Latest / The Pragmatic Engineer / The man behind the Big Tech comics – with Manu Cornet

Transcript

The man behind the Big Tech comics – with Manu Cornet · The Pragmatic Engineer

0:00
SPEAKER_0: Why do you think Google has such a problem naming things?
0:03
This is the bad consequence of a good company culture because if you are very
0:08
bottom up and employees,
0:10
engineers are really the powerful figures in the company.
0:13
People are going to start things and sometimes they'll start efforts that
0:18
will end up being competing.
0:20
When I first interviewed for Google and I asked my interviewer,
0:23
"What criticism would you say about Google or your job or what are some of the
0:27
drawbacks?"
0:28
And one of the guys that interviewed me and I asked him this question,
0:31
he said,
0:32
"Because it's so bottom up driven,
0:34
there ends up being some duplication.
0:36
SPEAKER_1: People who work on similar things or competing efforts."
0:39
SPEAKER_1: And that was in 2007, so it was even worse later on.
0:43
Have you ever seen this comic making fun of the org structure of Amazon,
0:46
Google,
0:47
Facebook,
0:48
Apple,
0:49
Oracle,
0:49
and Microsoft?
0:50
This comic was created more than 10 years ago,
0:52
and yet most of these are still spot on even today.
0:56
SPEAKER_0: But who created these and how are they so accurate and funny?
0:59
Today we reveal the cartoonists behind this comic and many other similar ones.
1:03
SPEAKER_0: He's Manu Cornet, who was a Google software engine for 14 years.
1:07
SPEAKER_0: He worked on Gmail, Android, Chrome, and Google Search.
1:11
Today we talk about the story behind the org chart comic and how Manu almost
1:15
did not publish it.
1:16
Lots and lots of cartoons about Google,
1:18
but we also dig behind the meaning of them,
1:20
like why Google is so bad at naming products.
1:23
We look at our cartoon where Manu correctly predicted Google's cloud gaming
1:26
console Stadia would eventually be killed,
1:28
and he did this on the date that Google launched this console.
1:32
It's safe to say that Manu is the highest profile cartoonist in tech world,
1:35
and I hope you enjoy him sharing the story behind some of his most famous comics.
1:39
If you enjoy the show,
1:40
please subscribe to the podcast on any podcast platform on YouTube.
1:43
SPEAKER_0: So let's jump in.
1:45
SPEAKER_0: Manu, welcome to the podcast.
1:47
SPEAKER_1: Thank you.
1:47
SPEAKER_1: Thank you for having me.
1:48
So one of your most referenced comics is the org structure one,
1:53
which I just sent over.
1:55
SPEAKER_0: Can we talk about when and how you created this and why?
1:59
SPEAKER_0: And it is been a very heavily referenced one.
2:02
I've spent the last 10 years trying to prove that I'm not just a one trick pony,
2:07
with this one cartoon that was successful.
2:11
So just for anyone listening,
2:13
this is in six panels of org charts of big tech companies.
2:20
SPEAKER_1: And it was published in 2011.
2:23
SPEAKER_1: So I almost didn't publish this because I didn't really find it funny.
2:30
It's always a big problem when you have an idea,
2:34
you find it funny on the moment and then you draw it and then you've spent time
2:38
on it.
2:39
SPEAKER_1: This one isn't too elaborate, so it didn't take too much time to draw.
2:43
Sometimes it takes a lot longer and by the time you're done drawing,
2:46
you have no idea whether it's funny or not.
2:50
SPEAKER_1: You've been staring at it for so long.
2:52
And for this one,
2:53
I didn't really find it very funny by the time it was done,
2:58
so I almost didn't publish it.
3:00
And the punchline,
3:02
which is the last company was supposed to be Oracle that has a large legal
3:08
department and a tiny engineering department.
3:12
SPEAKER_1: And initially, I had written engineering improperly.
3:15
SPEAKER_1: I had an E where the I should be somewhere.
3:18
SPEAKER_1: I fixed that later.
3:20
SPEAKER_1: I think it was at the time when Oracle and Google were battling on Java.
3:26
SPEAKER_1: So that was really...
3:28
SPEAKER_0: This was in 2011, 2010.
3:28
SPEAKER_0: So the one.
3:31
SPEAKER_1: Yeah.
3:31
SPEAKER_1: Yeah, so the drawing is from 2011.
3:34
SPEAKER_1: I'm pretty sure there were some losses going on at that time.
3:40
So Amazon was like the base case,
3:43
like a very standard tree-like structure,
3:47
very hierarchical.
3:49
The other ones were kind of easy to think about,
3:52
Apple being very centralized.
3:55
And initially,
3:56
I didn't have Microsoft,
3:58
but I couldn't really have only five companies because that would not have an
4:02
even number of panels.
4:04
SPEAKER_1: So I thought about a six one, which was Microsoft.
4:09
And I looked online on some forums on what the Microsoft culture was and I
4:16
heard people mentioning these battles between departments and that ended
4:22
up being the panel that caught people's attention more than the others.
4:30
This is old Microsoft,
4:30
but this comic was so talked about and I still see it referenced.
4:35
SPEAKER_0: I wonder if there's... Are you fine to do an updated version?
4:40
SPEAKER_0: Would you have inspiration?
4:42
SPEAKER_1: I guess I could.
4:43
I'm trying to look forward and not try to live off,
4:46
surf off previous waves.
4:50
But I know that it was mentioned by the CEO in his book without really
4:59
mentioning it by name or anything,
5:00
but he described the drawing pretty clearly on the first page of his book,
5:05
which it was a nice hat tip even without attribution.
5:10
SPEAKER_0: This episode is brought to you by WorkOS.
5:13
If you're building a SaaS app,
5:14
at some point your customers will start asking for enterprise features like
5:16
sample authentication,
5:18
skim provisioning,
5:19
and fine-grained authorization.
5:21
That's where WorkOS comes in,
5:22
making it fast and painless add enterprise features to your app.
5:26
Their APIs are easy to understand and you can ship quickly and get back to
5:29
building other features.
5:31
WorkOS also provides a free user management solution called AuthKit for up
5:34
to one million monthly active users.
5:37
It's a replacement for Auth0 and comes standard with useful features like
5:40
domain verification,
5:41
rule-based access control,
5:43
bot protection,
5:44
and MFA.
5:45
SPEAKER_0: It's powered by Radix Components, which means zero compromises in design.
5:49
You get limitless customizations as well as modular templates designed for
5:52
quick integrations.
5:53
Today,
5:54
hundreds of fast-growing startups are powered by WorkOS,
5:57
including ones you probably know like Cursor,
5:59
Vercel,
5:59
and Perplexity.
6:01
SPEAKER_0: Check it out at workos.com to learn more.
6:03
SPEAKER_0: That is workos.com.
6:07
This episode is brought to you by Formation,
6:08
a personalized interview prep program for experience software engineers.
6:12
Formation is built by ex-FAANG engineers who've conducted thousands of
6:16
interviews collectively so they know what it takes.
6:19
SPEAKER_0: In 2024, they helped engineers increase the comp by $110,000 on average.
6:24
They help you understand exactly how you stack up and then get you ready to ace
6:27
assistant design,
6:28
coding,
6:29
and behavioral interviews.
6:31
They assist in landing top tier roles and level up your composition through
6:34
unlimited career coaching and technical mentorship.
6:37
SPEAKER_0: Rigorous they drill problems that align with your personal skill gaps.
6:40
SPEAKER_0: Get unlimited mentorship to get unblocked.
6:43
SPEAKER_0: Do as many mocks as you need to ace the real thing.
6:45
Stop wasting time trying to figure out what to prep,
6:48
only to get to the interview and realize you had it wrong.
6:51
Formation helps put your interview prep on autopilot,
6:53
ensuring you're always focusing on the right thing.
6:56
SPEAKER_0: Apply today at fm.dev/pragmatic.
6:57
SPEAKER_0: That is fm.dev/pragmatic.
7:03
There's this comic that you also created also around probably 2011,
7:09
which is titled a much clear insight into who sues who.
7:14
And there's this big diagram of Apple,
7:18
Google,
7:19
Amazon,
7:19
Samsung,
7:20
HTC,
7:21
Microsoft with arrows,
7:22
one between the other and then on the bottom right corner,
7:28
there's Nokia by itself,
7:30
not sued by anyone,
7:31
not suing anyone.
7:31
SPEAKER_0: Just burning.
7:33
SPEAKER_1: Yes.
7:34
Yeah,
7:34
this was almost...
7:36
Yeah,
7:37
I guess it's not any better right now,
7:39
but it felt like it was a big mess of who was suing whom.
7:45
I don't really follow it closely any longer,
7:49
but I don't even remember this one so closely.
7:54
Oh,
7:54
I remember I drew a Starbucks on the side because I thought Oracle might be
7:59
suing Starbucks because of Java and the JavaBeans.
8:00
SPEAKER_1: Then Nokia was...
8:05
SPEAKER_0: This was the height where Oracle was suing everyone because of Java.
8:10
SPEAKER_1: Yes, exactly.
8:11
So I don't think they ever really sued Starbucks,
8:14
but I thought I would make that a thing.
8:18
And then yeah,
8:19
you're right,
8:19
Nokia was...
8:22
I think it made a bit of a comeback since then.
8:25
I mean,
8:26
it's not one of the most popular smartphone makers,
8:30
but it used to be really popular with feature phones and then it was really,
8:36
yeah,
8:37
as you said,
8:38
as it's in the drawing,
8:40
just by itself in its own corner,
8:42
almost sad that it's not being sued by anyone.
8:44
SPEAKER_1: He's like, "Yeah, I want to play too."
8:47
Well,
8:47
plus,
8:48
I don't know if you created this around that time,
8:50
but a famous Nokia's old CEO made this famous speech of Nokia being a burning platform.
8:56
So I don't know if it was because of that,
8:59
but you can associate it with that as well.
9:01
SPEAKER_0: It is burning by itself.
9:03
I don't remember the details,
9:05
but it might be very well be one of the inspirations for that.
9:09
SPEAKER_1: Yes.
9:13
And then we talked about is there a follow-up for the tech structure,
9:16
what companies are?
9:17
But recently,
9:18
I think only a few years ago you did this called Google versus Amazon,
9:23
which is not about all big tech companies,
9:26
but Google and Amazon,
9:27
and it was with guns and roses.
9:29
There's this circle with Google where there's guns held outside to
9:34
customers and roses to employees,
9:38
and Amazon is the opposite way where they hold roses to customers and guns
9:42
towards employees.
9:45
SPEAKER_1: Yes.
9:46
SPEAKER_1: This one didn't get shared quite so much as the initial org charts one.
9:55
I think it was a little bit more complex as in harder to parse,
10:01
like the org charts one,
10:02
you take a glance and it immediately is obvious what's happening.
10:07
For this one,
10:07
you have to read...
10:08
You have to notice that the customers and employees are in different places.
10:13
But yeah,
10:14
I do think it obviously oversimplifies which comics always often do.
10:22
But it does feel like Google has always...
10:25
Well,
10:26
maybe not right now,
10:27
I'm not sure,
10:28
but at least when I was there treating its employees insanely well and not
10:34
really caring that much about the customers because they're not the one
10:38
that's paying.
10:40
SPEAKER_1: Most people use back then Google products for free.
10:44
Now there's a little bit more of an enterprise business and cloud business,
10:48
but most people are still just users of things like Gmail or Search,
10:54
which are free products.
10:56
So Google has really no immediate incentive to be really nice to its customers.
11:02
I mean,
11:03
it does and it did,
11:05
and it still does focus on the user as in trying to build features that are
11:10
useful to the user and to make things nice and pretty and fast.
11:15
But if as a user you are unhappy and you're looking for customer service,
11:22
good luck ever reaching out and reaching people from Google to help you out.
11:30
Yeah,
11:30
in my experience,
11:31
I kept joking to people that as a business owner or as a startup,
11:38
if I'm buying from a company.
11:40
SPEAKER_0: The company that I trust the most is Amazon.
11:43
SPEAKER_0: If I have a problem, they bend over, they bend backwards.
11:48
SPEAKER_0: There's been so many stories.
11:49
An example of Steve Yegge did an interview with the pragmatic engineer,
11:55
and he was saying that...
11:56
He used to work at Amazon and then he went later to...
11:59
Well,
12:00
he worked at Google for a long time,
12:01
sorry,
12:02
Amazon and then Google for a very long time.
12:04
And then he went to Grab,
12:05
and when he worked at Grab,
12:08
they were on Amazon,
12:09
AWS,
12:11
they used services.
12:11
They were not the biggest customer by far,
12:14
but they were like know midsize or something like that.
12:16
And the Amazon team meeting wanted to meet them,
12:20
and he had the PMs for several products that they were using in the room,
12:26
in person meeting with them.
12:28
And he was just blinking why did they come here and they just wanted to talk to customers.
12:33
And again,
12:33
they were not the biggest customer and that's very specific to Amazon.
12:39
Whereas on the other hand,
12:39
for Google,
12:40
when I was researching,
12:41
for example,
12:42
on-call practices,
12:44
Amazon on-call is good luck.
12:45
SPEAKER_0: Amazon on-call you need to make sure your systems are there.
12:49
SPEAKER_0: It is very tough.
12:51
SPEAKER_0: If you have a small team, you still have an on-call.
12:53
SPEAKER_0: And Google on-call is probably as chill as it gets.
12:57
SPEAKER_0: They have a dedicated team that takes the majority of the on-call load.
13:02
SPEAKER_0: No other company does this because all the teams have it.
13:06
SPEAKER_0: They pay for on-call.
13:07
SPEAKER_1: All these are the SREs, right?
13:08
SPEAKER_0: Yeah, the SREs actually step in and take that, which is unheard of.
13:13
As an employee,
13:14
if you're looking for what is the company that pays the most,
13:18
and I have the most chill place to work until last year,
13:23
this was definitely Google or so.
13:25
As you said,
13:26
it might be changing a little bit,
13:27
butIthinkthiscomicisstillkindof-
13:31
It's interesting that the approach of really bending over backwards,
13:36
would be nice to customers also translates to company customers like B2B for
13:42
AWS and not just people who buy stuff on amazon.com.
13:47
SPEAKER_0: Was so surprised to learn.
13:48
SPEAKER_0: And this was Steve Yegge used to work at Amazon.
13:51
SPEAKER_0: He was surprised.
13:51
SPEAKER_0: He was like, "What?
13:52
SPEAKER_0: They're still doing it?"
13:56
I made another comic later on where showing how confusing it is to do anything
14:04
with Google products,
14:05
with all the different names,
14:07
and it shows a woman showing up in front of a Google employee or it's this one.
14:12
And she wants to buy something,
14:14
and then the Google guy keeps correcting her because then it's not the right
14:18
code name.
14:19
And then someone commented on this thing saying,
14:23
"Wait,
14:23
this shows a customer directly talking to a Google employee.
14:26
SPEAKER_1: That can't be right."
14:30
SPEAKER_0: Yeah.
14:31
There is this thing about,
14:34
I think still to this date,
14:36
people make fun of Google's naming and for some part,
14:40
rightfully so.
14:41
SPEAKER_0: I'll give you my example of Google.
14:44
When I worked at Uber and we were rolling out Apple Pay and an Android Pay.
14:51
So when I joined,
14:52
we had in 2016 or so there was Apple Pay and then there was Android Pay.
14:58
And then Google came back and said,
15:01
"Okay,
15:01
we're retiring Android Pay and it's going to be a new thing called Google
15:07
Wallet."
15:08
I think it was supposed to be called Google Wallet or something like that.
15:12
SPEAKER_0: And we're like, "Fine," or maybe Google Pay.
15:14
SPEAKER_0: I'm all confused.
15:15
SPEAKER_0: But it was interesting because it...
15:16
So in 2010 or 11,
15:18
when Uber started,
15:19
the first payment method was called Google Wallet,
15:22
then it became Android Pay.
15:24
SPEAKER_0: Then I think they said it is going to be Google Pay.
15:26
SPEAKER_0: Yes.
15:26
So it was Google Pay,
15:27
and then we agreed we were partners and we needed to upgrade to this new thing
15:32
and be part of the launch.
15:33
SPEAKER_0: And about two weeks before the launch, they changed the logo.
15:37
They send a brand new logo or a different logo to us,
15:40
and we're just thinking what's going on?
15:43
And then of course,
15:44
later,
15:44
I'm not sure,
15:46
I think Google Pay means something else.
15:47
SPEAKER_0: There's Google Wallet.
15:48
SPEAKER_0: You have a comic about this.
15:50
SPEAKER_1: Yeah, I have a comic about most things at Google.
15:54
SPEAKER_1: Even right now, there are two apps on Android that I have on my phone.
15:59
One is called GPay and the other one is called Google Pay,
16:04
maybe not anymore now.
16:05
At some point there was these two and one of them was functional and the other
16:09
ones was retired.
16:10
SPEAKER_1: So we were showing the deprecation process to users even then.
16:18
But then there's also Google Wallets and as you said,
16:21
Google Pay,
16:21
Android Pay,
16:23
and yeah.
16:24
This is another comic that I'm showing now,
16:26
which shows a double entry,
16:28
just a spreadsheet with rows being Google Chrome,
16:31
Android Nexus,
16:32
pixel play,
16:34
and of course the columns are [inaudible 00:16:39].
16:41
SPEAKER_0: This sounds very spot on.
16:42
SPEAKER_0: Now, jokes aside, you've actually been inside Google.
16:47
SPEAKER_0: Why do you think Google has such a problem naming things?
16:50
Because when I look at other companies,
16:52
they just seem to stick with it,
16:53
right?
16:53
SPEAKER_0: Look at Apple, they don't really change their naming.
16:57
I don't feel like I can appoint to any other large company that is so confusing.
17:01
Obviously we can always have funny stories about one product here or there,
17:04
but at this level it seems like a world record honestly.
17:09
So I think this is...
17:11
My interpretation,
17:12
and I don't have enough of a bird's eye view of things.
17:17
This is above my pay grade,
17:18
but my interpretation is that this is the bad consequence of a good company culture.
17:26
Because if you are very bottom up and employees,
17:32
engineers are really the powerful figures in the company.
17:36
People are going to start things and sometimes they'll start efforts that
17:42
will end up being competing.
17:45
And this is actually funnily enough,
17:47
when I first interviewed for Google for my job at Google in end of 2006,
17:54
early 2007.
17:56
Sometimes I like to be a bit sneaky and I ask my interviewer one day,
18:02
usually at the end of the interview,
18:04
they said,
18:04
"Doyouhaveanyquestionsforme?"
18:05
And I said,
18:06
"Yeah,
18:09
what criticism would you say about Google or your job or what are some of the
18:14
drawbacks?"
18:15
And one of the guys that interviewed me,
18:18
and I asked him this question,
18:20
he said,
18:21
"Because it's so bottom up driven,
18:24
there ends up being some duplication.
18:27
SPEAKER_1: People work on similar things or competing efforts."
18:30
SPEAKER_1: And even that was in 2007, so it was even worse later on.
18:34
SPEAKER_0: Wow.
18:34
SPEAKER_0: Yeah.
18:35
SPEAKER_0: That was so much smaller.
18:37
So,
18:37
that means that if you have different teams working on the same thing or
18:40
similar things,
18:41
there isn't that much top down pressure.
18:45
And this goes back to really being nice to employees and not so great about customers.
18:51
If you want to be really nice to employees and let people work on stuff that
18:55
matters to them,
18:56
you end up shipping your chart to the public and making consumers being a bit confused.
19:06
And I don't think it's much better now,
19:08
even though the company is very top down,
19:11
but maybe because the CEO isn't...
19:16
He's not like a Steve Jobs who would consolidate things.
19:20
He's more of an appeaser trying to arbitrate all the different efforts and
19:25
make people happy all at once.
19:30
SPEAKER_0: This is interesting though.
19:31
If we think about it,
19:32
I know we're looking at this the whole thing through the lens of humor,
19:35
but if you had to draw something about Apple,
19:37
it would probably a black box or just teams not talking about each other.
19:42
Because again,
19:42
one thing that we do know about Apple,
19:44
it's pretty well known.
19:45
I mean,
19:46
one day I'm going to do a deep dive on how Apple works,
19:49
but Apple is so damn secretive to the point of teams...
19:54
You work at a tech company,
19:45
99% of tech companies are 99.9%.
19:59
You work in a team,
20:01
you go have lunch,
20:02
you meet someone and you ask them,
20:03
"Hey,
20:04
whatareyouworkingon?"
20:04
And then they tell you,
20:05
you talk about it,
20:06
you might get some ideas,
20:07
you might help each other.
20:09
SPEAKER_0: And obviously everything you talk about stays in the company.
20:12
SPEAKER_0: This is without say.
20:13
At Apple,
20:14
you go and have lunch,
20:15
you ask someone,
20:15
what are you working on?
20:17
SPEAKER_0: And the first thing they're going to ask is, "Do you have clearance?"
20:20
They have a special word for...
20:22
And they're not going to talk to you because even though you're sitting next
20:25
to each other,
20:26
it's just a very different design.
20:28
So my point is just interesting how these trade-offs between how much
20:33
freedom do employees have?
20:37
Is the company prioritizing their needs versus how centralized things are?
20:45
SPEAKER_0: How much information flows?
20:46
SPEAKER_0: How much are we demanding from on-call so that we please our customers?
20:53
SPEAKER_0: They might be connected.
20:54
SPEAKER_0: You might not be able to have it all, right?
20:56
Yeah,
20:57
it's a difficult trade-off,
20:58
and I do think Google has been moving in the direction of where Apple is,
21:03
which is being more siloed and people not having access to other pieces of the
21:10
company or the code base.
21:12
SPEAKER_1: I think if I was to... I could update the Apple org chart.
21:16
I think it would look like fog of war in games where you're starting somewhere
21:22
and you're seeing what's around you,
21:24
but you don't really...
21:25
Everything else in the world in the map is blurred or just obscured.
21:31
I made fun of that movement of Google to move towards being more and more siloed,
21:37
but that also doesn't come free.
21:41
It's like,
21:41
okay,
21:42
this one is probably too long to read,
21:45
but it's especially ironical when Google's mission is supposed to organize
21:51
the world's information.
21:52
And when you're on the team,
21:53
you don't have access to the information of another team that could have
21:57
helped you,
21:59
but you need to prove that you need to know it in order to do your job.
22:03
And if you don't know it exists,
22:04
then you don't even know that you need to prove it.
22:08
SPEAKER_1: So anyway, I do think Google is moving towards that siloing as well.
22:15
SPEAKER_0: Yeah, well, it might be a necessity.
22:18
One other thing that is kind of,
22:20
I guess,
22:21
internally known about Google's engineering culture,
22:23
but externally might be a bit surprising with migrations,
22:26
and you did a comic about migrations.
22:30
SPEAKER_0: Can you explain what this is?
22:32
I didn't really mean it specifically about migrations,
22:35
but in general about competing efforts.
22:40
And it is the same thing that we were describing earlier with Google Pay
22:44
becoming GPay.
22:46
So what you see in the comic here is it's like a first person view of a car that is
22:53
encountering a choice between two roads.
22:58
SPEAKER_1: And then the top sign says, "Welcome to Google."
23:00
SPEAKER_1: And the right-hand side road says "Main road."
23:04
SPEAKER_1: And that's crossed out that says "Deprecated."
23:07
SPEAKER_1: And then don't even think about it.
23:09
And the road looks dirty and old,
23:13
and the right-hand side says,
23:14
"Newroadunderconstruction."
23:16
SPEAKER_1: And then there's a danger and then there's a construction problem.
23:20
This was initially,
23:23
I think this was initially a comment by the Zen CEO,
23:28
Eric Schmidt.
23:30
SPEAKER_1: I don't know if he's the one who said that, or maybe Johnson Rosenberg.
23:34
SPEAKER_1: I forgot.
23:35
One of the Google execs at that time who said,
23:38
"There's two ways to do things at Google.
23:40
SPEAKER_1: There's the deprecated way and there's the way that doesn't work yet."
23:45
SPEAKER_1: So this was an illustration of that line.
23:52
SPEAKER_1: And I do think that's actually quite relevant to lots of other companies.
24:00
When I was at Twitter,
24:01
people asked me to rebrand the welcome to Google to make it welcome to Twitter
24:06
because it was the same thing.
24:07
So I do think there's a tendency for...
24:11
Specifically for engineers,
24:12
software engineers,
24:14
but I think that might be even more general to try to make a new improved way,
24:20
and try to get people to use it even before it is ready to replace the old way.
24:27
To be fair,
24:28
I think you're right,
24:28
this is not specific to Google,
24:30
this is any midsize or larger tech company.
24:33
SPEAKER_0: It just keeps happening.
24:34
And after a while you go into...
24:37
There's even this sketch about microservices,
24:39
which the punchline is similar to this,
24:42
where one of the systems is not yet ready to handle something.
24:47
And again,
24:49
from the outside people working at startups or small teams,
24:53
they're kind of rolling their eyes,
24:54
"Howisthispossible?"
24:56
SPEAKER_0: And it is, you put together smart people and stuff like this happens.
25:00
SPEAKER_0: And obviously, everyone's surprised.
25:01
This is one of the reasons why projects often...
25:03
Not always,
25:04
but one of the reasons projects get delayed is,
25:06
"Oh yeah,
25:07
we thought this service will be ready and we could just use it.
25:10
Buttheexistingoneisbeingdeprecatedandthenewoneisnotyetready."
25:14
Another engineering culture related thing of course,
25:18
is if you're at a midsize or large company,
25:20
it's code reviews.
25:21
SPEAKER_0: And of course, you have a comic about code reviews as well.
25:24
SPEAKER_1: Yeah, I have several.
25:27
I have another comic where...
25:30
I can describe it,
25:31
where there's two people in two different panels.
25:33
There's the author of the change who says,
25:37
"Oh,
25:37
I don't need to be too cautious about my code.
25:40
SPEAKER_1: If there's some problem, the reviewer will catch it."
25:43
And the other panel is the reviewer who says,
25:45
"Oh,
25:45
I don't need to review this too carefully if the author must've know what he's
25:49
doingorwhatshe'sdoing."
25:53
SPEAKER_1: And yeah, same thing here.
25:54
It's like the scene is basically a table with two people standing up on the
26:05
table replacing a light bulb.
26:07
So this was a pun on the how many people does it take to screw in the light bulb,
26:12
which was maybe not quite as popular as the knock-knock jokes,
26:18
but similar,
26:20
you can have different version of it.
26:23
SPEAKER_1: So there's a software engineer who is trying to replace a light bulb.
26:28
And there's another software engineer also standing on the table because of
26:33
the lamp is really high up that says,
26:35
"Quote review," who is commenting on how to screw in the light bulb.
26:40
SPEAKER_1: And that must be another way.
26:43
And then there's a PM who is holding the light bulb as Shakespeare to be or not
26:49
to be holding a skull kind of thing.
26:53
And then there's the TLM,
26:55
the manager who is looking at his wristwatch because his is all taking too
26:59
long and we need to ship now.
27:01
And then there's the SRE,
27:04
site reliability engineer who is adding little pieces of wood to make the
27:09
table more stable.
27:11
And then there's a UX person who is bringing a light bulb with a hat and
27:16
decorations around it,
27:18
making things prettier.
27:19
SPEAKER_0: When you work at a team that warrants a TPM.
27:22
And also like SRE guess most companies will not have that,
27:25
but yeah,
27:26
that sounds about right.
27:28
SPEAKER_0: It's more complex than it looks.
27:30
SPEAKER_1: Yeah, of course.
27:31
SPEAKER_1: Of course.
27:32
These comics are always over simplifying and obviously the SRE has a beard
27:38
and long hair,
27:40
which was a good stereotype at that time.
27:46
So some of your comics,
27:47
they weren't just limited to making fun or pulling some jokes.
27:50
There was a comic that ended up being printed on every single door at Google,
27:54
as I understand.
27:56
SPEAKER_1: Yeah, that's right.
27:56
So at some point people noticed that things were being stolen from offices,
28:03
and that was because Google employees were letting people in behind them.
28:09
And I guess this being Disneyland or Wonderland,
28:12
because all the campuses are colorful and everything is free and nice and
28:18
people are nice to each other,
28:20
you don't really question who is coming after you and you just hold the door
28:23
for them.
28:25
And they ended up calling this tailgating,
28:27
even though I understand this is more of driving thing initially.
28:34
SPEAKER_1: So this first comic was a pun on this initial thing.
28:41
So it shows a crocodile or an alligator who is slithering its way into the
28:48
office through a half closed door that has Google written on it,
28:53
and it says,
28:53
"Beware the tailgator," with an O-R at the end.
28:59
SPEAKER_1: So that was the first one, and then that people liked it.
29:03
And because Google was trying to spread the word on increasing security,
29:09
and having people be aware that they should not let whoever is coming after
29:15
them into the building without making sure that they're actually an employee.
29:21
SPEAKER_1: They used this drawing initially.
29:23
And then because they wanted to keep people's attention,
29:27
they asked me for a refresh.
29:29
SPEAKER_1: Re-edition, yeah.
29:32
So they printed this and it was whenever I was visiting a Google office
29:38
elsewhere in the world,
29:39
there was the same...
29:40
My same drawing on all doors,
29:41
which was really cool to see.
29:44
But I ended up making...
29:46
Yeah,
29:47
every year I would make two or three versions and it was like trying to say the
29:54
same thing in different ways.
29:56
This episode is brought to you by Formation,
29:58
a personalized interview prep program for experienced software engineers.
30:02
Formation is built by ex-FAANG engineers who've conducted thousands of
30:05
interviews collectively,
30:06
so they know what it takes.
30:08
SPEAKER_0: In 2024, they helped engineers increase the comp by $110,000 on average.
30:13
They help you understand exactly how you stack up and then get you ready to ace
30:17
assistant design,
30:18
coding,
30:18
and behavioral interviews.
30:20
They assist in landing top-tier roles and level up your composition through
30:23
unlimited career coaching and technical mentorship,
30:26
rigorous they drill problems that align with your personal skill gaps.
30:30
Get unlimited mentorship to unblocked,
30:32
do as many mocks as you need to ace the real thing.
30:35
Stop wasting time trying to figure out what to prep,
30:37
only to get to the interview and realize you had it wrong.
30:40
Formation helps put your interview prep on autopilot,
30:43
ensuring you're always focusing on the right thing.
30:45
SPEAKER_0: Apply today at fm.dev/pragmatic, that is fm.dev/pragmatic.
30:53
SPEAKER_0: So 20% time, this is something when you joined Google, it existed, right?
30:58
SPEAKER_0: Can you tell us about what it was?
31:00
SPEAKER_0: How you saw people use it?
31:02
SPEAKER_0: What did you do in your 20% time?
31:04
SPEAKER_0: And how did it change over time?
31:06
SPEAKER_0: And obviously, you have a comic about the change as well.
31:10
SPEAKER_1: Yes.
31:12
SPEAKER_1: Yes.
31:12
So 20% was,
31:14
I think one of the things when you were back in maybe when I joined,
31:18
maybe in 2010 or so or a bit before when you were at Google,
31:22
these were among the perks that you would tell about maybe you would say
31:28
things about the free food and about you being able to walk up to a microphone
31:34
and ask questions to the CEO,
31:36
stuff like this.
31:37
And one of them was being allowed to dedicate 20% of your time to a project of
31:43
your choosing within reason.
31:46
You're not going to tell your boss that you're going to learn guitar one day
31:51
per week.
31:51
That's not relevant to your job,
31:53
but something that would be a reasonable project for you to work on.
31:58
SPEAKER_1: Something new.
31:59
And Google itself used to private self on having so many of their innovative
32:06
projects that were started as 20% projects.
32:10
SPEAKER_1: So as projects that were started that way, I think Gmail was one.
32:14
I think Google News was probably one,
32:18
and there's a bunch of other projects that were started that way.
32:24
But this is the part where I don't really understand where exactly the
32:30
pressure to become a more traditional company with more siloed people.
32:37
Where that comes into a clash with that 20% project thing,
32:44
I never really quite understood how that manifested itself.
32:47
So the drawing shows a big tree with one big branch labeled 20% projects,
32:56
and then there's a woman sitting on this branch and the woman has a T-shirt
33:01
labeled innovation.
33:02
Then there's a guy with a tuxedo and a necktie,
33:08
and smoking a cigar,
33:09
and sunglasses who is climbing on the ladder and starting to saw off that branch.
33:18
So my point was that really that was a source of innovation,
33:21
and if you cut those off,
33:24
if you don't encourage people to have those projects,
33:29
which was the case more and more.
33:31
Even though I don't remember it being called out explicitly,
33:36
that 20% projects are no longer a thing,
33:39
because that would be too obviously admitting that you're becoming a
33:43
traditional or conventional company.
33:47
But they were less and less encouraged and maybe more and more frowned upon.
33:52
SPEAKER_1: But so yeah, I didn't really quite understand.
33:56
SPEAKER_1: Maybe they didn't care about innovation that much.
33:59
SPEAKER_1: They just were happy with-
34:01
SPEAKER_1: Yes.
34:09
But one of the things outside,
34:11
again,
34:11
Google has a lot of perception,
34:13
but one of the perception is like,
34:14
oh,
34:14
if Google ships something,
34:16
they're going to shut it down.
34:17
And I do wonder,
34:19
just putting a little bit of devil's advocate in the sense that could it be
34:25
obviously 20% time we know brought a lot of really cool innovation with
34:29
Google famously,
34:32
Gmail apparently started...
34:33
Has a 20% project,
34:36
and there's going to be examples of other products that started off.
34:39
But having too many of those products might just mean that some of them
34:44
eventually will have no investment.
34:46
SPEAKER_0: They'll have to be shut down.
34:47
So I wonder if it is connected that Google has more products retired than any
34:51
other tech company,
34:52
and they're also the only company that allow 20% innovation time for many,
34:55
many years.
34:55
SPEAKER_0: I don't know.
34:55
SPEAKER_0: This is a question too.
34:56
SPEAKER_0: Do you think there's a connection that might be?
34:59
SPEAKER_1: Yeah, I do think there's a connection.
35:03
SPEAKER_1: That's a fair point.
35:04
SPEAKER_1: I would say then it should happen internally.
35:08
You should be throwing things at the wall and you should be experimenting
35:13
without necessarily launching stuff publicly.
35:17
SPEAKER_1: I think, if I knew how to run a company would be a lot richer.
35:24
SPEAKER_1: I wouldn't be here, I wouldn't be drawing comics.
35:29
But I do think it's good for companies to encourage innovation,
35:34
and I think 20% projects were a good way of doing this.
35:38
But you could let the projects select themselves more organically by either
35:48
having employees test them internally without necessarily launching them.
35:54
Gmail did something that was interesting at some point that I also worked on,
35:58
which was Gmail labs.
36:00
So you had a special section in your settings with these things called labs,
36:06
and it was by the name,
36:08
but also the labeling,
36:09
it was pretty obvious that these were not going to be officially supported
36:13
and they may go out at any point.
36:16
But that allowed people working on Gmail to launch with quotes,
36:21
"Launch" something that was with lower scrutiny.
36:27
So you knew,
36:28
I remember it was very clear that this is experimental,
36:31
it can go away like this,
36:32
use it at your own risk,
36:33
but we'd love your feedback.
36:35
And actually,
36:36
I remember as a user,
36:38
I thought that was fair because it was just super clear.
36:41
SPEAKER_0: This is alpha alpha.
36:43
SPEAKER_0: You get the latest cutting edge.
36:45
And I think they also made it clear that if it gets enough usage,
36:48
it might make,
36:48
right?
36:49
SPEAKER_0: No promises I think I have a point there that-
36:52
SPEAKER_1: I wrote, bunch of those-
36:53
SPEAKER_0: ...this probably could have been done differently.
36:57
Well,
36:57
so maybe Google could have some sort of Google Labs,
37:00
which they probably do,
37:01
but it probably mean something else to launch things that were not actual
37:07
products yet.
37:08
And to make the branding clear enough that people wouldn't complain if they
37:14
went away.
37:17
And this comic that you created,
37:18
we talked about that you must have a comic about the Google Graveyard,
37:21
which is also a website which has almost 300 products listed there now or so.
37:26
This seems oddly prophetic because you created this when Stadia was launched,
37:32
right?
37:32
SPEAKER_0: Yes, Google Stadia the games console where you could play remotely.
37:36
SPEAKER_0: It was a pretty cool concept.
37:38
SPEAKER_0: And then this comic shows a Google Graveyard and a man-
37:45
SPEAKER_1: So there's a sign saying "Welcome Stadia developers."
37:49
Because Google had to encourage people to actually develop for this platform,
37:54
make games for Stadia.
37:57
So this guy was waving towards the nice roads with rainbows and stars and a
38:04
unicorn showing,
38:07
"Look this way,
38:07
please."
38:08
But ignoring that on the side,
38:10
there's a huge graveyard of tombstones of all the dead Google products.
38:15
So developers had to focus on the nice things for Stadia,
38:21
and ignore all the risk involved with all the dead products.
38:26
SPEAKER_1: And yeah, Stadia eventually was also discontinued.
38:31
SPEAKER_1: I forgot exactly when.
38:33
SPEAKER_0: It came a little bit out of... No, it's hard to tell.
38:36
There's different analysis whether it actually had market fit or it wasn't
38:41
growing fast enough.
38:42
There's a lot of...
38:43
But it did take people by surprise and off Google did do one thing,
38:47
which they provided a full refund to anyone who asked as I understand.
38:52
So they ate the loss,
38:53
which was I guess a bit...
38:55
But still,
38:56
it just reinforces that it's hard to stress Google outside of ads products,
39:01
especially on areas that are not known for gaming.
39:05
SPEAKER_1: This was my interpretation of why Stadia failed.
39:09
So this comic is a guy with a t-shirt branded Stadia trying to launch
39:15
something by firing a big cannon.
39:18
And then just in front of a cannon,
39:20
there's a big wall labeled "Network effect" with a bunch of previously
39:25
failed cannonballs labeled Google Wave and Google Plus and Allo,
39:31
which was one of Google's instant messaging things.
39:34
That was my interpretation,
39:36
that you need critical mass,
39:39
you need some kind of network effect.
39:41
If your Google Plus,
39:43
even if your product is superior,
39:45
if everyone is already on Facebook,
39:47
then it's a lot harder.
39:50
SPEAKER_1: And I believe Stadia has a similar issue.
39:54
Everyone's on Steam and Steam isn't compatible with Stadia and developers
40:00
have to develop for your platform specifically.
40:03
SPEAKER_1: That's a-
40:04
It's interesting because even if you think outside of Google,
40:07
if you think about what are products that have become wildly successful that
40:12
are related to large companies.
40:14
SPEAKER_0: If I think of Meta, they bought them, right?
40:16
SPEAKER_0: Like Instagram.
40:18
SPEAKER_0: They bought a platform that was getting that network.
40:21
SPEAKER_0: WhatsApp, they also bought them.
40:23
And the only platform that you could argue has some level of social network or
40:30
used by hundreds of millions of consumers may be threads.
40:35
That one,
40:36
there's existing distribution that Facebook can use and there's arguments
40:40
of how successful it actually is compared to some of the other ones.
40:45
But it just shows how really difficult it is,
40:47
even for companies with billions of dollars to spend.
40:51
SPEAKER_0: Just crazy amounts.
40:54
SPEAKER_1: That's a comment that I want to make pretty soon if I have time.
40:58
I heard that Mark Zuckerberg was criticizing Apple saying that Apple hasn't
41:03
really innovated since they launched the iPhone.
41:06
SPEAKER_1: They made that innovation and then sat on it for 20 years.
41:10
So I want to show him saying that and then saying we on the other hand have been
41:15
acquiring innovating company every few years since then.
41:20
So not that they're being very innovative themselves,
41:23
but as you said,
41:24
they just buy them.
41:27
SPEAKER_1: Yes.
41:41
And obviously things happened afterwards there,
41:44
but when you joined,
41:45
can you share what did the culture feel like?
41:48
SPEAKER_0: This was before Elon Musk bought Twitter.
41:51
SPEAKER_0: How did it compare, especially for such a long time at Google?
41:59
Yeah,
42:00
I ended up,
42:01
for various reasons,
42:04
leaving Google,
42:05
even though it was a very nice,
42:08
stable job.
42:09
The company was just becoming less and less what it was in my naive dreams initially.
42:18
And Twitter was,
42:20
at the time,
42:21
it felt a lot like a young version of Google.
42:25
So it was very similar to what Google was when I joined,
42:31
smaller company less red tape friendly culture.
42:36
SPEAKER_1: Google was always friendly, but less siloing.
42:40
The problem was that in a capitalist society,
42:45
if you're trying to be the nice guy,
42:50
I think Twitter was,
42:52
but you don't have a huge amount of revenue,
42:56
then you are at a threat.
43:01
You're going to be threatened by all the sharks around,
43:04
including Elon Musk.
43:06
SPEAKER_1: And that happened.
43:07
Google had that nice endless cash cow of the ad revenue,
43:14
which meant that it could pursue a whole bunch of really interesting
43:19
projects even before they could be profitable.
43:22
Which Twitter never really had that luxury,
43:27
but culturally it felt similar.
43:30
And then you had this very interesting time where almost as soon as you joined,
43:34
or a few months after rumors started to happen that Elon Musk might buy Twitter,
43:42
and you started to make some comics about,
43:45
what I assume it must have felt inside the company.
43:47
What did it feel and how much did these comics actually convey and what it was
43:51
really like?
43:53
SPEAKER_1: Right?
43:54
SPEAKER_1: Yeah.
43:54
So I thought I had made a pretty smart move to Twitter,
43:57
but yeah,
43:58
six months after that.
44:01
So just to confirm,
44:02
did I say it correctly,
44:04
you went to Twitter thinking,
44:07
"Okay,
44:07
this is going to be..." You can rewind time a little bit,
44:09
go back to an earlier version of Google,
44:11
focus-build nice people,
44:14
chill,
44:15
not chill,
44:15
but focus and less corporate stuff,
44:16
right?
44:19
SPEAKER_1: Yeah, yeah, yeah.
44:20
SPEAKER_1: Chill in terms of a nice atmosphere, but not in terms of slacking off.
44:27
SPEAKER_1: Everyone was working pretty hard.
44:29
Well,
44:29
you actually were one of the...
44:32
Just speaking of the work that you did,
44:34
just a brief mention.
44:36
SPEAKER_0: So you worked on different web products.
44:39
One thing I heard from people who I talked with is that you were a very
44:42
productive developer in terms of building stuff,
44:47
fixing code,
44:47
those kind of things.
44:48
SPEAKER_0: Can you talk a little bit about your work?
44:51
Yeah,
44:51
so I worked mostly on the web clients as opposed to the iPhone app or Android app.
44:58
SPEAKER_1: So the website.
45:00
SPEAKER_1: I liked the product, I liked the team.
45:03
SPEAKER_1: That makes it a lot easier to be working hard and to be productive.
45:08
SPEAKER_1: The developer experience was pretty good, I would say.
45:16
When I left Google,
45:16
I was working on Google Search,
45:19
which was obviously the oldest part of the Google Code base and pretty hard to
45:26
work on.
45:27
SPEAKER_1: This is something that I also make fun of.
45:29
SPEAKER_1: This is what it felt like.
45:32
This is trying to ride a boat that looks like a house moving castle,
45:35
withlotsof-
45:42
SPEAKER_0: So that's what Google Search code base felt like.
45:44
SPEAKER_1: Yeah, that's Google search Code base.
45:45
SPEAKER_1: I mean, not to blame anyone, it's something-
45:48
SPEAKER_0: Yeah.
45:48
SPEAKER_0: Just what happens.
45:51
SPEAKER_1: ... any project that all would become like this, I think.
45:54
And the comic on this one was really more about leadership,
45:59
trying to blame it on poor people rowing for not moving fast enough.
46:05
But moving from this to Twitter,
46:07
it felt,
46:08
yeah,
46:09
the experience was a lot easier.
46:11
SPEAKER_1: You could make changes and iterate faster.
46:15
SPEAKER_1: The teams were awesome.
46:18
SPEAKER_1: And I was also becoming kind of a more senior person at that time.
46:23
So people also looked up to me in a way that increased my imposter syndrome.
46:31
But I tried to be hardworking and productive and a good co-worker and mentor.
46:39
SPEAKER_1: I liked that-
46:45
Yeah,
46:45
someone told me who I talked with Twitter that you were one of the most kind of
46:48
highest output web developers at web developers in terms of front end at Twitter.
46:50
SPEAKER_0: This just came after you're firing.
46:53
Someone just reflect like,
46:54
"Oh,
46:54
well,
46:54
I guess they're also getting rid of people who were actually pushing a lot of
46:59
codeandfixingalotofthings."
47:00
And also the one way we connected,
47:02
which we didn't even know each other is,
47:04
I complained on the web,
47:07
on Twitter web about a bug on the web client,
47:10
and you just messaged me saying,
47:12
"Oh,
47:14
can you help me reproduce it?
47:15
SPEAKER_0: And I'll try to fix it?"
47:16
Which I was really surprised about because Twitter at this time was already
47:20
pretty big,
47:21
and I never had anyone reach out to me about anything.
47:24
And then you actually went in debugged it,
47:28
and if I remember,
47:29
you might've fixed it.
47:30
SPEAKER_0: I am not sure about that part.
47:31
But you actually somehow had time to monitor social media for people's
47:37
problems and just go ahead and try to fix it.
47:39
SPEAKER_1: This is something I had tried doing, even at the Gmail time.
47:44
There were Gmail user forums where Gmail engineers would never,
47:50
never ever hang out unless they really were forced to by some social program
47:59
between engineers.
48:01
But this is something that I had always been a little bit frustrated at Google
48:06
and big companies in general.
48:08
You don't really get to meet your users,
48:11
which is for me,
48:13
it makes...
48:14
If you have a feedback loop that's just not closed at that time,
48:18
it's not a loop anymore.
48:20
If you are being told what to do by your team or management who have done a bunch
48:26
of studies and market research and they have metrics,
48:29
and I'm sure they know better on what people want,
48:33
but still,
48:34
it feels like when you're not connecting directly to users,
48:39
it feels off to me.
48:40
So I did a little bit of that on Gmail trying to ask for what people's problems
48:47
and feature requests were.
48:49
SPEAKER_1: And that led to a bunch of improvements.
48:52
And same thing with Twitter,
48:55
trying to see what the users were actually saying and trying to fix their issues.
49:02
So it was back to the news of Musk making an offer for Twitter,
49:08
and at that point,
49:08
no one knew if he would buy or not.
49:11
SPEAKER_0: It was not a binding offer.
49:12
SPEAKER_0: It was all sorts of noise.
49:14
SPEAKER_0: What was it like to work through those months of proper uncertainty?
49:24
SPEAKER_1: So this is what this comic is about.
49:26
SPEAKER_1: This is a bunch of other ones I'll show afterwards.
49:29
This one is really...
49:30
When you work on Twitter,
49:33
because of your job,
49:34
you stare at the news all day because even though you might be testing on
49:40
synthetic data or testing data,
49:42
usually you test things with actual feeds.
49:47
SPEAKER_1: So you see new stories coming in.
49:51
When Musk started trying to acquire Twitter,
49:55
then this was all over Twitter.
49:58
So even though as an employee you don't really know what's going to happen to you,
50:05
and Musk not having a great reputation for how he treated workers at Tesla,
50:12
for example.
50:13
You would have good reasons to be concerned,
50:16
but management Understandingly was trying to say,
50:20
"Hey,
50:20
don't worry about it.
50:22
We'll worry about when we get there and just tune out the noise,
50:27
and try to do your job," which is completely understandable,
50:31
but hard in this specific situation.
50:34
So the comic is really a...
50:38
It's like a sports field with a huge arena around it,
50:43
and apparently tens of thousands of people around it,
50:48
but there's no sport.
50:49
There's just a bunch of workers,
50:51
people sitting at their computer working,
50:55
and the maestro or whoever is the boss saying,
50:59
"Justtuneoutthenoise."
51:00
And everyone is around cheering and there's banners and there's even a
51:04
helicopter and airplanes and fireworks.
51:10
And I made a sign with the stock price at that time,
51:14
which was apparently $46 and 37 cents.
51:18
So it gets a bit hard to ignore the news when it's in your face and you are not
51:24
only watching the news,
51:25
but you're the object of that news.
51:30
SPEAKER_0: And then what Elon Musk did end up buying at Twitter.
51:32
SPEAKER_0: There was really massive firings.
51:34
I'm not sure there's anything comparable unless a company is shutting down
51:40
or just properly going out of business.
51:44
What are some comics that you drew to represent how you felt about this,
51:52
how people felt about it?
51:54
SPEAKER_1: So let's see.
51:55
SPEAKER_1: There's a bunch of those.
51:56
Because at least 50% of people were let go within a matter of weeks,
52:01
and then even more after.
52:06
SPEAKER_1: After.
52:06
SPEAKER_1: Yeah, I think it was 80%.
52:08
So initially there was a rumor that we had heard before he actually finalized
52:16
the deal that said something about 75%.
52:17
And then that prompted me to draw all these...
52:22
A bunch of references with these Damocles hanging on top of people's heads
52:29
saying 75% while you're trying to work and make your plans for next year.
52:36
SPEAKER_1: We all thought, "Oh, that can't be right.
52:38
SPEAKER_1: It can't be that high."
52:39
SPEAKER_1: Eventually it ended up being more.
52:42
SPEAKER_1: More like 80%.
52:43
Oh,
52:44
this is one about him visiting the office,
52:46
actually trying to find the best location.
52:52
SPEAKER_1: But yeah, it was pretty brutal.
52:57
So I do think that it was a time...
53:01
Just to be a bit cynical and not to be too dramatic,
53:06
it was a time when Silicon Valley in general had a whole bunch of layoffs.
53:11
Metra had some pretty big ones also,
53:15
and I do think Twitter had over hired for a little while,
53:23
and it was,
53:24
I do think firing some people,
53:28
or downsizing,
53:29
if you want to call it nicely,
53:30
was overdue at some point.
53:33
It may have been not possible because of the deal that was ongoing with Musk.
53:40
SPEAKER_1: So it froze everything and people couldn't get fired during that time.
53:45
So some layoffs were necessary,
53:47
but the way they were done was really not very nice to people at all.
53:53
So this comic here,
53:54
yeah,
53:56
it's a big dart board with the Twitter logo at the center of the circular dart
54:03
board and a big org chart of all people working on Twitter.
54:08
And then you see dozens of darts with the label fired,
54:13
red darts hitting all those.
54:16
SPEAKER_1: And let's see, the bubbles say, "Shall we stop now?"
54:20
SPEAKER_1: And another guy says, "Now this is fun."
54:23
SPEAKER_1: And then another guy says, "Your turn, Elon."
54:26
SPEAKER_1: So yeah, it felt random.
54:28
The reason I drew this is because,
54:30
well,
54:31
first everyone was getting fired,
54:32
but also it felt really random.
54:35
SPEAKER_1: I don't want to mention my own case.
54:39
I mean,
54:40
someone else who is on my team who was clearly the most productive guy I've
54:46
ever met in my career in terms of not just commitments and pushing code,
54:53
but also quality.
54:54
SPEAKER_1: He was really, really hardworking and he was fired.
54:58
SPEAKER_1: So that felt really random to me.
55:00
It really felt like,
55:02
okay,
55:02
I understand they need to fire a whole bunch of people,
55:05
but they're not doing it very nicely at all,
55:08
and they're maybe not firing the right people.
55:11
And he ended up being rehired and he accepted because he had a visa issue.
55:18
And so he really had a good reason to go back,
55:22
even though most people wouldn't go back being treated that way.
55:26
SPEAKER_0: Well, the visa is powerful enough.
55:28
So I mean,
55:29
it was clearly a very hard time both for Twitter obviously,
55:33
but also for the industry.
55:34
You did create a comic that I felt looked at it from a different angle,
55:39
and I heard that you mentioned that this was pretty well received by people.
55:45
SPEAKER_1: Yeah, so this was, let me see the dates on this one.
55:52
SPEAKER_1: 7th of November.
55:55
So yeah,
55:56
this shows playing on the sink idea of Elon Musk showing up with a big sink,
56:03
just for the sake of making a pun.
56:07
In the comic,
56:10
he is emptying a large can containing a bunch of small bluebirds into the sink,
56:19
and that's the metaphor for firing people and the bluebirds are the employees.
56:25
And then there's a hole in the pipe below the sink,
56:29
and then all the little birds,
56:31
they escape through that hole and they all fly away and they all look pretty happy.
56:35
So yeah,
56:38
it was trying to inject a bit of a happiness or just trying to say maybe we
56:47
dodged the bullet because maybe we wouldn't have been treated really well anyway.
56:53
I also spoke from a bit of a privileged position because I had a bunch of years
57:02
under my belt.
57:04
SPEAKER_1: I had some more financial stability, more seniority.
57:08
I didn't really have any health insurance issues or visa issues,
57:15
so I was really very lucky compared to other people.
57:18
So I felt like it was also my role to try and have some lighthearted comments as well.
57:27
SPEAKER_0: Yeah.
57:28
So I asked you earlier,
57:30
what's a comic that will be personal and maybe reflect on your professional career?
57:37
SPEAKER_0: And you pointed to this one.
57:41
What is this comic,
57:42
and how would you reflect now on your professional career of coming up closer
57:47
to 20 years?
57:49
Yeah,
57:51
I would say I've been insanely lucky in joining this industry at the right time.
57:59
SPEAKER_1: I hope it will get better again.
58:02
SPEAKER_1: I'm pretty sure it will.
58:05
SPEAKER_1: It seems it's a lot more rough right now than it was a few years ago.
58:11
SPEAKER_1: So initially I came from academia.
58:14
SPEAKER_1: I was trained to be a teacher or researcher.
58:20
SPEAKER_1: So it wasn't an obvious move for me to go into big tech companies.
58:28
Initially,
58:28
I was going to study physics or research physics or computer science or bioinformatics.
58:36
And I left my PhD program before I finished it because there was this company
58:41
called Google that looked pretty awesome.
58:44
It looked like it wasn't really just only after profits,
58:48
it was trying to be a good neighbor,
58:52
trying to be a good player in the world,
58:54
and making things to really for the betterment of society.
58:59
SPEAKER_1: Which was a bit of a naive view.
59:01
SPEAKER_1: But that was my view at the time.
59:04
And to be fair,
59:05
I think Google was quite close to that ideal view of such a company at the time.
59:13
And then it took a long time for Google to slowly move down into being more and
59:21
more of a traditional company,
59:26
even though the founder's letter initially said,
59:29
"We are not a..." What was the word?
59:33
SPEAKER_0: "We're not a traditional company."
59:34
SPEAKER_0: Or something like that or-
59:36
SPEAKER_1: It's probably another adjective.
59:38
SPEAKER_1: "We are not a traditional company and we don't intend to become one."
59:45
SPEAKER_1: I'll look for the word later.
59:46
So I do think there are a pretty traditional company now,
59:51
even though if you compare Google to most other companies,
59:54
it's still much better on average,
59:57
but it is not quite the naive ideal that I have.
1:00:02
But I still got away with working more than a decade for that company that was
1:00:08
really a dream job.
1:00:11
SPEAKER_1: And I was trying to find some of that in Twitter also.
1:00:14
So the comic shows,
1:00:18
two drunk characters on the bench sitting on a bench at nighttime in a park.
1:00:24
One of them is a human with a Google T-shirt,
1:00:28
and the other one is the Twitter blue bird.
1:00:33
And Google says,
1:00:35
the Google guy says,
1:00:36
"Itonlytookme10yearsbeforecapitalismkilledmyideals."
1:00:42
And there's a bunch of empty bottles of wine on the side,
1:00:45
and Twitter says,
1:00:46
"Good for you.
1:00:46
SPEAKER_1: It took me three months."
1:00:48
So it felt like a descent from the ideal company,
1:00:55
which I don't know if capitalism even makes this possible in general for such
1:01:00
a company to be nice with its employees.
1:01:03
And also it feels almost like an anomaly for it to have so much of a cash inflow,
1:01:10
and being able to do all those awesome things and trying to find some of that
1:01:19
also in Twitter.
1:01:20
SPEAKER_1: But the travel downwards was a lot faster.
1:01:29
SPEAKER_0: Yeah.
1:01:29
SPEAKER_0: Well, it sounds like it's been an interesting ride for you.
1:01:34
SPEAKER_1: Oh yeah, yeah, yeah, absolutely.
1:01:35
SPEAKER_1: And even though people were not treated really well at Twitter.
1:01:41
SPEAKER_1: Again, I was insanely lucky compared to others.
1:01:45
And the cartoonist side of myself was having a great time witnessing that.
1:01:54
Yeah,
1:01:54
and I got this book of yours called Twittoons where you actually summarize
1:02:00
the story of what happened to Twitter via comics.
1:02:03
It's actually very entertaining to read because the story as we know it
1:02:10
played out.
1:02:12
It sounds like it could be fiction,
1:02:16
but it's actually based on real events.
1:02:20
It does sound like it would be...
1:02:21
I almost want Netflix to make something out of this.
1:02:27
SPEAKER_0: We'll see if they do.
1:02:29
So to wrap up,
1:02:30
I just have a few rapid questions,
1:02:32
so I'll just shoot them out and then you answer whatever's on your mind.
1:02:37
SPEAKER_0: What is your favorite programming language and why?
1:02:40
SPEAKER_1: I don't have one.
1:02:43
SPEAKER_1: I feel like it is a means to an end.
1:02:46
SPEAKER_1: It depends on what I'm building.
1:02:48
Whenever I have interviews and people have this online coding platform,
1:02:52
before they paste in their templates of the interview question,
1:02:55
they say,
1:02:55
"Oh,
1:02:56
whichlanguageareyougoingtodo?"
1:02:57
And then I always say,
1:02:58
"Ithinktheonlysmartanswertothisisitdependsonyourquestion."
1:03:03
SPEAKER_1: So people are always a little bit confused.
1:03:06
SPEAKER_1: But yeah, I don't have a favorite language.
1:03:09
SPEAKER_1: I just use whatever is best suited for the job.
1:03:13
SPEAKER_0: Which languages have you used a lot in the past?
1:03:17
SPEAKER_1: I would say all the major ones.
1:03:19
So Java,
1:03:21
JavaScript,
1:03:22
Python,
1:03:24
C++,
1:03:25
Objective-C for Apple,
1:03:27
that would be more like Swift.
1:03:29
SPEAKER_1: I'm trying to learn Rust.
1:03:31
SPEAKER_1: I did a bunch of C also when I worked on lower level system stuff.
1:03:38
And then,
1:03:38
I don't know if you would call these programming languages,
1:03:41
but things that look like programming,
1:03:45
like HTML or CSS or even LaTeX.
1:03:48
SPEAKER_1: It's some kind of problem.
1:03:51
SPEAKER_0: Well, I like the right tool for the right job.
1:03:54
SPEAKER_1: Yeah.
1:03:55
SPEAKER_0: What's a book and a comic book that you liked and would recommend?
1:04:00
SPEAKER_1: I always have this issue of reverse indexing.
1:04:04
It's like if I'm looking at a bookshelf or a library,
1:04:10
I say,
1:04:11
"Oh,
1:04:11
I love this one.
1:04:12
SPEAKER_1: Oh, I love this one.
1:04:13
SPEAKER_1: Oh, I love this one."
1:04:13
SPEAKER_1: And then if you ask me, "Oh, what are your favorite ones?"
1:04:16
SPEAKER_1: I miss the reverse index.
1:04:19
SPEAKER_1: "Oh, what are my three favorites?"
1:04:20
SPEAKER_1: So actually I have a web page that helps me remembering.
1:04:25
SPEAKER_1: So it is ma.nu, which is my website /faves, as in favorites.
1:04:31
SPEAKER_0: Wonderful.
1:04:33
SPEAKER_1: So yeah, you could get some ideas from them.
1:04:35
For English speakers,
1:04:37
I would recommend the...
1:04:39
There's a comic book that is a series of books that is really famous in Europe,
1:04:44
but not much in the US,
1:04:45
but it's currently being translated,
1:04:47
I think pretty well.
1:04:49
SPEAKER_1: And the English title is Gomer Goof, so it's the name of a guy.
1:04:56
SPEAKER_1: His last name is Goof.
1:04:57
SPEAKER_1: So he goof all the time, and he is the anti-hero, and it's really funny.
1:05:04
SPEAKER_1: That would be my choice.
1:05:08
What is your favorite comic that you've created of all the ones,
1:05:11
or a favorite comic?
1:05:16
SPEAKER_1: It's like asking for the favorite child.
1:05:19
SPEAKER_1: I have one child, so that's an easy question.
1:05:25
SPEAKER_1: She's my favorite one.
1:05:28
SPEAKER_1: This one is what I answered earlier.
1:05:34
SPEAKER_1: Yeah, I think this one... Yeah, I could stand by this choice.
1:05:40
SPEAKER_1: My answer will probably depend on when you ask me.
1:05:42
SPEAKER_1: So this one is more about engineering.
1:05:44
SPEAKER_1: Do you want me to describe it?
1:05:47
SPEAKER_0: Yeah.
1:05:48
So this one has two panels,
1:05:50
and I think it's something software engineers can relate to.
1:05:54
The first panel is labeled,
1:05:55
The Life of a Software Engineer,
1:05:58
and it shows a guy in front of building that only has foundations and looks
1:06:06
nice and neat and clean.
1:06:07
And he has a hammer and a hand saw,
1:06:11
and there's a bunch of wood on the side.
1:06:14
SPEAKER_1: He says, "Clean slate, solid foundations.
1:06:17
SPEAKER_1: This time I will build things the right way."
1:06:20
SPEAKER_1: And then the next panel says, "Much later."
1:06:24
And then there's a big mess of buildings that all look completely different
1:06:29
connected by a bunch of weird ladders and staircases.
1:06:32
SPEAKER_1: And he says, "Oh my, I've done it again, haven't I?"
1:06:36
SPEAKER_1: So this relates to at least my experience of trying to make things work.
1:06:41
SPEAKER_1: And then you try to start from scratch or start afresh.
1:06:46
SPEAKER_1: I'll do things differently, is going to be a lot cleaner.
1:06:48
SPEAKER_1: And then six months later it's even worse than it was before.
1:06:53
SPEAKER_0: As engineers, I think we can all relate to it.
1:06:55
SPEAKER_0: If not, you've just been not doing it for long enough.
1:06:59
SPEAKER_1: That's a good point.
1:07:00
Or you're a genius software architect and you must be hired immediately by
1:07:05
all those companies.
1:07:08
SPEAKER_0: Yeah.
1:07:08
So Manu,
1:07:09
thank you for helping go with all these stories behind the comics and just
1:07:14
talking about them.
1:07:15
SPEAKER_0: This was very interesting and pretty rare.
1:07:17
SPEAKER_0: There's not many software cartoonists.
1:07:20
SPEAKER_1: Well, thank you so much.
1:07:21
SPEAKER_1: Thanks for having me.
1:07:22
SPEAKER_1: I hope this was interesting or at least entertaining.
1:07:25
Thank you to Manu for going through some of his favorite comics and the
1:07:28
stories behind them.
1:07:29
SPEAKER_0: To see more comics from Manu head over to his website at ma.nu.
1:07:34
SPEAKER_0: That's a pretty clever name.
1:07:35
For more stories that Manu shared on creating comics,
1:07:38
check out,
1:07:38
Deepdives in the Pragmatic engineer linked below in the show notes.
1:07:42
If you enjoyed this podcast,
1:07:43
please do subscribe on your favorite podcast platform and on YouTube.
1:07:47
SPEAKER_0: Thank you and see you in the next one.