Transcript: Greening Django - Chris Adams

I'm Carlton Gibson, joined as ever by Will Vincent.

Hello, Will.

Hi, Carlton.

And this week we've got Chris Adams with us.

So Chris, I I don't know how quite to introduce you, Chris, but I first saw you at DjangoCon where you gave an absolutely mind-blowing talk about greening Django and environmental impact of tech and all of these things.

So I'm hoping that's what we're going to talk about today.

Super good.

But thank you for coming on.

I guess look look, we let's get into the um climate stuff momentarily because that is just an amazing topic.

But we always ask and we want to know how did you get into Django and what's your backstory?

Give us your sort of your origin story.

So I initially studied things like biology, chemistry, and maths uh with the idea of maybe being a doctor or something.

I totally got distracted by like photography and like computer games and all these other things in high school.

And uh I ended up getting really into media instead as a result, and I didn't realise you could even study that.

So I did that for a couple of years, and uh then I ended up doing uh going to university to do a course called Contemporary Media Practice, whatever that is.

Brilliant, right?

I know, and uh we ended up doing uh I learned all about how to you how to make films with an actual Bolex camera that you wound, like you see people in World War II style movies.

We did um uh like digital video, and uh we also did this thing called digital because this is like uh I think early 2001, 2002.

So playing around with flash and stuff, that was my exposure to like computers and things like that, and uh okay.

That was kind of going to that and I ended up working, I took a very kind of artsy route into this, and uh at the same time I was kind of getting interested in the whole notion of people talking about blogging and the whole kind of whatever that kind of whatever the closest thing that might have been to a scene, I suppose, around this in London back in the early 2000s, and uh I got involved in things like digital civil liberties, and as a result, I ended up a lot of the kind of artistic practice went into that.

So I ended up working on a couple of wacky projects.

So we would do things like make films uh or do photography projects, but the thing that I did for my final year project was basically build what you might recognise as something that looks a bit like a Russian dash can these days, but with an insurance process built into it.

So uh it was it was designed as a kind of like comment about the whole kind of like startup text scene, I suppose, partly because we had a lecturer who was his name was Richard Barbrook, and he taught us all about this.

Yeah, there was a there's a particular course called Media Freedom.

So we talk about who who owned media and things like that, and uh so you learned a little bit about what kind of privacy laws were and what laws were tied to things like say copyright and stuff.

And the thing I realized when you're looking through this was that there was a load of really interesting and I don't know, quite I get like pressing now, I suppose, uh, discourse about like the impact of tech on society and stuff.

And we ended up building like this dash cam with the idea that we would create a kind of like faux mock startup around this with the idea that um what's the best way of describing it?

Um yeah, I I'll maybe I should step back to talk about where I lived at the time because that might help.

So I used to live in a pretty rough part of town, and uh journey, when you were driving, you wouldn't want to like beef at someone or something like that because there was a lot of aggro around there, and uh we I I we end up uh uh at the time I used to live with my mum because I was quite young, and I'd see her get really, really frustrated with this.

So as I so I basically built a tool which was literally like a little Linux box, uh this is for my art project, by the way, that run on the cigarette lighter in the car, was hooked into a kind of cheap webcam and a USB dongle and uh and a big fat guitar pedal that you could you could like hit.

And the idea was you'd be driving around, and when someone cuts you up, instead of like screaming at them and endangering your life, you pull over and just like mash this button, like, oh god, I can't believe I'm so angry.

And then by the time you get home, it'll take that video, upload it into the cloud, and then uh basically send uh uh do some number flip recognition and then expose that as a fee to the person's insurance company so they get priced off the road for driving like a terrible person.

And the idea behind this was that okay, this is an idea around this is the direction we're gonna heading to be heading towards, and uh the we built this and we thought this was actually an awful thing to build because it's everyone needs a break, right?

So this was that was it was start as a kind of like artistic statement.

So we after building this, we figured, well, this idea called things like prior art, right?

So what we did was we uh as because I was an art student, I basically opened the s open open sourced all this, and then we showed it literally in an art gallery, so it's literally prior art to make it really, really hard to patent.

And from there, I kind of got into like I guess like artsy techy stuff from there because it seemed like just a really interesting uh field to be working in, and uh that was kind of what I wanted to work in from with from from that's kind of what got me into tech, I suppose.

Yeah.

Uh partly I mean, like the purpose of the art project was deliberately to start a conversation around this kind of stuff because it felt like oh yeah, there's all these words like surveillance, which is like bottom-up surveillance, but at the same time, there's a whole discourse about okay, who gets to be the people doing this?

Anyway, this is a bit of a divergent, but that's kind of how I got into tech and playing the stuff.

Yeah.

It's exactly what we asked for.

But so how did how did you find Jiang?

Oh, okay, so

So when I was initially working with that, uh, I was uh that was roughly around Rails was starting to become like this cool thing.

And uh in my final years at university, I was like doing we didn't really do much, like the coding we did do was like dreamweaver and action script or stuff.

And uh I kind of bought a bot uh a book like Agile Web Development with Rails, and uh I was like um the a guy who was working with me, he was a bored cogn cog science student, cognitive science student who thought, Chris, the stuff you're doing sounds more interesting than the stuff I'm doing, so I'll help you on your digital project because it sounds like a laugh.

Yeah.

So he did all the kind of crunchy C stuff and like learning with servers, and I tried to do some of the front-end design bits, and uh that was Ruby on Rails, so I learned a bit from there, and then after graduation, I I got on really well with this guy.

His name is uh Fred Fix.

We set up a company together and uh we said we're only gonna work on wacky open source, make the world a better place, techie stuff, and uh he so I defaulted to doing I can like front-end stuff when he did a bit of like Rails, and then the space we're in, there was actually there's another agency who'd been doing a bunch of Django work.

I think it was like 0.96 back then, and uh like you share an office, and I was like, Oh, what do you do?

Oh that sounds cool.

Oh, we do this.

Um, is there any chance you could help us out?

Because I can see you guys can code, and like uh my friend Fred, he could code in Python and I could do some front end.

So I got my first exposure to Django back in 2008, Django 0.96, only working in the front end, uh, whilst I was working with other people who did all the back end, and this was before we had like migration, so people would like screw, like hey, like we're gonna kind of add we're gonna change the databases.

Uh, can you like paste this SQL file to like we had all that horrible stuff?

So that was my way into Python in Django.

So that was like 2008.

So oh my god, that does age me, doesn't it?

Wow, yeah.

Whereas when I was doing when I did the equivalent back then working in Rails, I had to learn all these other bits of it felt very much more kind of opinionated around Ruby.

So I was doing front-end stuff really, and uh from there I did a bit more kind of Python work, and then I think my first proper Python project was I went from there to working in Twisted, which I found extremely difficult because well yeah, there's asynchronous stuff and all this event-based stuff, which was oh god, really hard for a media study student to get their head around, or at least or at least this media study student to get their head around.

Andrew Godwin says it's hard, so I believe him.

So on that front.

I I love that project though, because that's that's so I when people email me about my my books and stuff and the projects they want to build, often it's it's something like that.

It's like, well, I want to have a back end and an API, and I want this and I want this and I want this, and it just sounds like impossibly complicated to me, but I'm thinking about how to actually build it as opposed to like, yeah, you could, you know, like and you just step by step through it.

Um so I yeah, it's I I love that.

That's it's like you're not constrained by like that sounds hard.

It's like, oh, like it's tech, like it c it can be done.

It was in it can be done, right?

You just gotta find the tools and work backwards to to the end goal.

I was in this case, I was extremely lucky to have someone who was both bored and really, really, really good at computers who kind of would help me.

So I came up with like the name and like the general like wacky concept.

In the end we called it grassroots as in like you would grass someone up on route to somewhere like that was kind of the it was like a multi-layered kind of artsy pun stuff.

So yeah that was it it was um you don't know what you don't know so you could have uh jump jump in and when you're like that young you don't need to sleep as much as you might do now when I'm like closer to 40.

So yeah it was uh it it was a learning experience and in some ways fun I suppose

um build building apps build building web web applications in the early 2000s are common common thing to happen but then many many years later not many years later probably 10 years later I saw you well eight years later I saw you giving the talk about environmentalism so what's the transition

end Fred we set up a company did that and then uh we didn't really know how to manage like expectations the commercial side uh all these other things and we're doing all these wacky projects like another one we did was access control for a co-working space so uh with a bunch of open source software written in Python the twisted thing so we built a door a daemon called DoorD which would make the door could you work back and forth right and it would read like RFID cards and the idea being that when you walk in it would check against uh an LDAP directory but also it it could play your th

eme music or could play any kind of mp3 so

but often they have fun with it especially here in Boston with the Red Sox so they'll just play ridiculous things.

So they actually have entrance music every time they come out it's like a 15 second little clip as they you know grab their crotch

So um that was so so answer the question Carlton that was um me doing a bit of Python stuff and I learned basically why it's not a good idea to put a Linux box connected to a building onto the public internet with SSH access right like a there were like I had we further down that we have you guys saying hey why is your door sending me spam right because you had like things like that that we realized well the but anyway that's another story.

So I went to bits of like Python and then after doing that for a while uh I mean Fred we kind of like decided to wind up that work and then I took a job at an agency where I at an agency called Headshift where I was working as an operations person because in all this stuff here I gradually learned uh basically all these difficult painful lessons under on my own money on how not to run servers and then by then I kind of had uh you know enough scar tissue to kind of work in that so I ended up

uh working as a kind of like junior operations analyst uh system administration dog's body person and then in about six months I ended up being moved into I think like consulting basically so I ended up being the person to I don't lead on as a kind of solutions architecty kind of person largely because I could talk to humans and talk about technology at the same time and that was useful

done.

So there's a bunch of like Drupal projects and WordPress projects and then like bits of chef uh DevOpsy stuff and then um I I there there were some projects which I didn't really like working on partly through because of technology choices and I basically look if you put me in another one of these projects I'll quit and they put me on another one of those projects and I quit and uh so I ended up moving to another company what was called Amy that stood for Avoid Mass Extinction Engine which was one of the companies I'd idolized when I was younger and uh they

were basically trying to work out the carbon footprint of everything and I was working as a Rails developer for them.

So my first work as like a Ruby coder and then I worked with them for like a year and a half and I saw what changes when VC companies when when companies raise too much VC and then realized oh god now we're gonna have to monetize or figure out how to make back the returns inside three years or so.

So I did that and then after 18 months I basically decided to go off on my own again as a kind of uh independent consultant and uh at that same time a friend of mine who had been doing a bunch of Django work he's like he was like Chris um this is one project this called Demand Logic which is turning into a proper company now can you help me out because I know that you've gone freelance recently so uh I basically did that I stood up working with him and then we did a kind of transition process and that was me going back into working

with Django again back after 2008 and like it was really nice because I had a background with Railsnap by that point and then I really appreciated lots of the ideas that were also implemented in in Django and there was lots of things that I really liked that I didn't see like the whole kind of notion of forms in Django is really really lovely and that was something that wasn't so well developed in Rails for a long time.

And then I ended up staying with Python because I found the explicitness much easier for me to kind of keep inside my head and then ended up working on that and then ended up I think proposing ideas to conferences because I figured we should probably should be talking about climate given that we're in a climate emergency and like you know you listen to kind of fellow travelers or trusted messengers so this seemed like a way to do that and also I really liked the kind of Django community back from like the P

yCon days back in 2008 so I ended up thinking like well if I'm gonna be if I'm gonna join one community Django seemed like the nicest and a really nice community and uh where did where did we speak?

I can't remember was it in Denmark or was it in Italy?

and it's like you know it's doable for you on the projects that you work on to try and consult this with and I was just like wow that that is that is amazing and then obviously it talked to you afterwards and it's like

to a Django Con or a PyCon or any of these things.

And then I went to Django Europe because it was in Florence and I went I'm going there and that sounds amazing and I was just blown away by you know the the the the talks and the the people and the just everything I've been like why haven't I been coming to this for years it's just absolutely the most foolish thing in the world to miss Django Con.

Anyway that was that was my that was how

so just we can jump hang on I was gonna jump around go ahead finish the the thought

being an abundance of in Rails was actually the tutorials, the really high quality videos and uh education uh educational support material so that you could be a few days ahead of being found out, right?

And the other thing that was actually quite nice was that the I found the Rails guides documentation absolutely stellar.

And uh when I I mean it was probably because I had 10 years difference between these two, yeah, but I found that when I was first coming to a lot of the kind of Django documentation, I wasn't able to make so much sense of it because I wasn't so confident in Python at the time.

Whereas with the Rails stuff I found it a bit easier for me to like copy and paste and kind of get to a sense of confidence, I suppose.

So those are some of the things that I found.

And I guess one other thing that was a really big thing, which I really appreciated in Ruby land was a CLI tool called Pry, which is very which was it made messing around on the command line like you would within using iPython or Rich, it made it really nice and and it gave you a really lovely experience.

And it took a while for me to find some things like that.

But I think those are the main things.

That and the testing, and back in uh I guess like early 2010s, was the confusion around what testing tools to use.

Uh, there wasn't like Rails and RSpec in the I mean Rails had RSpec and uh a few a few very very clear things, whereas in Python I was like doing a bit in Nose, doing a bit of PyTest, doing a bit in something else.

I mean, that feels like it's sped it down a lot more now, but I found myself a bit confused.

Yeah, there were at that stage.

Like the you know, should I be using nose?

What's what's so good about it?

Like, yeah, how is it any different from what you know is just no guidance there?

That's that's certainly the the case.

I mean, in part because I think Django's docs are really good, but in terms of official tutorials, there's Django Girls, poles.

I mean, it's an interesting thing because uh I have my own stuff and I would just as soon take like the very beginner stuff and be like, well, here's three, you know, three here's an official hello world, here's an official crud, you know.

But the docs are I probably like someone should do that, and like I could do that, but it's like it's it's the polls tutorial is sort of this almost like original source, and no one's really written those things.

So I feel like it would I feel like which is to say, I think it would be stronger for Django if there were more than one official tutorial, and probably I'm positioned to make something happen on that front, but there's this like hive mind of Django, I and someone else haven't.

But I think having you know, because right now people go to polls and they sort of get like they probably get confused unless they've done something web before, and then they go to Django Girls, or maybe they find my stuff.

But in terms of Django itself, would probably be stronger to have some yeah, more tutorials as opposed to just docs.

Um yeah, I think once you you're right, once you're in, you that you appreciate the extensiveness of the of the Django docs, but when you're working your way in, it is a little bit more difficult.

And then you think about how much it would cost to hire like an editor, like someone who's actually got specialist skills in this field.

Like, I kind of feel like these are things that it's a case of us deciding to allocate some funds to have people with the protect with the correct expertise to do this stuff and maybe with a different perspective than what we might have because a lot of us are really, really conf really comfortable with them, and we're coming from a different place of people who are just coming to this for the first time.

So, yeah, I think there's something like that.

There's a bit more work to go on there to make it more, but like that was a nice thing.

But I think there's there's room, I think, for you know, a a full-time, not full-time, but like a professional um uh what they call the technical writer, professional technical writer to you know be funded to do you know a much more substantial amount of work on the Django docs.

I think it is the introductory level because you know at the reference level it's perfect, but those those tutorials and how-to levels, they're not quite as as rich for simple reason it's really hard to write those and time consuming.

And yeah,

I mean, on some level it should just like I mean, I already give away the first like few chapters of all my books.

Like this is me being incredibly biased, but like you want something like that, like kind of just copy and paste it over, but I can't I I can't do that because I'm on the board, I'm I'm conflicted.

Someone else should come in and write something like that.

But you know, Django Hello World, Django Crud, maybe Django Auth something.

Um and you know, the issue is I mean, we we've talked about this in the show, is you know, the the Django board is is volunteers and it's small, and like already, just speaking personally, I'm doing way too much right now.

It's unsustainable how much I'm doing for the board.

And I think Anna is probably the president in the same boat.

So there just needs to be a way to do that.

But I think it would help Django and to have beyond just Django girls, yeah, to have those.

We probably could even, I mean, we had Eva from the Python Software Foundation, you know, there are grants available probably to do something like this, but someone needs to, you know, do all that work to make it happen is the thing.

But maybe someone listening can have a suggestion because I think you know, and even for me as an educator, I prefer teaching intermediate stuff than the beginner stuff.

I mean, I th uh I think it would be stronger for the very beginner stuff to be within Django itself as opposed to having to go to a third party like me and the other author, you know, few authors out there.

Because even for us, like someone yesterday was telling me he was like, Yeah, you know, you content creators make all this money.

And I was like, I was like, are you being sarcastic?

Because I can't I can't tell very via email, but it's um I love doing it, but like, you know, I take a discount over just being an entry-level Django developer to do my stuff.

Um, so I think that's part of the part of the issue.

So, anyways, long way of saying, Yeah, we should Django, we should, we should we Django should do that, and there's probably funding to have someone, a technical writer, write up those things to kind of match rails and friendliness for beginners.

Anyways.

Right, so oh, I wanted to ask you, yes, quick, the environment.

Um yeah.

Quick.

Is this just like a doomsday engine in terms of like creating a virtual currency to like use up insane resources and not really do anything?

Like what is what is your take on it?

Because you know how people, you know how have you ever seen that kind of that kind of meme where people are saying it's it's got um a proof on the Simpsons diving uh in in in front of someone with like Elon Musk and valid criticism and going on the internet, right?

You have the same thing with Bitcoin, people get really, really, really, really emotionally invested in it such that if you say anything against it, it's very difficult to have any kind of meaningful, kind of informed discussion around it.

Watch, watch, watch.

All right, so generally, I think that there are much better uses of uh all the compute and the resources and everything like that.

And uh I I I kind of feel like if you really care about the decentralization aspect of it, then you should be able to decouple the notions of decentralization from proof of work because the proof of work part, which basically creates scarcity when there needn't be scarcity, not just in compute, but also now in kind of bandwidth, because there's kind of tokenization of bandwidth and tokenization of storage with SHIA.

Every single fundamental building block we use to create digital services to make the world better, there are tools and there are various schemes which essentially create scarcity when there needn't be.

And that pushes up the costs onto literally everyone else to enrich a small number of people.

I find that problematic, but this is uh I if possible, it'd be really nice not to talk about Bitcoin for the podcast because I think there are other people who do a much better job of talking about it, and uh there's there's you know we could talk about basically.

I think it's really easy for me to get on a on a high horse and start talking about it, and then suddenly the hour's over, and I and I'm gonna and another thing.

Yeah, I feel

Tell us about climate and Django's history.

Like give us your give us your your your Yeah, okay.

And uh I can see uh the way I see it, like if we're kind of if you're over the age of like say 18 and you consider you considered an adult you know we're kind of heading into like we're literally in a climate emergency like we have you know more than half of all the kind of local councils in the UK and lots of places around Europe and in the America saying hey we need to do something you see all these things and I kind of feel that we tend not to talk too much about this as technologists because we don't really have the language for using this and I feel

that it's kind of what should be part of your your part part of what you consider being a professional is to be aware of these changes and avoiding and and take steps to avoid needless harm if you are in position of power when you can make these decisions.

And that's kind of how I framed it for the Django talk specifically because I kind of feel having a sense of control over your own destiny is really important and I and that's why I think lots of people respond nicely to the kind of climate Django talks because of like oh wow there's things I can do I'm no longer powerless and I think this I think there was also something interesting at that conference is that later on someone was saying well I'm just a developer what can I do and I kind of feel well that's like saying someone I'm just someone who gets paid multiples of a median salary in

my place where I'm able to pick and choose my jobs and able to basically dictate the extent of how I work such that things like 20% time are a thing or if I get severance it's six months of severance rather than being fired on the spot in many cases.

And I feel that like yeah if we're professionals then we should be able to I guess rather than just kind of like feel bad about this weird kind of dissonance between what we do professionally and what's happening around us it makes more sense for to align these ways and there are ways to align this stuff and I think that's a more healthful healthy way of thinking about this and also I think if you can frame the stuff that you do in these terms then I guess it makes you like more for healed as a as a technologist but

in many ways the things that you might value as someone who cares about I guess us moving to an advanced and humane future are the same things you'd value as an engineer like you know avoiding needless waste avoiding needless harm.

Yeah all this stuff here right and it often makes you look like a rock star if you're able to I don't know halve the bills of things you're at or something like that.

I mean rockstar in a good way then but you know what I mean.

the early 2000s you meet various people and I think uh I can't remember where I met Simon Simon Willison but I heard about Django from and like we bump into each other I think some of the open rights group style things and uh from that I met people like Tom Dyson and these people who I really looked up to and uh they were very much like I guess they had a halo effect for the framework that made me want to uh go towards that because I thought like the things that were being built like every block and all these things here which were about okay let's talk

about an agenda in terms of okay let's address some of these balances of power these seemed like really cool and important things to do and I feel I I that's kind of where some of this came from and like I think this is the other reason why I I end up speaking that a lot of the Django stuff in addition to me being enjoy like interested in like the framework and everything is that I feel like there's a kind of there's a there's something of a kind of climate tradition with Django.

I know there are people who have worked on things like this.

So the people who I kind of look up to or had looked up to a very long period of time so I look at say like I remind you like Simon Willison Andrew Godwin uh Tom Dyson uh oh Natalie Down I think she's Nat Bat on Twitter these are the people who all worked at say an organization called Torchbox when they worked on a thing called the carbon account which was like carbon uh people trying to work out personal carbon footprint stuff back in 2008 and they were using Django to build with this stuff and uh and now you look at say some of the companies that are doing some really really cool st

uff like uh in the UK in particular there's a company called Octopus Energy uh they're one of the fastest growing energy companies and uh they also have a really really lovely style guide which I can always always reference when I'm working on projects and uh yeah that's um you know there's a climate angle there as well and I kind of feel like even when I was in Germany where I work I worked at a company called the Mondo and like once we've realized that we need to get off fossil fuels for energy one of the next problems is figuring out how to heat space heat things uh or

like in figuring out heating as a problem and like they were working on that which I worked briefly and again that was a big Django thing Django team and I kind of thought that like it makes uh Django is a really really good tool for this but also a lot of the aesthetics or the people I look up to seem to be associated with this.

So I figured yeah that's um thread I should probably run with that because I figure like it would be awesome if Django just became like the obvious choice to build any kind of climate the next gajillion dollar climate tech startup or something like that.

You know and it just feels like because there are you can reach into data science and because it's so ubiquitous it feels like it's a really nice place to have lead reach I suppose and there's a really nice scope for cross pollination as well like there's loads of interesting stuff with like say machine learning and people figured out that yeah machine learning uses huge amounts of electricity for what it's doing.

But there's also responsible scientists and technologists and innovators there.

And there's things like codecarbon.io which is basically a Python context manager that will basically run a process or run something then tell you what the carbon footprint of that would be and it's initially used for like working out say for you to kind of think about how you might change your machine learning run.

But this is exactly the kind of stuff you might use to kind of either put into like CI or anything that can run a process like you could I can I can imagine this being tied to say anywhere that anywhere you kind of run a Django process so you can understand what the environmental impact of that might be so that you can like tune it and do things like that.

complicated imagine how many yeah uh there's a really good post that the Firefox team wrote about how they basically used they applied machine learning to look at the similarities between all these different kinds of similarities in their test matrix their targets right and uh they basically okay I don't I'm not a data scientist or a machine learning specialist they did like some AI magic to work out which ones tend to be like meaningful failures compar

So there's no point to no no they they they were doing some stuff like that basically but they were to able to achieve really drastic reductions and the use of CI there.

And there's all this space for ingenuity and innovation here.

And I kind of feel like these are the things that speak to our values as like technologists and people who want to be building the future and I feel like these are the things it would make sense to make consistent with how we feel as well what what what we want to do in our day job really.

And I feel like if we kind of end up building Star Trek instead of Mad Max along the way, this is also a win right like yeah.

I mean but I I almost wonder if we had a list you know so if if someone wants to tackle this for Django like I almost feel like we should have a list of like open like millennium problems that someone for Summer of Code or else were elsewhere has wanted to to tackle because um yeah maybe that's an interesting idea to I mean certainly the sorry just to say like yeah that just how long it takes productivity hit as well as an environmental hit to take so much time to run tests.

Like you can talk about like efficiency uh all the time.

Uh, but it's also important to think about the other things that you kind of value, I suppose, because we can talk about how we and I this is one of the things that kind of I guess drew me very much to like the Django community was when I went to the conferences, you know, there was a sense of inclusion and sense of camaraderie and community, and I I find that I I I really found that a real plus.

And uh when I I ended up like helping at some of the Django girls things, and I saw how generous people were with their time and expertise, and I thought I definitely really want to be a part of that.

And I again I think that there's a lot of ideas that I think are really I think we don't celebrate so much or really appreciate quite so much that we could celebrate more or really kind of make more of a thing of.

And I kind of I've got it.

I'm just not like just a fanboy of like kind of torchbox and people like this, aren't I?

But I kind of feel like we should be like we celebrate when you celebrate, you know, um companies doing a big sale or like or or raising a bunch of money or VC or set or going to an IPO or something.

Yeah, that's good and everything, but I think there's also things when we see people decide like okay, we're not going to sell on my agency, but I'm gonna sell it back to the employees so they can focus on doing really, really meaningful work that other that might not necessarily be as profitable as other work, but is really vital work.

Like I know that Torchbox did something like this recently where they the founders basically took a step back, and I thought, like, this is a kind of classy move that I wish we would see more of and be celebrating more and be talking about because I feel like these tie into the ideas of community and things that we seem to like really value uh uh when I when I speak to people at conferences, and it feels like these are things are worth being aware of in the same way that when you look at I guess people working at things like 18F in America or GDS, these

feel like things which are also things that we we do celebrate, but I feel that we could celebrate more because they are like vitally important services and things which improve the lives of so many people, so yeah, it's stuff like that.

You know, we we don't call them out.

All the people who help a Django Girls workshop, we don't call them out.

All the people who just are there on the the mailing list or the forum that we don't call out.

And the Django community is so much more than just the repo.

And yeah, I mean to to call out cool things in the ecosystem too.

That's yeah, it's all part of the same you know, we need to recognize that what a great community we have here.

And most people spend their time doing the work, but um it's like who's gonna promote it.

I mean, one of the things I I hope that comes out of this podcast is introducing people who do fill those roles and and calling attention to it, because there isn't an official Django way to say, hey, here's who organizes these conferences, here's who are the people who have done all these things, and this is what they're like.

Uh so that's partly, you know, in a way to take it to your question of, you know, owner occupied, there isn't really like a certainly not a monetization, but there isn't a a way for someone or some organization to to spend the time to promote things because that is also a form of work beyond saying, do come to a Django Con, do get involved.

And once you're in the community, you can experience it.

But you sort of have to take that first step to um to have it.

But I think but certainly certainly everyone on this podcast, you know, is big on the Django community.

So I think it is the case once you take that first step, whatever whatever it is.

In any event, I I did want to call out one thing just in we have links to your talks.

Um for someone who I wasn't actually aware aware, I wasn't thinking about this, but the fact that you have a slide saying the internet is the biggest machine in the world and it m mostly runs on fossil fuels, like in the US, I don't even know how much the internet counts for energy consumption, but I think it's like other percentages out there for like actually how much is the internet versus

So, first of all, the internet uh is roughly you can think of it in terms like the if you read all the academic literature, you'll find figures from between one and three percent of CO2 emissions and one percent and uh five or ten percent of electricity.

Now, it generally speaking, it's I kind of feel that the amount of electricity it uses, I'm not that worried in uh that.

I mean, basically it's it's it's I think it's more useful to think of it in terms of what is it comparable to, right?

Because this helps us figure out like, is this a good use of that?

You know, is it a good deal or is it not a good deal?

So the internet and like technology in general, it's about two percent of global CO2 emissions.

So that's like roughly all of aviation or all of shipping, right?

And uh it's also like the equivalent to like, I mean, if you want to compare it to other things, that's like Canada or Germany, or the okay, the increase in emissions from us switching from mainly cars to mainly SUVs is round about the same kind of ballpark, right?

So you think about okay, if I want to have these things, if all consider all the kind of uh utility I get from this, I would say 2% of emissions uh for all the things that digital does to me is actually a pretty good deal, especially in the middle of a pandemic, right?

When I think about okay, switching from cars to SUVs in that same time period and it being the same kind of footprint, I feel that's slightly less of a good deal, right?

And I kind of feel like this is why you need to think about what you're comparing it to.

And the other thing to bear in mind is that we need to get this to zero, right?

So even if it isn't a huge amount, the scale of the challenge ahead of us is that we need to get to zero emissions by okay, the as soon as possible.

But uh most governments are saying uh the 20 2050 at the absolute latest, and uh in in America it's like half you know, 50% by 2030.

So we not we we it's not a case of it just being level, it's having to get down to something like that.

And uh the good news is that technology, assuming you're not gonna be doing something silly with proof of work-based currencies, right?

You can actually it it does feel like it's the you know, you're moving in the right direction.

Like I'd say tech is actually pretty good for a bunch of this stuff, and if you're gonna talk about you wouldn't necessarily be talking about the environment, the carbon footprint of tech so much as like what who are we choosing to enable and what behaviour are we uh which needs are we choosing to prioritize?

Because these are the decisions that are probably gonna have much farther reaching implications, but at the same time, it's kind of a professional hygiene thing to like not be running your cut your all your infrastructure on fossil fuels if you can help it, because it's needlessly causing harm.

Like it's something in the order of like you know, one in five early deaths comes from burning fossil fuels.

So even if you don't talk about carbon emissions, there's still all these needless deaths which happen just because we've got this un this destructive default that we can and should be fixing.

So but I go to I go to cloudprovider.com um you know forward slash provision and I I I I click my instance.

There's no column for you know um renewable power.

They do some really cool stuff where they'll basically look at the kind of job that you just did and then say, well, this is what the carbon footprint would have been, or these other regions which we can find on our list of cloud providers.

And ThoughtWorks recently did some have released a project called Cloud Carbon Footprint, which basically will plug into your AWS Google Cloud or Microsoft billing tools and will basically tell you this and say, Well, if you are running, you know, this is what the carbon footprint of your job is, of like down to like service level, so that you have these figures out there, but it's generally right now only the huge companies provide these kinds of numbers right now, and it feels like there's a real need for other ones because you know, healthy

ecosystems are diverse ecosystems.

And uh we had the discussion before about okay, well, there's also the idea that these are the cloud providers who are very, very efficient, but they're also the people who are the absolute leaders in accelerating the extraction of fossil fuels which we burned, uh, which have these problems further down.

And right now we have no way of decoupling the cool stuff from the bad stuff, like accelerating fossil fuels or not paying taxes.

And I kind of feel that this is something that we tend not to talk about too much, and I feel like there is something that we can actually address by by basically, I think the phrase I used uh was stactivism rather than activism, right?

Thinking about okay, well, if you're gonna do this, what steps can you take?

And like Amazon have been on the record saying, if you care about this, please use the green regions more, and like they are moving, but they're only moving as fast as they really need to.

And I also feel that like it kind of feels like wait, so suddenly the cost of a migration is on me to move all my stuff because you, a trillion dollar company, aren't in a hurry to make these changes.

And I feel like while we're stuck with no way of moving away from this, I kind of feel it's very difficult to have much leverage.

And I feel like this is a thing about us being more strategically aware as professionals about what we who who we're enabling when we do this kind of stuff, or at least be able to talk about and say, Well, this is a real priority for us.

Please do something to do this.

Because I'm not saying that we should disengage, but even engaging, well, we I don't think there are that many developed people who are even engaging at this point or know how to start these conversations.

Well, I think it's gonna have to, I mean, as you said, it's gonna have to come from the consumers who say, you know, this is important to me, and that will feedback, at least to the American companies, they're not you know, they care about the bottom line.

So if customers care, they'll add it.

But so

this is the thing.

So we often think in terms of this has to come from consumers first, but in many ways, it's much more of a kind of symbiotic thing.

Where in say the let's look at accessibility, for example.

Accessibility is another case where there's lots of people who are who have been previously shut out from services which would improve their lives, and it wasn't necessarily consumers who were able to get this, it was a concerted effort by professionals saying no accessibility is important, but also speaking to the people who procure and say these are the things you need to ask for.

Like having something like the the W, you know, the WCAG, the web content authoring guidelines, these make it easy for people who write checks to say you need to make this accessible, otherwise, we're not going to do this work.

And like this was this was initially pioneered by public sector, so uh, and then this has led down to trickle down to like even Domino's Pizzas having to care about this and private sector companies needing to take this account into account where previously didn't have to have to have to.

And uh, there's really good examples of stuff in public sector where they are working for this, and like I mean, I speak as a European, right?

Where I think 45 billion euros are spent on IT each year, which is a pretty big carrot, right?

But also in the UK, there is a load of really interesting work where the people who spend money there, like in the orders of hundreds, tens of millions, are also saying we spend loads of money and we have our own target that we need to meet.

Uh, we should be speaking.

So it's a case of consumers, but there's also, I think there's a wider that it's it's it's more complicated than that.

And I kind of feel that I mean the really good example actually will be America right now.

So uh America, we've had this massive uh infrastructure announcement, right, where they're gonna spend something like two trillion dollars.

Now, even Google themselves, Google are like one of the leaders in this fit in terms of greening their infrastructure, right?

So they said we want to be 24-7 matching power, so not on an annual basis, they want to have all the time, all their infrastructure running, all the time on renewable power.

And they said, We we're doing this and we are setting a target of 2030 because even us, a trillion-dollar company by ourselves, can't do this.

We need to work with the government, and uh the US government have basically included the same language of 24-7 power into all their buildings from now on, right?

Now, the biggest buyer of renewable electricity of electricity has now said we're gonna do what Google is doing, so that's gonna change the market much more than just consumers in this.

And I kind of feel like as people who build systems and maintain systems, there's all this useful systems thinking we could draw upon and like expand our understanding of this.

And I feel like as professionals, we could totally do this.

It's like it's like so much you could like dive into if you're interested,

What are we gonna do to raise your hand and mention those points is actually working on the front line here?

It's it's making this.

First of all, we've done some work with the Internet Society and we've landed a bunch of funding to basically build an open syllabus, a syllabus aimed at technologists saying, these are the questions to ask, these are how to think about it.

We're gonna be doing it over the next year, yeah, and we'll be running a series of fellowships who we basically hire them to learn in the open with us to build this open syllabus that could be used for pretty much any kind of documentation.

But the thing that I wanted to speak to you folks about is that we're considering doing something similar to have like a kind of green web Django fellow where we would basically pay, we think you're some like between 10 and 15k for someone to basically write the guides and write the information for this.

And uh we can probably find we can put up like I don't know 7k already because some we we we had some funding land.

So if we can find some other people to put in some of the other stuff, then we could absolutely pay for someone to say these are the things to think about as a specialist, as a responsible technologist.

So just like we have things like the defaults from like hangar from I think uh Jacob, what's Jacob's her name?

Do you know uh Jacobin on Twitter?

I forget his uh he's the

So uh the one of the places he worked, there is like this this document called the defaults, which is this is what we use for our stack, right?

Uh all things being equal, right?

Yeah, it would be so nice to have something like like these are the things to think about as a Django developer because there's stuff out there.

There's some work by a guy called Asim Hussein on a thing called he's written something called principles.green, which is basically how to think about architecture as a developer in this field.

And uh he is now the head of Green Cloud Advocacy at Microsoft who are doing some really good work.

There's totally stuff out there, but it'd be really nice to have something specifically aimed at the communities because in the land of WordPress, there's a bunch of real trailblazers there, and I kind of feel like with Django's background and all the people I've mentioned before, there's absolutely space to have like people saying, This is how we do kind of green Django, or the phrase I've used is gold Django, because there was like green, open, lean, and distributed.

That was like the stuff I used in the talk to make it sound cool.

So this was uh based on the idea of poor, because if you uh work with accessibility, there's uh perceivable, operable, uh understandable, or I think it's understandable, I might be wrong, and uh I forgot what's the last one, uh resilient, right?

So you you you have the word poor, it makes it easy to speak to developers.

Uh it's a it's a useful kind of memory a, but it's also a thing that you can say this is what I'm asking for.

Like it would be cool to have.

I want a golden Django website where you can so people who commissioned.

Uh so green inputs, which is basic basically running computers and making computers because computers are sand plus IP, remember, right?

Like sand with lots and lots of energy, and then thinking about it, turn you know, with computers.

So green, open, open, you like open source, open culture, open data, these are really useful.

L lean.

So if you're going to emit carbon, and we kind of have to because everything uses energy, at least make it count and don't waste it so much, at least be able to respond.

And then distributed, because well, once you because you don't want to be reliant on just one uh one provider, and uh there's also once you understand a little bit more about electricity and how like the underlying things that create a carbon footprint as a technologist, you realize that when things are distributed, you can basically scale up demand in where's places where there's abundant an abundance of green power and not reliant on elsewhere.

And uh, so that's it.

Green, open, lean, distributed.

Yeah, and uh this is the uh that's what I spoke about in Green and Django, the talk for Django days, and oh yeah, I should probably talk about it now.

Um the keynote for DjangoCon Europe.

Yeah, so we'll be speaking about this, and if there's interest, I could probably expand upon that a bit more because yeah, there is um that's the thing that we'll be talking about in a bit more detail.

Yes.

I've got to make my I've gotta make my my project able to be.

This is the thing that's really really cool around distributed, right?

So there's a company called Lanceum, and uh what they do is they basically put a load of data with we we used to think of data centers as big box things, a little bit like a kind of big box out-of-town Walmart, right?

It's miles and no one can see it, right?

And it's the energy is really really cheap, and uh they spend loads of money and loads of time trying to get rid of the heat in the building uh to the out to to where it's outside, and that has some unfortunate side effects.

But there's also this is like the heat is a really really useful thing to actually have.

You can do stuff with that.

So, like in uh there's a whole movement to have much much smaller data centers in urban places where the heat that is generated ends up being used to kind of help with say uh greenhouses, for example, or something like that, when you take a systemic view.

But there are also places where because uh we're in a kind of world where renewable energy is increasingly common and you have these kind of natural cycles, you have cases where if you can get the data centers near where, say, there are solar wind farms or I mean say solar farms or wind farms, when there are places where they're unable to sell electricity into the grid where it'd otherwise be wasted, you can make use of that, which means that you can have much, much cheaper compute than what you might otherwise pay.

So Lancy and use this, they take advantage of this, and they end up providing essentially EC style compute for jobs at something like an something like an 80% discount compared to Amazon's own prices because their entire economies are different to the the economics of running a massive big box data center somewhere else.

So there's all these things that we're not really aware of yet that we could take advantage of if we think about the distributed part.

And that's why I think it's kind of cool because yeah, all this work uh by people in ML have basically made it easy to be aware of this stuff.

So whenever there's anything that we could do which isn't super time sensitive, you can shove it to somewhere where it's gonna be way, way, way cheaper and way, way greener, and that'll still be part of your process.

But you need to design it from the very beginning or at least have that.

And these are the kind of principles that I think we should be aware of as technologists because it's just freaking cool if nothing else, right?

Yeah, dude.

And the only reason that it was possible was that the Internet Society Foundation who are funding us, they basically said, we want to do something about the environmental footprint of the internet.

And we said, Well, the problem is people don't quite know what to talk about it in terms of in terms of beyond just efficiency.

So uh they said, Okay, well, we will fund you and pay $200,000 to fund a series of fellowships to build this kind of stuff.

So there is funding out there to do this stuff, and there's absolutely funding in my in my view to kind of pay for like say people to like uh you know pay for content designers or the people like that for the documentation for Django, all this stuff here.

These are things we value, and I feel like that's worth doing.

And I kind of when I look at how how useful the Django framework has been and the community and how how valued the valuable the community has been to us as well, I'm kind of amazed by just like it wouldn't take that much to kind of double the foot uh the the funding going to the Django Software Foundation, right?

And there's so much you could do with that, right?

And if you think about like a day rate of a of a developer, right, like it feels like this isn't that hard to do, really, right?

It's just a case of explicitly choosing what we decide to to fund and support, I suppose.

But it's seems like that's actually not the case.

Um and we actually had Karen Tracy, um, who was on the board and at at Cactus and written a great book, r uh reached out to Carlton and to me to give us the backstory because the current board wasn't around then.

Um so, anyways, I think these things are changing to open things up.

The issue for Django is always a kind of a management thing, right?

We gotta find a working group or someone to to lead these things um to service the grants.

But there are all these grants out there, and if we're able to do that, we could have people, you know, beyond so Carlton does a lot of work with the summer, summer of codes, season of docs.

There's certainly level steps up in terms of money and features and things that could that could be added.

So so we're not limited by that anymore, which is I guess is what I'm saying.

We're confirming that seems to be the case.

Um, but you know, it's always been the case with Django.

Well we could go out and raise more money, perhaps.

We you know, ask companies that are using come on to contribute, but what for?

You know, the fellowship program runs, and then beyond that we didn't have anything to raise the money for.

But if there's a super project to raise the money for, then you know we could all jump on the bandwagon and you know, shout from the things that I'm sure companies will.

I remember there's at least been one th one thing that's been kick started to actually get uh to create some kind of

So I've heard people you talk about things like uh data, you know, there's critical infrastructure which we're aware of, like buildings and bridges and things like that.

But in the UK, people talk about data infrastructure as a way of talking about like addresses or energy data or things like this, and uh in some other places I've heard people talk about uh code as like logical infrastructure, as in the stuff that we rely on.

And I think it's a case of uh that's I think that came from the low-risk project who used this term, which I thought, oh that's kind of nifty.

And I think we need to be able to talk about these things that society is increasingly built upon, and uh, once if we're able to use that language, it becomes possible to fund this kind of stuff, or it makes it much easier to think about this.

It's just that we probably don't have.

I mean, I guess there is it's it's guessing being open to figuring out how to frame it or how to speak the language of people who are prepared to fund this stuff or are looking to fund this stuff because right now they're not quite sure what to do in many cases.

I try and cut out cut be efficient at home and all these kind of things, but I don't really know what to do.

Like people, there are people who are really good specialists at this.

So let's say if there's someone who specializes in grant in grant fundraising, I don't know, like this is something we've been considering at the Greenwood Foundation, speaking to someone because I've tried writing funding applications and oh my god, it's like pulling teeth, right?

Like uh sometimes it works, and I'm like, it worked uh this time, but it's taken so many other times, and it's like, oh my god, it's taken so much time.

It's not you know, like when you're a backend developer and you do that CSS and you're like, this is not the most effective use of my time, right?

I kind of feel like there is going for us to kind of realize there are all these other functions which are absolutely deaf, you know, real deep domains, and being able to tap into people for that would be would make so much sense.

Like we're gonna try this because it it makes sense now that we for the first time we can afford to, but it would be lovely to see this kind of like maturity, I suppose, uh expanding because yeah, I I really, really am kind of quite emotionally invested in the Django community now, and like also technically, because a bunch of the infrastructure now uses Django at the Greenbow Foundation, and I that's another podcast, right?

Uh but I kind of feel like there is you know there there is an urgency around this, and it would be cool to have like you know a goal at something like that.

Anyway, I've realized we've gone a bit over time, so I'm gonna wind up because I I'm a little bit more than a lot of things.

But um if we have, you know, because it's chicken and egg, if we know these are the problems we have on the one hand, and then these are the groups that are looking for ways to give money, um it's just a matter of putting that together.

Um in any case, so we we will have um links to everything on here.

This will come out in June.

You're keynoting DjangoCon Europe, which I think will happen before then.

Um we'll have a link to I think they'll have the videos up, so we'll put we'll put your talk up when this comes out.

By the way, thank you.

It's so you know, that's the privilege of having the podcast as you know, people were like, wow, that person's really interesting.

And then we get to we get to meet virtually.

So thank you for taking the time.

I'm just like, you know, thinking about all these things that we've been talking about.

So thank Chris, thank you for so much for coming on.

Thank you for joining us, everybody, with um chat on Twitter.

DjangoChat.com.

Yeah, yeah, DjangoChat.com, chat jango on Twitter.

Join us next time.

Bye.