Transcript: Kraken - Çağıl Uluşahin Sönmez

Hi. Welcome to another episode of Django Chats, a podcast on the Django web framework. I'm

Carlton Gibson, joined as ever by Will Vincent. Hello, Will.

Hi, Carlton.

Hello, Will. And today we've got with us Chal Ushahin Sonmez, who's lead developer at Kraken

with Octopus and member of the DSF board member, and many more things beside. Welcome to the

show charl how are you i'm good carlton how are you i'm marvelous marvelous excited to talk with

you um we i've just given you a whole load of things that you are but tell us tell for the

audience who are you how how do you how did you find django what's your what's your origin story

we always i'll start with that so let's go there yeah you only start with that question right um

yeah yeah um i'm chill i'm a software developer of uh over 20 years almost now uh uh i'm a

computer science degree uh also uh master's and phd um so nothing um surprising there um

no that is surprising most people are like i self-taught myself doing php and mysql back in

so actual you know i've no just got the phd in computer science that's you know

okay maybe phd and django doesn't go as well as php and django very good very good very good

but yeah that's the thing i didn't know php good enough so i thought maybe i

No, go on, carry on. I've totally know what you're up to.

Wait, let me ask. Let me ask. It's a big question when we just say, like, what's your origin story? Start talking. Let me be more specific. So did you work with computers before you went to university or is that something you discovered there?

I did have a computer since when I'm 10 years old.

And when I was going into uni, my claim was, I do know everything about computers, so I don't need to study.

I didn't even know it was called computer science back then.

But by chance, I got into the computer science department in Big University, and I think that was the best thing that happened to me.

I love programming.

I love Django.

I think it was around third or fourth grade of uni when I first got exposed to Django.

It was the web programming course.

But when I first built my app, I remember it very well.

I was a faculty member and I've been asked to build this web page for students to submit their assignments.

I don't know why because we were using Moodle, a good CRM for, like, everything course management.

I don't, like, why we couldn't use it.

I really don't remember.

And I think, like, 2007, 2008, I worked, again, like, not full-time.

I worked at Marcafoni, built their authentication system.

It was a Django-powered app.

It then turned out a really, really big company, like, I don't know, 20,000 to 30,000 employees.

It was a big, big hit.

It was a Turkish e-commerce company.

So, yeah, since 2008, I would say, I am using Django.

What was it that appealed at the beginning?

What was the bit that you remember building?

You think, oh, yeah, I like this.

Okay.

I mean, coming from a computer science background,

Like, you don't get to build software until, like, very, very, very late in the process.

Like, I was super lucky.

Like, they started teaching us Python on the first day of uni.

Okay.

That's quite progressive back then because back then it was all C and Java.

yeah i mean like uh really want to mention some of those uh people uh who who then i then i work

like that's that's that's the good thing if you if you build a good curriculum if you build a good

software uh then like you you you get the other people like like good people people interested

people like like-minded around you so yeah uh then like then i stayed in faculty as a faculty

member so i was saying um you said talking about just actually building things with

as a with software as opposed to abstract said studying yeah uh so they they taught us python

They taught us computer science using Python, Java, Lisp, Scheme, but like end of third year or something like, what can I build with all that knowledge?

It was almost impossible to build like desktop applications.

And then suddenly I met with Django, like with just like 10, 20, I don't know, maximum 15, 50 lines of code.

Boom, you have a website working with, like, a database connection and everything.

Fascinating.

That's it.

This is my thing.

Okay.

So the reason I wanted to ask, I want to ask, all this time later, do you think Django still got that?

I think Django still got that.

My position in life has changed.

Yeah, I guess you're not building apps in 50 lines with Octopus.

Less interest in prototyping.

I have less time for teaching.

But I think Django still has it.

Maybe not as like this incredible framework that works super easy out of box for even a beginner with no previous web development, web application experience.

But with all the ecosystem, community, knowledge, inspiration, I think it is actually more than what it used to be.

Can I ask, so you said you started when you were about 10 with computers. What were you building at that age, right? Because I have an 11-year-old daughter, and I can't even think of anything she would want to build on a computer now, let alone back then. But there must have been something that, like we often hear with guests that maybe it was building video games, or they had a parent to expose them to it. It must have been something, right, that hooked you in at that age.

Well, sorry, I'm going to disappoint you.

So when I was 10 years old, it was only me who had a desktop around my friends at their home.

I wasn't building any application.

I was just like using it for...

Okay, well, that's normal.

That's normal.

Like for navigating in the web, for chat, for like solitaire.

sure playing games uh what is funny is uh as a teenager i didn't know uh that

using a desktop is different than what programming is about so like back then i was super naive and

like thinking so being able to solve uh problems with windows like uh you probably remember like

sometimes even connecting to internet was a challenge um yeah i remember when dial-up came

that was a huge thing and then the sound and then it was slow installing application like i

so what i taught computers and programming was was limited to using windows in an advanced level

then that and super naively i thought like okay that's it i don't need to know anything more about

uh computers or programming uh that's it well maybe it's maybe it's still the same like uh

maybe there are people who think like programming is about tiktok uh like content or minecraft or

something yeah but there were as well there are courses that you know i remember a few years ago

at local libraries of course it's like computer driving license thing and it was like how to use

um like word and how to use excel and how to use powerpoint and how to use these apps and yeah okay

all that stuff's important but they didn't even touch on any of the other stuff and it's like but

this that stuff's important but surely you could mention you know i don't know one of these

development environments python being the one i would you know if i was running that course i

would throw it in but there are plenty of others actually uh there there's uh something i always

remember that uh no i think it's actually super related to jango so when i first started my

bachelor's degree um chris steffensen our head of engineering at cambridge alumni

uh i think again it was first few weeks he told us like computer science like computers can use you

or you can use computers yeah nice and like sometimes i feel like uh over the years

when i use django i feel like like i am using the framework and some of those other like

i don't want to give a name because i'm not super um attacking any other framework but sometimes i

play with some other uh maybe easy to say some let's say microsoft programming i feel like they

are using like uh i don't know if you if you know what i mean well the dot net framework is very

rigid right it's like you've got to do it this way and it's part of the class-based thing you

but i don't know that's the thing you need you need to decide like if you're like your framework

your mindset will lead you to a way either you will use the tools for yourself or

tools will use you so you need to pick carefully and i think jeng jeng like jango and all the python

ecosystem we are super lucky about that yeah this this one of the things that enriched initially

appealed to me about python was the lack of ceremony about it so i was programming at the

time objective c and there's a lot of boilerplate and you know at the time there was a lot of boiler

plate there's a lot of long method names there's a lot of class construction before you get to write

the code you get want to write and you fire up a python script and it's like i just want to bang

it out and it's like 20 lines and it's like oh yeah and it does exactly what i want and you know

it might be a throwaway script but you'd never get that expressiveness from you know one of these

other environments before we move on i have to ask your research topics for your phd i see that

like deep neural networks and stuff this is back you know a while ago yeah you know what what was

that like right because you were you were a little bit early to that party right i mean what do you

what was what was the appeal what was it like studying that then and then what do you make of

you know the current chat gpt world which is sort of like all that promise now actually being

actualized i don't know like i'm still skeptical if it's been uh or not uh but that's the thing

uh my phd uh uh was between uh 2013 and 2018 uh yeah i did a lot of work about uh deep learning

um deep neural networks uh cnn's in particular uh but i also have uh a slightly still burnout

to be honest like i didn't i didn't finish my phd uh i like it was one of the top toughest

decision i did uh after five years uh into my phd and moving to london i i decided to uh leave it

um so uh but yeah uh super hot topics uh lately and uh as a as a researcher in the area um

yeah the story is a bit different from my perspective uh i think um yeah there there

there there are uh great success and improvements on the field uh but it's not magic like

uh it's not it's not working hundred percent it's it's not even working 80 percent of the time

but that's not how they mark like do the marketing to you no yeah and that's that's actually why i

didn't want to like after my phd i didn't want to work around um data science or data engineering

or what they call it at the moment uh it's mostly about uh using the tools of like uh big companies

so you like um and you need to like uh you need to go with how good their performance is it's

it's not like from my perspective i was building models and um evaluating systems and like uh

as a researcher that's what you need to do you need to put something new on top of what's done

but in the industry it's more like using that output at the moment and i see only little

progress from like little companies other than the big ones on improving on that but with the

data all those big companies have it's actually now easier to produce better results and that's

what we are enjoying at the moment and of course they will keep getting more and more and more and

more data and we'll improve

the

results.

But still, I don't

like...

Let's say I'm a bit skeptical on the

area, being

too much into the topic.

I quite like being

on the web application

development side of things rather

than

AI, let's say.

Maybe super unpopular

opinion tell us about your work because you're working at kraken octopus kraken which is what

like quite seem to employing everybody it seems one of the biggest users of django you know high

high high traffic really stressing it to its limit etc it's quite exciting so tell us about kraken

i think most exciting thing about kraken tech is it is almost uh maybe after instagram i i don't

I don't know, maybe bigger than Instagram now, biggest Django shop.

And yeah, I've been working in Kraken Tag as a lead backend engineer for almost two years now.

And we have approximately 900 engineers.

That seems like quite a lot.

Yeah, under Kraken Tag.

This includes data engineers, back-end, front-end, customer

success engineers, et cetera.

I would say roughly we have 500, 600 Python developers globally.

So that is why it looks like we are hiring everybody.

Yeah, it's bigger because it's not a quite big team, and we have like 10, like we have

engineering, we have tech teams in 10 different countries now, so the headquarters of engineering

team is the UK London

but then

from Tokyo to

Australia

many cities

Germany, Italy, France

Spain

the United States

US

getting bigger and bigger there

and even Dubai now

so yeah

and I don't know I quite like it

I see Kraken Tech

as the future

google in terms of like technology like building technology uh and i quite like

our engineering culture that's uh that's why i joined the team of course every company every

engineering team have like a team stay to grade and and it's improvement on other other points

But, yeah, it's quite a good team I'm working with, like the wider team I'm working with.

We have good conventions, good culture.

I think that's why retention rates are super low and easy to convince my other fellow Django Python developers who are fantastic to join the team.

i think for a big company the trick is retention rate what's the turnover do people stay or do

they move on and if people stay then you know okay that's a good sign yeah i think yeah

like we're doing some things right like we are working super closely with the community

we're using django python which is the best thing about crack and play

can i ask what um are what version of jango are you on or are you on a private fork already

because that usually is what happens uh no uh we are not on uh sorry let me actually check i'm

i'm curious because uh we are uh we are trying to uh upgrade and having issues with uh my pie

which we can talk in detail.

Well, because as powerful as Django is,

when a company reaches internet scale...

4.1?

Yeah, it's pretty good.

That's very impressive for a large company to be...

Latest patch version as well, we know.

So, yeah, we are actually trying to upgrade Django.

You know, currently we have major issues with Django Stops and MyPi.

Right.

There's actually a team of volunteers working towards the upgrade.

So probably there are other small or big issues blocking the upgrade,

but it's going well.

I think we're down to 1,000 issues, 10,000, 5,000 issues.

but let me remind you it's a it's a it's a big monolith it's a it's a big um uh repo uh

so yeah no but those are cascade as well you'll fix one thing and all of a sudden

a thousand of those will clear up no no to start with we had thousands

yeah so yeah maybe that team can throw a good talk on how they upgraded from django

four point something to django five point something in later in the year okay so do you

can you talk to the typing question so i mean a lot of the the issues are around django stubs and

typing compatibilities you update or or you're not sure yeah so that's not something i'm working

day to day uh but uh for example like i remember uh from a chat from last week's uh one of the

issues was uh django uh object manager uh uh so apparently you can't you can uh

Jungle Start isn't willing to type objects of Django models,

object model of Django models,

because with some not super recent feature,

you can actually change the objects class of a model.

For example, this one took us weeks back.

a forward um upgrade because uh i don't remember exactly but you need to then type it with a super

unfavorable way uh or like there were other there were two other alternatives um so coding practices

best practices library restrictions django features uh actually they reach out to me and ask if django

can change

like the default

assignment of reassignment

of the objects.

Yeah, you're on the board. You can do it, right?

I am

on the board of directors.

I'm not on the steering council.

Yeah, yeah.

I don't think

I mean, that was a super valid

question.

The correct place to

talk about this is the forum.

It's currently forum

used over the uh male groups so i did the retire action i think like that's half of the uh half of

my jobs in the board like uh connecting uh people with the correct resources or correct ways and

like making sure like we have a a sustainable way of doing this as like not everybody has access to

a board member sure but as well i think that there's there's like with you with you there

there is a um you know not just you there are members of the contributor community who are

working at kraken or in contact there it'd be massively interesting i think to get

octopus kraken's experience about typing in django and what you know what we can learn what are the

difficulties and is there a way forward i think there's you know there's some there's kind of

like a chicken and egg problem at the moment it's like people know that it's nice but how do we

begin and well one way we can begin is by looking at a super big installation and saying well you

know these are the actual problems we face can we address some of those um i think uh yeah i agree

and uh i really wish we can like work a little bit closer with the python and django community

like Microsoft does.

I mean, I know what you were referring to

wasn't that some kind of different feedback rule,

but I think like we can also feed in our experience

directly into the like Django code base

with proposals, improvements, ideas.

And even engineering time to make those happen

because there's lots of ideas

and little bandwidth to get them done.

So, yeah, that's what I really like to see happen.

But as every mid-to-big-size company, there's always another agenda item.

But I think I'm getting into other people's minds on that as well.

I think, yeah, soon we might have some steps towards that.

Also, like, it's not just us, which is something.

Okay, you can cut this part out later.

Oh, no.

We can decide.

It's not only about Kraken Tech or people in the team.

It's also about the Django community.

The ways of working is different.

So we always promote contributions into the Django,

but we don't realize it's not that easy to, like, yes,

we are all using Django and Python,

but contributing to django it's it's not only time or willingness it's also like ways of

workings are different there are uh hidden conventions in the django community and it's not

super easy for uh people who doesn't have any contribution background into that community to

Like, come up with super advanced feature requests.

Yeah, no, totally.

So, like, that's something that I've been thinking lately about.

I made a talk last year in Europython and PyCon Spain about coding conventions.

I think it was the first day of EuroPython.

I was in a workshop about PyTest,

which is my favorite test set.

And I was listening to workshops,

and I just realized we need conventions

for open-source communities, too.

I don't know if you're familiar with our coding conventions.

It follows Google's Go coding conventions, basically just some recipes around best practices,

the ones that we get to do wrong often.

that like it's it's a good reminder uh and we use it very much during our day-to-day

practice as part of mostly part of peer reviews well we really need a convention

for for example like for django for even like for django project.com like even for small projects

like django project.com like what's the convention here like how do we contribute or like what's the

preferred way of a specific thing that way i uh as a first timer and again first timer doesn't

necessarily mean i'm a junior or mid-level dev i can be like a lead dev and still be a first timer

like what how should i uh contribute to be welcome because also like over the last few years

in most in like jungle con sprints or like other sprints or some other events like i have

pushed people towards contribution like more senior people and like sometimes they received

feedback super surprising feedback and then like uh i've started to yeah

like motivating people to contribute contribute is step one then do we also need to like manage

the communication or like relationship after the how shall we do it well what's the sustainable way

um yeah no i mean i remember um we were in copenhagen at the sprints and it was a bit like

well you know how do we i was a newish fellow only a year in then and i spent lots of time

going around to individual people trying to help people get going but i was doing the same thing

time and time again and i was by the end of the day i was exhausted i could hardly speak but yet

there were still people there you included were like well how do i get you know how do i you know

join in a project here how do we get because it wasn't it wasn't organized off the back of that

i hadn't you know i did a couple of workshops which were getting contributed starting workshops

at the sprints and they went well and my sort of hope is to you know that that can be something

that other people will pick up the baton on but we don't have a good on-ramp still i think i mean

the django nauts project the django nauts space project is an amazing thing is to try and bring

people on board and help them over these these hurdles but they're real hurdles right it's not

oh yeah i'm just going to contribute to django it's it's not as simple as that yeah but i i'm

super optimistic like uh uh being in a team this size and uh following a super um uh unconventional

uh ways of engineering like uh we are super flat yeah we don't have managers uh like we don't have

product owners

we trust

we put a lot of trust

to developers

and like

but still we are super organized

like there are teams, there are conventions

still we have clients

etc and looking at

the open source community

and seeing

how it works

it's a

miracle like

uh they're not paid actually they're not part of a team or like official teams uh

they only use their like personal time and still produce such quality

software so i'm super optimistic that we can fix things forward um

but yeah going going forward in my software development journey now like

different topics is are becoming my interest and like different i'm looking into improving

the ways of working of community in different ways okay good well that's a nice segue there

because you've been massively active

in the Django and Python communities for years now.

So, can you tell us how you got involved in the community side?

That's thanks to Django Girls, I think.

If we go really back in time,

I think I've been taught this community feeling

And by, again, the team in my uni, I was happily part of some free software days for years, which was a big event back then in Istanbul.

And I met people like I'm Murdoch, Richard Stallman, and like literally help, sorry, what's the word?

Chaperone.

Help chaperone.

No, but that's the thing.

I was a second or third grade computer science department.

department was running this uh local uh free software days with like like a jungle con europe

uh but on free software and we were having like fsf europe uh um presence or like i'm murdoch uh

i even met that like that like debbie of debbie uh i'm murdoch is the uh author of uh devian

I'm a long time Linux person since I switched to OS X a few years ago.

I think I've been told about the value of community and role models back then, but really

it was Django Girls.

my PhD back then and some other women in the department reached out to me to organize the

first ever Django Girls Istanbul and over the course of five years we organized and

helped organize over 10 workshops in Turkey.

We built a small community of our own, of organizers and mentors, not only women, like a mixed, a good diverse, like not so small group.

Then we revived Pai Istanbul, which was a meetup group that was deserted, I don't know how many years ago, started weekly meetups.

Like, I went to Paris-Dambourg meetups for six months every week with my other six Django Girls women fellows.

And, like, I think things like this has big impact.

We talk a lot about diversity, a lot about feeling included.

It doesn't always happen organically.

That's also how I became a volunteer to become an organizer in London Jungle Meetup.

I spent like five years with the community in Istanbul before moving to London, then in London.

of course the first thing i did was to like come to london jungle meetup

and i was a bit disappointed because it was a big one like we don't we don't have that many people

every month it was a big one or like 120 or 100 people and i was the only other woman in the room

and the feeling wasn't great so i'm like okay i need to do something to change this and like

over the course of the years i learned if you want change you you become the change

and uh yeah really i think like that six women changed the scene and like their commitment

to the meetups changed the scene uh long before they changed the scene by organizing the uh

Jungle Girls workshops

but also like

it works

I try the same thing

also my other co-organizers in London Jungle

meetup was super

allies of diversity

and like

representation in the scene

so we work

together towards making the

meetup more diverse

and last

meetup I think they were like

14 15 women out of 80 attendees and i think that's like that's that's good

uh one out of 100 to 15 out of 80 in you know that's certainly that's certainly the right

direction but this is what i want to ask you i think you know we do well in the python community

we do well in the jungle community compared to you know how it used to be and in other other

areas of tech where it's not so um open but where are we still failing what more can we do like

what's the like where do you think do you know what if we did this it would help if we did that

it would help because we want to keep promoting keep working right uh yeah we need to keep working

we need to keep educating ourselves uh we need we need like uh good role models not necessarily

a woman, but a life.

Yeah.

Karsten, you're

one of the

few people in the

community that made it super

welcome for me, both

into the community and both into the

contribution side of

Django, but

I can almost hear

you were doing

great, but I think

you also see

that it wasn't super sustainable.

So we need like hundreds of Carlsons around, not just like one, two, three, four.

We need to we need to work more towards it.

We need to keep it always in the in the agenda, always like in the top five priority, both on the technical board and on the board of directors.

Like, that's the only way.

We need to make sure we have women all over the processes.

Like, for example, now we are looking for a new fellow.

And now I'm a year, over a year into being a director on the board.

It's time for me to, like, dig deeper into our, for example, interview process.

Like, for example, who writes the ad for the new fellow?

Is there enough diverse proofreaders?

Who does the selection?

Are there enough diverse, like, people in that part of the process and the last part of the process?

And I think having a diverse group of us help us come up with those questions.

Not all of us brings the same question into the table, right?

All of us will bring different questions and more diverse we are, more different questions,

more different perspectives hopefully towards improving and hopefully towards on the to the

constructive side of the things but yeah diversity is the key to diversity you know i i think that's

just absolutely brilliant you you said i you know i was welcoming and whatnot i spent

the whole time when I was a fellow trying to bang the drum of getting more contributors from

different areas of the community because you know you go and stand at a DjangoCon you stand on stage

and you look at the the audience it's quite a diverse audience but then the contributor base

it's still the same white English-speaking men and that's not it's not to criticize anybody but

it's largely that way and I always felt though that I was stood there going come on you know

come and help but i'm a white english-speaking man and it's like i can't stand there and

improve things i can say you're welcome i can put out a welcome mat but it doesn't change anything

because i'm not the right person i disagree we need a life i like my my role models there are

great woman but also like great male uh allies without them like we can we can we can we can

make this work i don't know maybe we need to build a like bigger feedback group i don't know if you

know of your impact uh uh all the efforts you put in actually had a lot of impact in the community

And, for example, we changed the, I don't know if you have seen, the DSF membership guidance has changed.

Now, community contributions are stated as valuable as code contributions.

And I think that's a big milestone.

We need to start giving those people some credit because they're building the future of the community.

They're building the next generation of contributors.

It's not any less than code contribution.

And I tell this from an engineering management perspective.

I'm a hands-on tech lead.

So, like, my 30% is hands-on, 40% is people management,

and other 30% is just project management, networking.

Working in a big company.

And one has less impact than the other.

I don't want to quit hands-on development

because that has a big impact.

If you don't have contributors, Django will not improve.

But also, if you don't have people running the community, I also don't see how we can improve further.

Right, but Django is not the code base, right?

It really isn't.

Yeah, that's one part of it.

It's a good framework.

It's great for building your web apps.

But there are other options out there.

So why do we keep going with Django?

because there's more to it than just the web i i think there's a just just to build on what you

were saying i think there's also a i mean carlton has given talks on this the sustainability piece

of being involved in the community so you can come in and and be on the board and do a lot of code

but the it shouldn't be implied like it shouldn't be expected that you do that indefinitely because

people do burn out so one thing i've been happy to see and i guess carlton and i are both living is

having been very actively involved and then stepping away a little bit, but still being

able to contribute and help so that there's a, people can see the ramp and the process where

you can still be part of the community, but you don't always have to be driving the change because

it's unrealistic for someone to, to do that all the time, right? So the things you're doing right

now, I mean, I hope you keep doing these for years and years, but maybe you just want to

attend Django London, you don't have to be the organizer for forever, because you won't be able

to maintain that. Right. So I think it's also important to for people to see that there's an

option other than just stepping 100% away, and not going to conferences and not contributing,

which is what I have seen with some previous generation people. Instead, there's a way to

be involved, but less involved. If that makes sense.

Oh, yeah, it makes sense.

And also, those are things that I keep thinking a lot lately.

But it's natural.

It's natural, right?

That's life, right?

You're not the will.

You used to be like 10 years ago.

I'm not the child.

I used to be 10 years ago.

It was a challenge to build, for example, the community in Istanbul, the community in London.

I put a lot of effort into that.

But, like, of course, it's not the same as it was six years ago for me.

I've reached out of ideas.

I think community had enough leadership from me.

At some point, it will be a good time for me to step down

and for someone else with more enthusiasm and more energy to come in

and share their wisdom, ideas, energy, motivation, time.

But how do we keep the loop?

How do we motivate people?

Now maybe it's time at least for me to think more about those things.

But also, we see comebacks, like Jacob came back into the board after years, and he's

working with a great enthusiasm.

That's also inspiring, because it's not always the newcomers sometimes ask.

I don't know, I think I can call myself an old-timer now.

we also need to see see see the future like uh uh what is it for me in in 10 years time

yeah yeah no entirely you saying that makes me think too of just with meetups in i i live in

this in boston and there was a there was a very active django boston meetup before covid and then

covid happened an organizer left and it didn't happen for three or four years and now it's

coming back again and there's some lead organizers who I've been helping, but it's, yeah, it's

interesting to be a long timer and you get these, take these pauses, but then like right now I'm

very active in trying to help. Like I just gave a talk helping find speakers because there has

been that period of rejuvenation or of rest on, you know, for me. And yeah, it's not, you can do

that, right? That should be the cycle, right? You're very active and out there, and then you

step away a little bit. So I'm now living that, right? Where for a while I thought, okay, I need

to take a break from the local conferences, but now I'm like, no, I really want to do it and I

have the energy. So it's good to not just say these things, but live these things. So I feel

like I'm living that a little bit, but it's a lot of work, right? We've only had one meetup

and we're going to have more and more and more, but even in Boston, which is a big programming

town, it just takes, I don't think I can go weekly for six months, but it takes a lot of effort

to get it off the ground. But then you find these people who say, I feel like every meetup I meet

someone who's been using Django for a long time at a company I didn't even know of that's quite

large and they say oh i didn't even know about the community and yet they're using django every day

so it helps you know helps bring things together yeah uh hearing about the boston

london uh london london boston django meetup is that the name yeah uh yeah yeah django boston

boston django meetup yeah uh one yeah one thing happened recently that i've been pushing forward

since a year now is now we have

a list of local Django

groups meetups

under djangoproject.com

which has an interesting

story. Do you want to hear about it?

Yes, please.

So a non-Django meetup, I think

it was actually not last year, but

previous year.

Not every, but last

few years we were running Hectoberfest.

So instead of a meetup, we

do a

hackathon.

And that edition, John Gold came in and said,

okay, I want to do a non-coding contribution.

What shall I do?

And literally, I wasn't prepared.

I was like, can you make a list of all the Django meetups globally

so we can put it on the website?

He prepared the list.

in the Django Con Sprints,

Dave Turner,

actually again from Kraken Tech,

worked on the page

and the backend required

to display it

on the Django project website.

And I think recently,

Sarah filled in the event

and now we have a good list

which we need to improve

and keep maintaining.

This stuff like local Django

um communities meetups uh which i think is uh one of the next steps that the sf should be focusing

on uh like how we can improve how we can help and contribute to local communities one good thing is

telling people they exist because it's to get the word out it's quite difficult so there's a thing

on the jenga website that's kind of handy well one of the i don't know what how the time is spent on

the board currently, but when I first joined, there was quite a lot of time was

spent on funding requests for events.

And that's one thing that the board can do.

But also, as you've said, facilitating and telling people about existing

things is another thing that the board and the community, you know,

people are doing the work but if nobody sees it then it's not fulfilling its uh full potential

yeah that's but well that's the that's the other thing i'm sitting in the board i'm running the

london jungle meetup a good comment is like going to jungle con europe europe item uh that's one of

the things i'm living in uh london like yeah i don't know uh how many people you know in the

community based in turkey based in sudan based in china korea thailand uh there are great communities

they're great people uh it's important that uh we build a better connection yeah

i can't keep talking about this no but you're hitting one of my old bugbears as well is that

we've got we still have this kind of english language silo it's not 100 but it's kind of

still english dominated us dominated australia dominant like just because it's that's the old

base and there are so many users around the world and what can we do to a facilitate them and be

connect with them it's ah yeah well you know so so you're you're uh not a native english speaker

like what what can we the native english speakers do or are there some are there any obvious things

that you and the board are working on to to facilitate this obviously having having showing

showcasing the meetups is is one but are there like what are we what are carlton and i not seeing

as, you know, stereotypical Django contributors?

We need to work on that more.

Unfortunately, there is not a structured work happening

in the agenda of the board at the moment.

There is a slight move on DSF members group.

We are going to open a few non-English speaking channels.

I think we are going to start with Turkish and French.

I'm not 100% sure about the second language.

I know about Turkish because I volunteered for it.

yeah i i being on the board as a turkish woman is my first step i don't know we need to like

conferences uh um has actually a big impact on the community um i saw in your notes about

my talk in jungle com yes uh that's actually a good example so that talk uh um was on a topic

that i wasn't super comfortable with we uh i led the team building a mobile backend with django for

the first time and i was super aware like um if you're doing um a new thing for the first time

uh your lesson learns are super fresh i wanted to give the talk even though like i wasn't an expert

on the ARIA. Also, like, put a lot of extra information in mobile development with Django

that I learned over the course of a year. I was a mom of a newborn, having super, super,

super difficult time and i don't i i think that's the worst talk ever

uh i remember recording in a super difficult condition like uh

i don't know robin was in the other room crying i was having a hard time coming from a maternity

leave uh but really like that talk should be there because like life is not always

flawless

and not all the people

are great speakers

or like native English speakers

and that's

one thing I bring into the

speakers committees of the conferences

I volunteer for if I do

that task

I try to

motivate

the group into not picking

always the best of the best

of the talks and trying to see

like what other like maybe the speaker is not the uh uh expert of the domain but like what

else they can bring um and sometimes you should just have like not great speakers that's that's

life that's the reality that's the person uh or that's the person like sometimes working next to

you uh well i was just thinking that i wish i wish it was less of a barrier entry to giving

a talk. And I don't know if that means that, you know, if that we like having a channel where

people who don't get accepted for a talk, but still have one they want to give can be on an

official Django YouTube account. Or maybe there's a way that meetups can showcase the video. And

the expectation is not that it has to be in English. So if the Django, you know, Istanbul

meetup happens, we should show that these things are happening and you're not forcing someone to

speak a second or third language for it because it's all happening, but maybe there's a way to

find it. I mean, even just the talk I just gave, we didn't record it, but I'd like to see the Django

London meetup talks. And again, maybe just the fact that these talks are happening again in

other countries where you're not forcing people to speak English would be helpful just to have it

somewhere or just say if you are running a meetup and you want a talk to be put on a channel

like here's a way to do it because the bar is the bar is is too high for the django con events i

think there's way more people who want to talk than there are spaces for them and the communities

yeah bigger than that yeah i mean europe python is doing a great job uh about it but like compared

to DjangoCon they are super privileged

they have like I don't know 500

speaking slots or something like that

it's like six

six rooms going for

four days

it's a big urban

but we like we still

need to hear from

those

speakers too

yeah having it live helps I think

right because I don't know like your talk

you gave virtual

it's a lot of work to give a virtual talk and not get any feedback right we like you need that needs

to be part of it especially in real time or maybe just having a clear way for people to give feedback

because otherwise it feels like you're just giving and not receiving in the moment i mean what i'm

trying to uh uh um support uh i'm trying to mitigate in london general meetup is you want

give a conference talk come and practice your talk in the meetup like uh less people more

comfortable uh surrounding you can receive feedback and you can improve your talk uh but

yeah we should we should we should have uh more of those but um yeah i think uh we also need to

improve uh not improve but like we are great as a community but need to need to still still work on

it like uh that's that's that's the uh jungle chat right like i i like to i like to hear more

diverse speakers i like to hear better uh keynote speakers when i say better i mean uh more diverse

Yeah. Well, I think it's just about the board or someone having the infrastructure so that it's easier for people, right? Easier to contribute to Django. Also easier to give a talk. Easier to do a podcast, right? I mean, it's not that Carlton and I are the only two Django people who should have a podcast, but it's a lot of work to do a podcast.

It's Django Bell's podcast.

Sorry. Yep. Thank you. Yeah, that's true. Yep. Yeah. Stepping up a level and thinking about, you know, from an organization, how do you make it easier to do these things is important. And again, especially when it's volunteers helping volunteers volunteer. Right.

Yeah, I mean, the one way that I know is please, please, please do support and motivate those speakers around you who's not as comfortable with the language or with their competency.

because i think that's that's the only way forward uh telling a person you can do it

is the yeah yeah it's the best way that so far i know

to support them yeah brilliant yes you can okay i think we're coming up slightly on time so i just

wanted to pique your mind um on the technical more technical side perhaps we can for a few

seconds you mentioned pytest which is your your favorite framework there is that's your go-to

right yeah that's my go-to uh i i think i've been using it since i don't know eight years

ten years i don't remember and it's something i use day to day i think i know something about it

i never did but yeah uh i think it's one of those software that works out of my box seamlessly

supports us to the end and when you ask the question like to be honest i don't know anything

super super super technical about it other than all those features that i use without even like

noticing so can you can you spell parameterize without looking it up

i can't say it doesn't mean i can spell it but every time i write that i have to look up

i used to look up yeah i used to look up um pi test is great but uh it's also it's also the like

uh like the engineering culture you're in uh for example um i'm like i'm writing more tests

uh since last six seven years um and uh i still i think there are a lot of things that i don't know

uh yeah sure sure but that's always true though uh yeah uh one thing for sure that's also something

that i'm looking into uh uh interviews tech test as as well like the more more more senior to canada

the more senior the developer including myself they they write rely more on tests and write

better tests and uh i i think i also quite like uh um the django uh uh package uh uh

PyTest Django, Django PyTest?

Yeah, yeah, yeah, yeah, because it's got the helpers

and whatnot.

Yeah, it's just

those are things like Django, it works

out of box and

it's super seamless.

Were you expecting

a different answer, Carlton?

No, no, no, I just...

I don't know if I can talk more technically about this.

No, no, no, I just

wanted to ask, you mentioned it,

you said it was your go-to, I just wanted to

sort of get you to paint

your picture you like you like it give you a more technical then or different one octopus you're

using drf for um is that the case if you're going to go into django ninja stuff i i don't know

anything about because i was just going to ask we do more graph call at the moment we

this is off the record uh to join the team uh but like yeah we do have drf but we don't build

new endpoints using DRF.

Okay.

Well, next time we have you on, we'll just talk about

building GraphQL

stuff. Yeah.

Okay, go on. Go with the magic wand.

Wrap this up. Okay, so final

question is always, you have a magic

wand. What do you fix about Django?

It could be more

than one thing, but what

would you want to fix?

Oh,

that's a tough question. I wasn't prepared

for this.

one thing i will ah i know

good i would fix uh class-based views okay how would you fix them what specifically not not that

they need any fixing uh but uh currently we are having a lot of uh issues uh people just like

picking any class basically if they think appropriate and using it but on a scale

uh that we have actually it's not uh uh feasible to push django objects to templates i know for

most of projects that's a super great feature of django but for us that means extra hidden

database queries

and

has been a few

occurrences that took

a few clients down

so we have super strict

conventions towards not

pushing Django

objects to templates but

like half

of class-based views

seamlessly contributes towards that

which we don't want so sometimes

you don't need things to happen seamlessly

in our scale.

Okay, so that's a fascinating question then.

So just quickly, how do you enforce serialization

prior to the view?

Like in the view or prior to rendering the templates?

You must be fetching all the related properties

and serializing to something.

Yes, we do use, we used to use,

instead of Django objects,

We used to do data class classes.

Now we are using utters to serialize all the concepts.

And also domain-driven design is big in Teams and Kraken.

So it's a team-based option,

but there are a lot of components in the monolith

that actually doesn't persist

the way we write data to database

into domain and interface layers.

That's fascinating.

Cool.

We could have another podcast just on that.

Actually, yeah.

I wanted to give a talk

in layered architecture with Django

in last year's DjangoCon

and you didn't get into the list.

Well, you do know this, as we record this,

there's two days left for the DjangoCon Europe CFP, yeah?

No, but yeah, we can talk about more technical Django stuff

in another episode.

The community stuff is equally important,

so I'm glad that we focused on that,

especially given you're on the board

and all the work that the board is doing,

which also can be hidden, just like code contributors

code contributions can be hidden

a lot of community work also

isn't seen but it should be

celebrated because it's important

so

I think we are out of time

we have links to everything

in the show notes

and thank you so much for

coming on and taking the time to

chat with us

thank you also I'd like to celebrate

you for all your contributions to the

community by not

only the podcast but all all years of work thank you thanks for coming on i'm really inspired

um django chat.com and we'll see everyone next time bye bye bye