that becomes visible to other humans. And

13:20

of course that's not just writing, it

13:22

could be photos or whatever. And you

13:24

can think if there's any venue for

13:26

users of your app to write things

13:29

visible to other users, then you create

13:31

the possibility for spam and abuse and

13:33

copyright violation and all sorts of, you

13:35

know. problems that you'd have to monitor

13:37

and moderate and deal with and filter.

13:40

And so again, in Overcast, there is

13:42

no user review feature. I've intentionally left

13:44

that out. So there is no way

13:46

in Overcast for users to post or

13:49

create or write things to other users.

13:51

That way, I don't have to deal

13:53

with all of that. Because then there's

13:55

all sorts of complexity there. You have

13:57

to deal with not only what's decent

14:00

and ripe. and you have to deal

14:02

with what's legal, and you have to

14:04

deal with what's legal and decent and

14:06

right all over the world in all

14:09

different languages around the world that you

14:11

don't speak. So there's lots of potential

14:13

mess there if you can design your

14:15

app not to have the user-facing content

14:17

made by other users. And then you

14:20

start looking at other areas of

14:22

what can make an app sufficient

14:24

or not. Infrastructure is a big

14:26

one. And again, we're talking things

14:28

like servers like servers. infrastructure

14:30

can go into other areas too,

14:33

like where, if your app shows

14:35

content, where is that content coming

14:37

from? Does that require some kind

14:39

of maintenance or special casing ever?

14:41

You have to deal with that.

14:43

Then there's business roles of your

14:46

app. This is like, where does

14:48

your money come from or do

14:50

you have business relationships that you

14:52

need to maintain? In my case,

14:54

Overcast has ad banners for the free

14:56

version and I sell those ads on the

14:58

website. So I have set that up so

15:01

that every every new ad I have to

15:03

manually review and approve. So I have then

15:05

created a job for myself that like

15:07

when people buy an ad it does not

15:10

go live immediately. It goes into a

15:12

Q and I have to go a

15:14

couple times a day and go refresh

15:16

my admin page for that Q and

15:18

see if there's any ads there and

15:20

approve any that are there. If I'm really

15:22

busy, suppose, say I'm traveling with

15:24

family or something and I'm really

15:26

busy, I still have to do that. If I

15:29

don't do that, those ads will just

15:31

sit there and the advertisers might get

15:33

mad if they're sitting there for more

15:35

than 12 or 24 hours, like they

15:37

might start to get annoyed, like why

15:39

isn't my ad approved yet? And I

15:41

don't get money until it's approved also.

15:44

So there's, you know, I have created

15:46

a job for myself, but it's a

15:48

fairly simple one. you know, disconnected,

15:50

that I couldn't go do that at least once a

15:52

day, I would have to have somebody do it

15:54

for me. Like that's how important that is.

15:56

And then as you move on the ladder from that,

15:59

I think the next... up is support, like

16:01

user technical or whatever support. This is

16:03

probably the biggest one of these, but

16:05

I think it's the easiest to deal

16:07

with in the sense that like you

16:10

will get the most support, like if

16:12

you've designed your app to not have

16:14

editorial, to not have human to human

16:16

contact, to not have like tricky or

16:19

high maintenance infrastructure, to not have high

16:21

business development needs. Support will then be

16:23

the thing that you have to deal

16:25

with the most. And the good thing

16:28

is that support is also usually the

16:30

least pressing and the easiest to outsource.

16:32

So if you fall behind in your

16:34

support inbox by three or four days,

16:37

probably doesn't matter that much, unless you

16:39

have serious problems, you're probably okay with

16:41

that. It's also the easiest thing. If

16:43

you want to hire a service or

16:46

a person to cover your support for

16:48

you, that is not... I wouldn't say

16:50

it's easy in absolute terms. I've had

16:52

very mixed success there myself, but it

16:54

is the easiest to ask of all

16:57

these other things typically. So that's something

16:59

you can cover, you know, somewhat reasonably.

17:01

And you can also choose or design

17:03

your app in such a way that

17:06

it either doesn't need a lot of

17:08

support, hopefully, or in a slightly worse

17:10

case, you can do what I do

17:12

and set expectations to people that you

17:15

don't really provide support. and that most

17:17

emails will not be answered. And then

17:19

it becomes more of a kind of

17:21

feedback or user request channel, which is

17:24

very valuable in its own way, but

17:26

it's far less engaging and far less

17:28

demanding of your constant never-ending time, then

17:30

support. And then finally after all of

17:33

that, if you cover all those things,

17:35

then you have actually the code of

17:37

the app, the writing of the app,

17:39

bug fixes, updates. And that is... the

17:42

best thing to be the thing that's

17:44

fun on you. because that's the part

17:46

that you can do best and that

17:48

you can often do on your own

17:50

time and your own schedule and your

17:53

own priorities. So that part, like we

17:55

all have to deal with occasional bug

17:57

fixes, you know, but for the most

17:59

part, until the OS changes under our

18:02

feet, we can usually do that when

18:04

we are good and ready. And so,

18:06

you know, in my case with overcast.

18:08

I did a whole lot of work

18:11

on the app during all of last

18:13

year and the year before. So last

18:15

year was a massive overcast year. And

18:17

because I did all that work, I'm

18:20

able to take a breather now. Like

18:22

the app is in a pretty good

18:24

place now. You know, a lot of

18:26

a lot of the old problems of

18:29

the old app are gone. Most of

18:31

the problems with the new app are

18:33

resolved. And so I'm able to take.

18:35

a brief time off now for a

18:38

few months while I get all this

18:40

stuff sorted. And then I'll be able

18:42

to go back to the app on

18:44

my own time, most likely this summer,

18:47

and get back into the full bandwidth

18:49

swing of things with the actual bug

18:51

fixes and improvements and new features and

18:53

new designs and stuff like that. And

18:55

I think in that it's like you've

18:58

made those choices as you work your

19:00

way up that. And I think the

19:02

interesting aspect is, hey, that's a choice

19:04

you have to make. And I think

19:07

it's important to probably to understand that.

19:09

If at those, as you're working up

19:11

that hierarchy, if you have a project

19:13

or something in mind that requires you

19:16

to not, you know, to not decline

19:18

a feature or to deal with something,

19:20

like if you need user-generated content, if

19:22

you need to deal with moderation or

19:25

regular updates or business to business stuff,

19:27

like if that is essential for your

19:29

thing. In my moment, he's like, most

19:31

likely, it's good to be, keep in

19:34

mind, you're likely going to need a

19:36

staff. You're likely going to need people

19:38

to, or maybe not necessarily a staff,

19:40

but you are going to need outside

19:43

help to accomplish those things if they

19:45

become important. Because it is unlikely or

19:47

unwise, perhaps moreover. to think that you

19:49

would be able to just indefinitely manage

19:52

that yourself. It's like maybe you don't

19:54

need the outside support initially while you're

19:56

until the thing actually happens, but as

19:58

soon as it actually happens and you've

20:00

suddenly created this sort of situation that

20:03

you're existing, you may need that outside

20:05

help. You may need that extra thing.

20:07

And something like on the servers and

20:09

servers and services side, it's like something

20:12

I've gotten into a lot more recently

20:14

is like a lot of my services.

20:16

they rely on the small dedicated sort

20:18

of one-stop APIs that are necessary for

20:21

something. So like in, you know, Pidometer

20:23

Plus Plus, there's a thing that does

20:25

directions for hikes as you walk around.

20:27

And that was could have been as

20:30

I was deciding, it's like, could have

20:32

been as I was deciding, like, do

20:34

I want to build something like this

20:36

myself, you know, sort of explored, and

20:39

it's like technically possible, I could have

20:41

used open street map box, which is

20:43

a, like a dedicated API provider, who

20:45

that's what they specialize in. And it's

20:48

now become, in some ways, their problem,

20:50

that if their service goes down or

20:52

there's some data change or some aspect

20:54

of this needs to be updated, it's

20:57

like they have an expertise and a

20:59

specialty and a sort of obligation to

21:01

some degree to take care of that

21:03

for me. And I think that has

21:05

worked out really well for me in

21:08

a lot of ways where in some

21:10

of our things, like I think both

21:12

you and I, we tend on our,

21:14

the actual, you know, the, the, the

21:17

code at the bottom of our app,

21:19

we have a lot of tend to

21:21

exert a fair amount of control on.

21:23

We don't tend to use a lot

21:26

of third-party libraries. We tend to roll

21:28

a lot of things ourselves. And I

21:30

think that's, for both of us, has

21:32

kind of worked well insofar as we

21:35

have a deep understanding and control over

21:37

that part. isn't really ongoing in the

21:39

same way. It's ongoing in the sense

21:41

that we have to update our app

21:44

and do bug fixes, but it doesn't

21:46

have the day-to-day, hour-by-hour sense of that.

21:48

But as soon as it gets outside

21:50

of the iPhone itself, I look for

21:53

other ways that I can do this.

21:55

I can get external help. And last

21:57

year, or yeah, I've subsequently hired someone

21:59

who helps me with a lot of

22:01

the other things that I can't do

22:04

because some of these roles are essential

22:06

and are inevitable and are just part

22:08

of the deal. But it's important, I

22:10

think, to understand that. where you're going

22:13

to need external help, like that's going

22:15

to happen. If you don't think about

22:17

it ahead of time, it's like the

22:19

eventually the situation will present itself where

22:22

you will have to think about it.

22:24

But you can decide how you're going

22:26

to go. And it's like for me,

22:28

I spend money to avoid this. Maybe

22:31

it's another way, the important aspect of

22:33

this that like one of the things

22:35

that is easy for me to get

22:37

stuck sucked into in the early days

22:40

of being Indy, was that I was

22:42

very adverse to spending money on anything.

22:44

I would always think, well, I can

22:46

do it myself. I should just do

22:49

it myself. And while, like, there may

22:51

have been a very narrow brief window

22:53

where that was actually necessary in terms

22:55

of getting the business up and running,

22:58

that window was probably a matter of

23:00

few months, and I probably held onto

23:02

that mindset for a few years. that

23:04

I was holding on to this view

23:06

that I needed to be involved in

23:09

all aspects of things and didn't get

23:11

the help I needed or tried to

23:13

take on things that I probably shouldn't

23:15

have and not treated it like If

23:18

I can exchange money for my time,

23:20

that is usually a pretty good exchange.

23:22

That is usually a thing that is

23:24

going to come back to benefit me.

23:27

And in terms of making my apps

23:29

more sustainable, it's like as long as

23:31

the app is generating enough income to

23:33

do that thing, that to pull me

23:36

out of the critical path of something,

23:38

that's a great thing. And if it

23:40

doesn't, if the business doesn't generate that

23:42

income to provide that, you know, to

23:45

externalize that need, it's like... the business

23:47

isn't really a business. Like if it's

23:49

only held together by me being the

23:51

duct tape, it's not a business. It's

23:54

not sustainable and it's not going to

23:56

live up to the long-term turbulence of

23:58

life, I suppose. Yeah, I mean, and

24:00

that's, I think it's really important to

24:03

recognize. that like no matter what you

24:05

think when you're designing something when you're

24:07

you know in your 20s no matter

24:09

what you think You will have times

24:11

where you are needed to, like you

24:14

need to pull away from things, you

24:16

need to not be available for various

24:18

reasons. You know, maybe it's, you know,

24:20

a health thing or, you know, a

24:23

major family event or a loss or

24:25

just something, something really important is pulling

24:27

you away. And even, you know, look,

24:29

every night you have to sleep at

24:32

some point, you know, so every night,

24:34

like, there's, like, there's, like, there's small

24:36

scale of versions of this happening. possible

24:38

big scale versions of it happening you

24:41

know when life throws you a curveball

24:43

and so you should plan for this

24:45

whether you think you need it or

24:47

not and then it's a question of

24:50

just how much and on the other

24:52

side of that what I'm not saying

24:54

is that you can just set things

24:56

in motion and then just never think

24:59

about your business again like I'm definitely

25:01

not saying that you know like the

25:03

way the way I view a lot

25:05

of these these kind of you know

25:08

life curveballs is more like, you know,

25:10

you have like the short-term versions of

25:12

sleeping every night and maybe, you know,

25:14

having a weekend to yourself, but you

25:16

have longer-term versions and those are kind

25:19

of like sabbaticals or, you know, like

25:21

parental or, you know, maternity leave. It's

25:23

like, well, you know, you can take

25:25

leaves from your work here and there.

25:28

If you are always on leave for

25:30

many years, that's, you're not really doing

25:32

that job anymore. But you should build

25:34

your job in such a way that

25:37

you can take leave when you need

25:39

to and the whole thing doesn't fall

25:41

apart. And there's, you know, a lot

25:43

of distance between those two extremes. And

25:46

so, you know, if you are never

25:48

taking time off, that's not healthy for

25:50

you. And if you are always off,

25:52

that's not going to be healthy for

25:55

your business. But a lot of things

25:57

in the middle are just fine and

25:59

you can make... that work if you

26:01

think about it. Yeah, and a weird

26:04

way reminds me of, I had a

26:06

friend who used to work in finance,

26:08

like doing kind of stock trading and

26:10

all kinds of things, and his company

26:12

had a policy where every year he

26:15

had to take two weeks consecutive. leave

26:17

where they completely shut him out of

26:19

all of his work systems. That's awesome.

26:21

And for the company, it was a

26:24

sort of a safe, it's like this

26:26

way for them to safeguard against someone

26:28

who was doing dodgy trading or doing

26:30

things and was essentially like doing this

26:33

thing where they were you know wash

26:35

trading or finding ways to hide a

26:37

big loss or a big problem by

26:39

continuously doing things that you know sort

26:42

of hit it but if after you

26:44

know if most of those kinds of

26:46

schemes or trades or things would unravel

26:48

if they weren't sustained for in this

26:51

case, a two-week period. And it was

26:53

a way for them to safeguard against

26:55

that. And in some ways, that makes

26:57

me think of as a mental exercise

27:00

for, it's probably useful for us all

27:02

to have. It's like if we had

27:04

a two-week period where we were completely

27:06

disconnected from our work, What would suffer

27:09

as a result of that? What problems

27:11

would be revealed? In this case, it's

27:13

what are these, you know, maybe in

27:15

your case, like your ads wouldn't be

27:17

sold. And like that has a certain

27:20

cost and maybe that's okay, maybe that

27:22

isn't, but it's probably just a useful

27:24

mental exercise to think about. If I

27:26

completely logged out of all of my,

27:29

you know, all of my systems never

27:31

went, didn't check my email, didn't log

27:33

into my servers, didn't look at my

27:35

stats, all those kinds of things. What

27:38

would be the consequence for, you know,

27:40

I think two weeks is a reasonable

27:42

amount of time to think about where

27:44

so many things, you know, will probably

27:47

be fine for a couple of days,

27:49

maybe a week, after two weeks, maybe

27:51

it's sort of run into trouble. And

27:53

if you have things in your app

27:56

and your business in your situation that

27:58

couldn't survive two weeks of... being away,

28:00

that's probably something that's worth looking at

28:02

and understanding and making sure that you have

28:04

at least a vague notion of what you

28:07

could do to address that. Because sometimes that's

28:09

going to happen on a positive sense, you're

28:11

going to buy a restaurant, you're going to want

28:13

to set it up, it's going to be fun,

28:15

there's going to be a great reason. Sometimes

28:17

life is going to cause you to have

28:20

to do that. And so being ready for

28:22

it is just a wise move to do

28:24

when, you know, when the sun is shining

28:26

and things are straightforward is much better when,

28:28

you know, you're in the middle of the

28:30

hurricane and things aren't as great.

28:32

I love the idea of that as

28:34

like a safeguard against, you know, fraud

28:36

in the financial industry, but also like

28:38

showing you, like, are you irreplaceable?

28:40

And if you design something like that,

28:43

you're also making sure... like in the

28:45

case of your friend's business there,

28:47

they're also making sure nobody's irreplaceable.

28:50

Like that, you know, if somebody got

28:52

hit by a bus, like no one

28:54

is irreplaceable because if their job just

28:57

doesn't get done for a while, the

28:59

business can continue and, you know, the

29:01

system will not fail. And that's a

29:03

hard thing to design for when

29:05

you are a one-person business, but

29:07

it's not totally impossible. So I

29:09

like the idea of challenging yourself

29:11

to start thinking like... If I

29:13

had to drop everything with no

29:15

preparation, if I just got pulled away, maybe

29:17

I was knocked unconscious for two weeks. So

29:19

I had no time to set things up

29:22

or change things. What would happen? And

29:24

how can I prevent the bad outcomes

29:26

there? Yeah, and hopefully it never happens,

29:28

but if you're ready for it, then it

29:30

won't matter if it does. So yeah, just one

29:32

of these things to keep in the back

29:34

of our mind, especially in perhaps the quieter time

29:36

that we're in before, you know, the summer

29:39

business, it's like maybe it's a good time

29:41

to think about how we can make ourselves, you

29:43

know, a bit less essential and a bit

29:45

more replaceable in that way, I suppose. Thanks for

29:47

listening, everybody. And we'll talk to you

29:49

in two weeks. Bye.