Latest / Coffee with Developers / WeAreDevelopers LIVE - Open Source Agent Toolkits - Eddie Jaoude

Transcript

WeAreDevelopers LIVE - Open Source Agent Toolkits - Eddie Jaoude · Coffee with Developers

• 0:05
  SPEAKER_00: Hello and welcome to another Wear Developers Live.
• 0:07
  SPEAKER_00: How are you doing, guys?
• 0:08
  SPEAKER_02: Yes, I'm doing well, apart from managing lots and lots of uh assets, bits and bobs, uh things that are happening ahead of the uh kind of our first North American event.
• 0:16
  SPEAKER_02: So yeah, lots happening.
• 0:18
  SPEAKER_02: Busy, busy.
• 0:19
  SPEAKER_00: End of the month.
• 0:20
  SPEAKER_00: I'm alright, end of the month, I'm already putting my suitcase together as well.
• 0:24
  SPEAKER_00: So I have to check if the hotel is a pool or not.
• 0:26
  SPEAKER_00: So it's the thing, you carry so many things.
• 0:29
  SPEAKER_00: I I I love packing for for vacations and business things because you pack yourself as if you're planning to soil yourself three times a day and like you have like uh double backups for everything, and then you come back and it's all messy.
• 0:42
  SPEAKER_02: Completely.
• 0:43
  SPEAKER_02: I mean, I will say, and I won't jump the gun and kind of introduce our guests, but I will say they've probably got a few tips when it comes to kind of packing and living out of a suitcase because they've uh definitely been a bit of a digital nomad.
• 0:52
  SPEAKER_02: I'm probably giving some clues here, but yeah, we could definitely do with some of those tips, especially on how to pack clothes um and just kind of fold them properly so they don't get creased when you arrive at the room.
• 1:00
  SPEAKER_00: Oh I'm totally on that.
• 1:02
  SPEAKER_00: I mean, I'm so good at that, and uh yeah, my wife is like a raccoon on crack doing the same thing.
• 1:06
  SPEAKER_00: So when whenever we come with like I got she's got two suitcases and I got this one business suitcase completely packed, and it's like, yeah, it really works.
• 1:14
  SPEAKER_00: And then I mean also she as soon as we get to your hotel room, she actually unpacks it and puts it all into cupboards and stuff.
• 1:21
  SPEAKER_00: I never bother, I just live out of the suitcase, so it feels like cementing.
• 1:26
  SPEAKER_02: Yeah, I think I I would say it probably is in general.
• 1:28
  SPEAKER_02: I used to do that, I used to just live out of the suitcase.
• 1:31
  SPEAKER_02: Um, and then I find that actually it's like I don't you end up treating the hotel room or something that's just like a like a place you're there, you know, you don't really make it nice for yourself.
• 1:38
  SPEAKER_02: Um so I think that I'm trying to sort of actually do that.
• 1:40
  SPEAKER_02: But then if you're only there for two days, there's no point unpacking, right?
• 1:45
  SPEAKER_00: Yeah, but then uh the other thing is you never used to use everything up because you forget the things on the bottom, you just take the first thing out.
• 1:51
  SPEAKER_00: So if you really want to take your uh uh take your different outfits for this for the week, then you basically should be uh should be uh unpacking them out of the suitcase.
• 1:59
  SPEAKER_00: So that's a proper proper thing to do that.
• 2:01
  SPEAKER_00: But yeah, it's uh uh it's a messy, uh it's a messy idea all in all, how to do these things.
• 2:07
  SPEAKER_00: But yeah, uh digital nomady things.
• 2:09
  SPEAKER_00: I mean, it's uh it was an interesting concept, but let's see if it's still one.
• 2:14
  SPEAKER_00: So without further ado, we got Eddie Ejawood today here uh from PayPal.
• 2:18
  SPEAKER_00: Uh just freshly at PayPal, right?
• 2:21
  SPEAKER_00: You've been freelancing and doing your own stuff for a while, and now you're going back into the the bosom of a large corporation.
• 2:28
  SPEAKER_01: Wow, uh thank you for the intro.
• 2:30
  SPEAKER_01: I love that conversation, by the way.
• 2:31
  SPEAKER_01: I've got to share some thoughts, and and just so much you said resonated, which I thought was um it was brilliant.
• 2:36
  SPEAKER_01: Yeah, so PayPal.
• 2:37
  SPEAKER_01: I've actually been there about five months now.
• 2:39
  SPEAKER_01: Um and so time has really flown by, and I've been a freelancer and contractor for uh 12 years before that.
• 2:46
  SPEAKER_01: So yeah, it's uh it's interesting.
• 2:48
  SPEAKER_01: It's great to be part of a team and to continually build on something.
• 2:51
  SPEAKER_01: I I I I love contracting and freelancing, but when you when you kind of build relationships with um the people in the company, your team and the community, and then after three or six months or even 12 months, you just kind of drop it.
• 3:03
  SPEAKER_01: It just it just felt really hard.
• 3:04
  SPEAKER_01: In the beginning, it was exciting, it was always something new and shiny.
• 3:07
  SPEAKER_01: But then after doing that for over a decade, I was like, you know what?
• 3:10
  SPEAKER_01: I actually want to want to really invest time in something and build on those relationships and build further.
• 3:15
  SPEAKER_01: So yes, um, so I was super excited to be part of PayPal and part of their uh uh DevRel team and uh yeah, really enjoying.
• 3:22
  SPEAKER_01: We've got great teams, so it's quite quite fun.
• 3:24
  SPEAKER_01: Lots of exciting things happening soon.
• 3:26
  SPEAKER_00: It's interesting because it was this conglomerate at the beginning, you know, it was eBay and eBay PayPal and eBay PayPal Skype.
• 3:32
  SPEAKER_00: They were all the same company, and I knew a lot of people from back then because I worked in Yahoo, we worked with Skype, and we worked with PayPal on stuff as well, and of course, eBay was a thing that promoted with us, and it was just that especially in London, there was this big group of people that were all eBay, PayPal, or Skype at one time, and they all spread out and they're all now at Meta, at Google, and at Tropic, and these kind of things.
• 3:55
  SPEAKER_00: So it's once again an interesting thing where people say, like, this is the boring dot-com first thing.
• 4:01
  SPEAKER_00: Uh, and you're like, no, but it actually is infrastructure that is absolutely necessary.
• 4:05
  SPEAKER_00: And I do uh I do agree with you.
• 4:07
  SPEAKER_00: It's great to be part of a team, to be, to be uh feeling of uh putting something in and basically um um being part of a larger group and being able to delegate to experts that are better than you, and not having to uh when you're a contractor, you come in in, they always expect you to do everything.
• 4:26
  SPEAKER_00: And you have to you have to make sure that you actually actually believe that you're the only one who can do it, so otherwise you don't get the contract anymore.
• 4:32
  SPEAKER_00: So, in companies, it's much better to spread yourself out and actually avoid the red bus problem.
• 4:37
  SPEAKER_00: Like if you get run over by a bus, there's lots of people that still know what you've done and what you've uh worked on, and especially with developer advocacy, this is the bit that that uh got lost by the wayside.
• 4:50
  SPEAKER_00: I think the internal communication, the being there for developers inside your company and uh to allow them to shine outside as well is the is to me a huge part of developer relations.
• 5:00
  SPEAKER_00: But when you talk to people at DevRelcon, for example, they're all like, Yeah, I've I've worked for this startup for three weeks, and now I'm working for that startup for three weeks, that startup.
• 5:08
  SPEAKER_00: And I'm like, that's just marketing, you're not doing developer relations there because you don't even get yourself time to get into the relations.
• 5:15
  SPEAKER_01: Absolutely.
• 5:16
  SPEAKER_01: You it takes time to like understand um the product, the platform, the nuances, and also the other people, like you said, there are people who are experts in specific parts of the product and also what interests them as well.
• 5:29
  SPEAKER_01: And so even if you had you know four, so we've got a team of four devrails, but we're all very different, we're all interested in different parts of the platform, we're all interested in in doing things slightly differently, but we all add value.
• 5:39
  SPEAKER_01: And I think that's the important thing.
• 5:40
  SPEAKER_01: And and we should work to our strengths and also try to improve our weaknesses is that if that's something that you know we want to do, and I think that that is important.
• 5:47
  SPEAKER_01: Um, so yeah, it's it is great.
• 5:49
  SPEAKER_01: And I love it when the team is collaborative and supportive.
• 5:52
  SPEAKER_01: I think that just uh makes it so much more fun and you enjoy it.
• 5:55
  SPEAKER_01: And it does uh tie into what I love doing on the side as well, like you know, building in public, open source, and and I love it when they can complement each other when my day job and my kind of evenings and weekends for those of you shouting at the screen, I do love work-life balance and we will get into the digital nomad side and and you know, working from the beach of a laptop, all that great stuff, uh and bad stuff as well.
• 6:17
  SPEAKER_01: But um yes, so I just love it.
• 6:19
  SPEAKER_01: They complement each other.
• 6:20
  SPEAKER_01: So that's that's really useful.
• 6:22
  SPEAKER_01: Um, so I feel like both work faster, as in my my side and hobby projects, my open source projects can go faster, but also my my day job can go faster.
• 6:30
  SPEAKER_01: So I do like that overlap.
• 6:31
  SPEAKER_01: I know some people don't, they like to keep it very, very separate.
• 6:34
  SPEAKER_01: So if their day job is a I don't know, full stack or you know, full stack devil, they want to do mobile or something completely different, or say completely different to people outside tech, it all probably all sounds the same, but but people in tech, it's different to integrity, or or make games or something like that, as in it's still probably you know coding and vibe coding and all that good stuff, but in a different area, a different subniche, I suppose.
• 6:56
  SPEAKER_01: But I like to keep it very similar.
• 6:58
  SPEAKER_01: Um, so yeah, that's so that's me.
• 6:59
  SPEAKER_01: Um but I when you were both talking about the whole packing thing, and I was like, oh, been a digital nomad for now six years, and um my packing has got better, but I am the worst packer.
• 7:10
  SPEAKER_01: I pack way too much, like Chris said.
• 7:13
  SPEAKER_01: I always want like a backup of a backup of a backup.
• 7:16
  SPEAKER_01: And when you're when you're traveling for I like to do slow travel, so um I like to travel for at least one month to a location because I feel that gives me time to unpack.
• 7:23
  SPEAKER_01: Otherwise, anything less than that, I live out of a suitcase, just like Chris said.
• 7:26
  SPEAKER_01: I just take what's on top and that's it.
• 7:28
  SPEAKER_01: Um, so at least a month.
• 7:30
  SPEAKER_01: And um I take too much with camera equipment, you know, clothes, tech stuff, and the extra cables that I don't need, but just in case a cable breaks, um, it does add up to weight.
• 7:42
  SPEAKER_01: So my first trip was during the pandemic.
• 7:44
  SPEAKER_01: I went to Bali for six months and I traveled with about 160 kilos of stuff, and it was just, you know, way overkill.
• 7:52
  SPEAKER_01: I took so many lights, I took three cameras just in case one camera broke, and oh my god, I just took way too much.
• 7:57
  SPEAKER_01: Looking back, I spent so much in excess luggage, like it just wasn't worth it.
• 8:01
  SPEAKER_01: And then also the insurances and everything else.
• 8:03
  SPEAKER_01: Um, and then on my my next trip for six months was a bit lighter and lighter, and now we've got it even lighter.
• 8:08
  SPEAKER_01: So we've even got it now down to some giant duffel bags.
• 8:11
  SPEAKER_01: But on our next trip in a few months to to Thailand for um at the next year over winter, we are gonna try and make it even lighter.
• 8:19
  SPEAKER_01: We try and knock off like at least kind of, I don't know, 25% every trip.
• 8:24
  SPEAKER_01: And we've we have been doing that, we have been getting better.
• 8:26
  SPEAKER_01: So we're we're not experts yet, but we are getting there slowly.
• 8:30
  SPEAKER_02: Eddie, at the end, you left out the numbers.
• 8:32
  SPEAKER_02: Uh he said two big uh duffel bags, and I'm going, we started at the 160 kilos.
• 8:36
  SPEAKER_02: I mean, are we working with a 23 kilo allowance or what are we talking about?
• 8:40
  SPEAKER_01: Yeah, so it was 20, it's actually it was 23 kilo allowance, but I think it was about 24 and a half, I say 25.
• 8:47
  SPEAKER_01: Um, and I am also in addition to my tech stuff, which I have reduced a lot and and upgraded to to kind of combine uh devices.
• 8:55
  SPEAKER_01: Um, we do also travel with motorbike gear now as well, which is full bikes, jeans, jackets, helmets, everything.
• 9:02
  SPEAKER_01: So that does add an extra kind of seven kilos.
• 9:05
  SPEAKER_01: Although I cheat, but I wear all mine on the plane.
• 9:08
  SPEAKER_01: Kid you not, my wife thinks I'm crazy, but I literally wear armor jackets and everything in the plane.
• 9:13
  SPEAKER_01: You see me sit, I'm protected, right?
• 9:14
  SPEAKER_01: It's extra safe.
• 9:16
  SPEAKER_01: So that is 70, but our next trip, we want to get down to 50.
• 9:19
  SPEAKER_00: That's our goal combined.
• 9:21
  SPEAKER_00: I always want to, I always want to, I always want the next James Bond to be like that, to be like more grounded in reality.
• 9:27
  SPEAKER_00: He's only allowed to fly with Ryanair.
• 9:29
  SPEAKER_00: So you see like James Bond in a tuxedo and 12 pack, 12, 12 coats on top and everything in his pockets.
• 9:35
  SPEAKER_00: And like it's always quite fun when you see people flying with the EasyChat or Ryanair or all the other bits.
• 9:40
  SPEAKER_00: Like, it's just amazing how they tried to get to get away with like five layers of clothing and putting things into in the pockets.
• 9:47
  SPEAKER_00: The other thing is, of course, the Instagram ads.
• 9:50
  SPEAKER_00: Like I had that over the years as well, especially when I traveled so much, uh, as well.
• 9:54
  SPEAKER_00: The the amount of horse shit that Instagram offers you as like a travel great clothing hack, suitcase, foldable, black backpack, blah.
• 10:03
  SPEAKER_00: They're all shit.
• 10:05
  SPEAKER_00: Not a single one of them was good.
• 10:06
  SPEAKER_00: It's just like it's unbelievable.
• 10:09
  SPEAKER_00: Just go to go to a container store in America, buy everything from Eagle uh Creek, and that basically is the stuff that you pack your clothes in, and that's all you need.
• 10:17
  SPEAKER_00: But the the amount of like the this jacket has like 16 different compartments and do these kind of things.
• 10:22
  SPEAKER_00: You're like, yeah, this means you're gonna you're gonna you're gonna look for your house keys for four and a half hours when you get back because you have to go through all these pockets.
• 10:32
  SPEAKER_01: One travel hack that my wife did, which works and doesn't work.
• 10:35
  SPEAKER_01: Um, and this is the maybe the engineer in me.
• 10:37
  SPEAKER_01: So she said, You want to do it all in the stuffle bag?
• 10:40
  SPEAKER_01: I've reduced it all in the stuffle bag, and she bought these really vacuum-tight packing cubes that make everything like super small.
• 10:47
  SPEAKER_01: And that's great, but it still weighs the same.
• 10:49
  SPEAKER_01: Like it didn't get any lighter.
• 10:50
  SPEAKER_01: So it's like it's smaller, it fits in the bag, but it still weighs too much.
• 10:55
  SPEAKER_01: Is it also super wrinkly?
• 10:56
  SPEAKER_00: Like when you when you do the ones with the air out.
• 10:59
  SPEAKER_00: I think double basically crees everything up, right?
• 11:02
  SPEAKER_01: Uh, I don't, I don't really look.
• 11:04
  SPEAKER_01: I just give it a shake and I and I'll wear it and I'm happy.
• 11:06
  SPEAKER_01: Like this is island style, like everyone's in wrinkly clothes or something, so it's not a problem.
• 11:10
  SPEAKER_01: By the time I find a bike jacket on it, on top of it, it's going to be wrinkled anyway.
• 11:14
  SPEAKER_01: So she does iron stuff when she's out there, but I'm not too fast.
• 11:17
  SPEAKER_01: I I love the island style vibe of just just being super chilled.
• 11:22
  SPEAKER_02: I love the fact that we're developers talking about how we uh um how we can in the age of AI, we had somebody last week saying something like, you know, it's amazing that we can do everything with AI, but can we get something to fold my clothes?
• 11:32
  SPEAKER_02: And I love that we've like reverted back to like the various ways that we sort of pack our clothes and fold them, you know.
• 11:37
  SPEAKER_02: We need AI to sort this out.
• 11:39
  SPEAKER_00: But that's why every humanoid robot is like the test thing right now, the folding folding a shirt, right?
• 11:44
  SPEAKER_00: That's what all the different demos from China are showing and stuff.
• 11:47
  SPEAKER_00: And you're like, yeah, but you know, they just want to buy buy your robots for killing people, they don't care about the folding ones, but that's a different story.
• 11:55
  SPEAKER_00: Okay, first thing, Euronews.
• 11:57
  SPEAKER_00: Um, EU places chat GPT Reddit and Roblox under strictest digital safety rules.
• 12:03
  SPEAKER_00: The interesting bit in that article I found is that they actually rule ChatGPT as a search engine and thus give it the same uh uh responsibilities as a search engine.
• 12:16
  SPEAKER_00: And I wondered if that's actually not a proper uh evaluation of it because people use chat GPT instead of search, right?
• 12:25
  SPEAKER_00: Or what do you sue ChatGPT as?
• 12:28
  SPEAKER_00: Is it a replacement for search or is it more like a generative tool?
• 12:32
  SPEAKER_00: Or what do you think it would be the when it comes to like having to license it or having to give it the right uh legal environment?
• 12:39
  SPEAKER_00: What would chat GPT be?
• 12:42
  SPEAKER_01: I I think it's a bit blurred.
• 12:43
  SPEAKER_01: I don't think it's like clean cut anymore because you're right, some people use it for generative, some people use it for research, and some people use it for search.
• 12:50
  SPEAKER_01: And I probably combine those depending on what I'm doing.
• 12:54
  SPEAKER_01: Um, I use Claude as well, and uh it's a similar sort of thing.
• 12:58
  SPEAKER_01: Um, I still use Google.
• 13:00
  SPEAKER_01: I don't know why for certain things I still search.
• 13:02
  SPEAKER_01: I'm a bit old school with that.
• 13:03
  SPEAKER_01: Maybe that's just like old habit.
• 13:04
  SPEAKER_01: So I think it's tough.
• 13:05
  SPEAKER_01: I think the lines are really blurred.
• 13:07
  SPEAKER_01: Um be everyone wants to make the I mean the what I took away from this was that you know, everyone's like, I want to make the next big massive company, be the next first solo billionaire.
• 13:17
  SPEAKER_01: It's like, well, you've got problems like this to deal with.
• 13:19
  SPEAKER_01: Um, I just want a lifestyle business that's just you know making enough to live off and and all the rest.
• 13:24
  SPEAKER_01: So I don't know, being the biggest isn't always the best.
• 13:27
  SPEAKER_01: I know it'll trickle down into smaller um companies as well later on.
• 13:31
  SPEAKER_01: But yeah, it's tough.
• 13:32
  SPEAKER_01: I don't know.
• 13:32
  SPEAKER_01: What what do you think, Daniel?
• 13:34
  SPEAKER_02: I mean, I I I think that you kind of have to accept that on some level it's a search engine, engine, whether or not it's your primary use of it, it's it's something that's a core feature of it.
• 13:44
  SPEAKER_02: You know, you can find something, it will return results.
• 13:46
  SPEAKER_02: Um we know as well that there's now become it's become the new kind of SEO, things like GEO, people wanting to kind of be the uh the I suppose the link of choice for these kind of chatbots.
• 13:57
  SPEAKER_02: Um so I think it's it's not really that different to what it, you know, the old the old days of SEO and just trying to be at the top of the Google results or whatever your search engine was.
• 14:05
  SPEAKER_02: Um I mean I suppose that uh for me it's that there will be a lot of knock-on effects from the legal side of this that like I that I'm not gonna be, you know, that I'm not gonna try and speak on.
• 14:13
  SPEAKER_02: Um but it's interesting because it does show really that these things are moving really fast and regulation feels like it's actually moving more quickly than we would expect.
• 14:22
  SPEAKER_02: Like that's always the criticism.
• 14:23
  SPEAKER_02: Things move very slow, we don't regulate fast enough.
• 14:26
  SPEAKER_02: But it does sound like actually that you know that there is a recognition that these things are like powerful, but we need to try and figure out what they are sooner rather than later, because the impact could be disastrous if we don't.
• 14:36
  SPEAKER_00: I mean, here's the reason.
• 14:37
  SPEAKER_00: I mean, it already resulted in a in 870 million euro in fines.
• 14:42
  SPEAKER_00: So, yeah, of course, it moves faster if that's the outcome of it.
• 14:45
  SPEAKER_00: Other regulation that has less of that impact might not be as fast, but it's interesting to see.
• 14:50
  SPEAKER_00: Uh that the Reddit thing is also fascinating to me because Reddit to me was always like a glorified forum, you know, like it was just basically like people write random stuff and there's wonderful fallouts of it.
• 15:03
  SPEAKER_00: I mean, I'm a big fan of Imager, which sadly enough, England uh Danny can't see anymore.
• 15:08
  SPEAKER_00: But I love, for example, that there's the 30 wholesome memes of the day, and people just show nice things.
• 15:14
  SPEAKER_00: And I love that there is these niches in there, but there's also apparently a dark side to Reddit that that needs regulation that I never encountered because I never got into it that much.
• 15:24
  SPEAKER_00: But it's fascinating.
• 15:26
  SPEAKER_00: I don't even know what Roblox is.
• 15:27
  SPEAKER_00: That was a kid's game, wasn't it?
• 15:29
  SPEAKER_01: I think so, and I think there's a problem with um because kids are using it and there's there's a social aspect to it, so I think there is some challenges around that.
• 15:37
  SPEAKER_01: I don't go deep into that either because that's not something I use and I don't have kids, but that's from what I've under under briefly heard from the grapevine.
• 15:44
  SPEAKER_00: Now, the other big thing that a lot of people talked about, of course, is that open AI leaving Cursor because Cursor got bought uh by SpaceX and now OpenAI pulls everything out, and everybody who used Cursor as their main driver to code things now can't use all the models anymore.
• 16:00
  SPEAKER_00: And basically they have to prepare to adapt to it when it happens.
• 16:03
  SPEAKER_00: And I find that a very cynical way of saying it.
• 16:06
  SPEAKER_00: Um but also at the same time, you shouldn't be surprised, you know.
• 16:10
  SPEAKER_00: Like we we make ourselves dependent as developers to certain companies and IDEs and uh and models, and then when they pull the rug from under us, it's like there's no point in whining.
• 16:21
  SPEAKER_00: It's like it's a bit like uh uh doing cloud doing cloud computing or hosting, like always be prepared to have a backup or to to to s use your data somewhere else and have your data somewhere else.
• 16:34
  SPEAKER_00: That's why I keep telling people put everything on your personal blog and then put it on social media, because the Facebooks, the uh the Instagrams, and all these sorts of world might change immediately and get away from it.
• 16:45
  SPEAKER_00: Um, I mean, I don't use cursor.
• 16:47
  SPEAKER_00: Uh Dan, I guess you don't use cursor either.
• 16:50
  SPEAKER_00: Do you use cursor?
• 16:51
  SPEAKER_01: No, I used to use it um when I think when it first came out, it was it was quite interesting.
• 16:55
  SPEAKER_01: I used it then, but no, I haven't used it in a in a long time.
• 16:59
  SPEAKER_01: Um and and my my my products that use AI, um, I've created adapters, so I've abstracted away so I can switch between um any model very, very any model, any platform very easily.
• 17:11
  SPEAKER_01: And I purposely do that just to make sure it's it's still working.
• 17:15
  SPEAKER_01: So I I rotate monthly, or if my um if I want to, you know, use a certain um uh model or I want to get certain result or or you know, I don't know, I spend too much on one platform, I I divide it up.
• 17:28
  SPEAKER_01: So yeah, in in my admin pages, I can just switch across between you know OpenAI, even places like Venice AI, um or uh um or anthropic.
• 17:38
  SPEAKER_01: So um yeah, I try to abstract that away so I'm never at the mercy of that.
• 17:42
  SPEAKER_01: Because if you put uh you know all your eggs in one basket, or as you were saying, Chris, if you put your blog posts on at least certain platforms and not on your own, you're kind of social media, you're on rented land, you know, they can just not renew the what's the what's the term, not renew the the lease on the land, let's say, and then you know what you what you're gonna do.
• 17:59
  SPEAKER_01: So yeah, I think people need to use like you know, design as an architectural design patterns that that help have this abstraction.
• 18:06
  SPEAKER_01: And I think that's uh yeah, really important.
• 18:10
  SPEAKER_00: I mean, it says here it's November, so it's really not nobody got hit immediately by that.
• 18:15
  SPEAKER_00: Given given the outrage that I saw on LinkedIn, I thought it would be the end of the world for so many people.
• 18:20
  SPEAKER_00: But yeah, at the same time, I mean they they violated their terms of contract and all kind of things.
• 18:25
  SPEAKER_00: But yeah, it's uh I'm a big fan of local hosting and uh of models and stuff and using local inference and these kind of things.
• 18:33
  SPEAKER_00: I mean, so I'm I'm still waiting for my Nvidia GTX part, but once it's there, I'm gonna be have my internet on my house, and that's basically there.
• 18:41
  SPEAKER_00: They yeah, they uh they they uh they gave me one uh at at our congress as a thank you for doing all the stuff with them.
• 18:47
  SPEAKER_02: So keep that quiet because that's the sort of thing, Chris.
• 18:51
  SPEAKER_02: Your nice new house.
• 18:52
  SPEAKER_02: You don't want people knowing that's there because that's gonna be very valuable, if only for the copper, but it's gonna be more valuable for other reasons.
• 18:58
  SPEAKER_02: But yeah, I wouldn't tell too many people.
• 19:00
  SPEAKER_00: The area I live in Berlin is only people above 60.
• 19:03
  SPEAKER_00: They they're just actually more worried that they have to buy electric cars now and not normal ones.
• 19:07
  SPEAKER_00: So they're they have other issues to deal with and the elections at the moment and these kind of things.
• 19:12
  SPEAKER_00: So there's a new Twitter, uh, because apparently um it sounds it looks a bit weird.
• 19:19
  SPEAKER_00: Um it looks like uh uh Master Elon uh uh didn't quite take uh care of uh renewing the the design and everything, so somebody took that on.
• 19:30
  SPEAKER_00: I tried it, I tried to try it out, but um it was offline when I tried it.
• 19:36
  SPEAKER_00: I hope it's gonna be the same as the old one, but um yeah, um I don't think that people flock to it that that easily, but uh it's gonna be interesting to see.
• 19:48
  SPEAKER_01: I think I think it's really interesting, I think it's really funny.
• 19:51
  SPEAKER_01: Um, because I'm a huge you know Twitter fan.
• 19:54
  SPEAKER_01: I have been on the platform for years, and and since my digital nomad life, I actually invested a lot of time on it, had a whole strategy on it, you know, grew to 200,000 followers really quickly on it.
• 20:06
  SPEAKER_01: Um and I and I like the old Twitter.
• 20:07
  SPEAKER_01: I mean, I still use the the new Twitter if that's what we're gonna call it.
• 20:10
  SPEAKER_01: But I thought this was really funny because you know, it didn't seem like for I'm no lawyer, but didn't seem like they they broke any any rules per se.
• 20:18
  SPEAKER_01: Um, but no one's ever gonna go to a new platform that quickly.
• 20:21
  SPEAKER_01: I mean, there was Blue Sky, there was what some of the other ones I forget around um you know, like five minutes.
• 20:27
  SPEAKER_00: There's Mastodon, there is uh uh Dick said they're kind of come back as well.
• 20:32
  SPEAKER_00: And uh yeah, there's uh there's uh um there's threats, which is basically threads, yeah.
• 20:38
  SPEAKER_00: Yeah, which basically everything you put on Instagram without the images.
• 20:43
  SPEAKER_01: Yeah, so they didn't, I don't feel they were ever a threat to Twitter or X.
• 20:48
  SPEAKER_01: So I don't know, I think they're overreacting, in my opinion.
• 20:51
  SPEAKER_01: I think they shouldn't just worry about it.
• 20:52
  SPEAKER_01: It's it's so hard to have a two-sided like marketplace as it were, right?
• 20:57
  SPEAKER_01: It's not just like a blog, you're you're it's one-sided, you hope people visit, but it still works by itself.
• 21:03
  SPEAKER_01: But in this, you need both sides, you need people creating content and people engaging in the content, otherwise, people won't post more content.
• 21:08
  SPEAKER_01: So it's very hard to create and get moving, get momentum.
• 21:11
  SPEAKER_01: So I don't know, I think they're overreacting.
• 21:12
  SPEAKER_00: Yeah, I mean, you got the got the Fattiverse stuff.
• 21:15
  SPEAKER_00: I mean, I have now all the Fattiverse plugins for my WordPress blog, so I get a lot of comments and I get a lot of people reposting and these kind of things, and I don't have time to look at that.
• 21:24
  SPEAKER_00: But yeah, I've got like 8,300 comments out of a sudden I haven't read, and most of the time it's just like like this, re-block this, these kind of things.
• 21:31
  SPEAKER_00: And I could display all their images, but I don't see the point of that to a degree.
• 21:35
  SPEAKER_00: But I like the fact that it goes back to the uh that you want to go back to the original Twitter because Twitter was a warm and interesting uh uh um uh environment.
• 21:46
  SPEAKER_00: Like you felt fuzzy, like the whole like fail whale and these kind of things, and taking follow Thursdays and the follow Fridays and these kind of things from the audience, much like uh Imager has as well.
• 21:56
  SPEAKER_00: There's a lot of things there that basically the community comes up with, and then they make a feature.
• 22:00
  SPEAKER_00: the platform and uh with X I don't see that and I don't have I'm still I'm still locked out of X at the moment and I'm trying to get in because they charged me again uh but uh I don't see the point of it right now.
• 22:14
  SPEAKER_00: I mean you said like yeah for I had 65,000 followers there as well and uh I basically basically because I curated them I didn't let anybody just go there.
• 22:23
  SPEAKER_00: I mean I could have done like a lot much more as well but now it seems to be like bots and uh and growth hackers uh shaking hands there and that's about it and in between some horrible political uh and uh uh and very limited views that basically get a lot of hits.
• 22:39
  SPEAKER_00: I've I've I've written about it on my blog a few weeks ago that there's an antisocial network happening right now instead of antisocial web instead of social web because it's all about like rage baiting and making a a blanket statement so people get annoyed with it rather than just inviting a conversation.
• 22:57
  SPEAKER_01: I think the reason why that happened was when they started offering payouts on X, I noticed a massive shift and people just wanted to like you said rage bait just get not to get the followers but they just wanted to get the views so they can then get a payout and I I do see people with considerably less followers than me getting a bigger payout.
• 23:15
  SPEAKER_01: And I don't mind you know I'm not there for to get a payout I'm there to socialize with the tech community and and I've curated my timeline really well so I don't see any politics I've muted certain words I just I'm there for tech.
• 23:26
  SPEAKER_01: But then you do occasionally see the the people that are just you know content was it click farming just to try and get the the the payouts and it's just the things being said are getting more outrageous and outrageous all the time.
• 23:39
  SPEAKER_01: I mean I think I saw something come up the other day goes oh I made my I vibe credited my first app in a weekend and I made my first seven million mm r in the first three days I was like no you didn't like you didn't like it's just getting more exaggerated every time and if they did you know fair play to them but cost me I don't think you did um and it's just I love to see the success stories and the failure stories but I don't like to see the BS stories and it's just very hard to tell I think the the lines are quite I mean we can tell sometimes but when it's like a bit in the middle it's more difficult to tell um but on that note I was watching a YouTube video in the gym the other day and um the the the host opened the video podcast and said thank you so much for joining us you know person X etc and so on uh but you know you were overnight success.
• 24:26
  SPEAKER_01: Tell us how you did it.
• 24:27
  SPEAKER_01: And they went I've been doing this for 10 years.
• 24:29
  SPEAKER_01: You all only heard about me when I became successful.
• 24:31
  SPEAKER_01: So I'm not an overnight success.
• 24:33
  SPEAKER_01: And I thought it was really interesting.
• 24:34
  SPEAKER_01: And that's the stuff that I want to hear because people don't hear the 10 years before you knew the person if that makes sense.
• 24:41
  SPEAKER_00: Claude cutting down 50 70% weekly limits.
• 24:45
  SPEAKER_00: I haven't hit them yet or my uh I don't know my boss hasn't hit me yet so because that I I didn't hit them.
• 24:51
  SPEAKER_00: But yeah it's uh uh it's it this is another weekly thing I see right now like everybody was like oh everything is free and everything is new and everything is now of a sudden all the companies realize that they're burning money so a lot of these limitations go down as well we've been hit by that by by a tool that we use for social media so we now have to have to do every two days instead of daily and kind of thing but yeah limits are a thing and uh I I would not be uh I would always uh err on the side of caution and say like okay this will not be available to me the whole time I mean this hit me hard I'm a massive clawed user like a like a heavy clawed user and I wasn't focusing on being efficient with it I was just in focus getting me output because I never hit my limit and all of a sudden doing the same thing I started hitting my limit so I was like right I need not do as much like exploratory prototypes just for fun.
• 25:43
  SPEAKER_01: Reduced that and I was under the limit and then you know a few weeks passed or very recently I'm being really cautious and I started hitting the limit I'm doing even less but the tokens just didn't seem to go as far and they reduced the limits um and now I'm trying to get more efficient with like keeping my context you know a lot cleaner and not wanting the context put up to a million and all this sorts of stuff making a bit more of a conscious effort to be efficient.
• 26:05
  SPEAKER_01: But this hit me quite hard and uh I just find it goes very fast now and I'm having to be super conscious.
• 26:12
  SPEAKER_01: I got some tips on this um which I'll I'll share later on but uh yeah I it hit me hard and it's like you said we can't expect this forever because you know they were working in a loss and they were just trying to get us all hooked and I am hooked I need my tokens okay uh so there's two things vibe coding and quality and AI coding will prevent expertise and Dan you were really thanking me for sending that to you because you were really excited about that article right yeah I really liked it.
• 26:41
  SPEAKER_02: I think that there's um a couple of really good kind of points in there and I actually just want to bring attention to that quote at the at the top because it's the one that's uh yeah we see a future where intelligence is a utility like electricity or water and people buy it from us on a meter and use it for whatever they want to use it for.
• 26:55
  SPEAKER_02: That's less so about kind of this topic in in particular but more so I just thought that's a that's a really kind of shocking thing for us all to be reminded of that's kind of what you were saying before about getting hooked and it becomes a utility like every single day you use it and that now we kind of need it, you know.
• 27:10
  SPEAKER_02: Now this article was thought was really interesting because uh it's essentially like looking really at kind of how people learn and how people move from beginner to expert.
• 27:18
  SPEAKER_02: And the thing that I thought was really interesting in there was that they're essentially saying that because we're using AI that it becomes like you're saying Eddie about the output and it's not really about the journal the journey along the way or it's not about learning through pain.
• 27:32
  SPEAKER_02: And there was a really good quote that I just want to see if I can find on there which was um it's about taste uh there it is it's under the expert novice um header and it's a really good one where it's um in the sort of third paragraph it says about the skills to to kind of become an expert um or a function of someone who's experienced the friction and challenges over time that culminate in good taste.
• 27:53
  SPEAKER_02: And I thought that is such a good way of sort of helping us to understand some of those things that we sort of just think people have as like a natural thing.
• 28:02
  SPEAKER_02: Like I almost think people look at it as like fashion taste.
• 28:05
  SPEAKER_02: You just think somebody knows how to dress like inherently and so they know how to build good software and it's like no this is this has come through wearing bad outfits lots of times and people telling you not to wear them or maybe you shouldn't wear that.
• 28:18
  SPEAKER_02: It's kind of in in the software world you have to have gone you have to have been in the trenches effectively and and really learn what the the the the detailed problems are that you're trying to solve not just the higher level stuff.
• 28:30
  SPEAKER_02: And that when we're using AI we tend to think in these kind of uh higher level uh vague kind of terms and we think the AI will fill in the gaps and that gap is actually the bit that I think is the important bit.
• 28:41
  SPEAKER_00: So it's a great article yeah and the buying intelligence on a meter is just very arrogant.
• 28:48
  SPEAKER_00: I think it's just a very it's like okay there's already pre-filtered intelligence according to your reality rather than just like a lot of things where where different people can get from different walks of life can put input in there.
• 29:00
  SPEAKER_00: Emotional intelligence is a different thing than just facts and uh uh taste is a very big problem as well.
• 29:06
  SPEAKER_00: And I think yeah the the the overall article is uh we've we've talked about that before as well it's just the overconfidence that people have right now in uh in getting their answers right for them rather than having to come to the conclusion why that is the right answer is a big problem when it comes to debugging later on or it comes to maintaining and also uh I think it uh in in uh it's it's a bit like social media addiction it's the same with that like you you basically you get this wow effect of being able to release something really quickly and fast but sooner or later it will kick in that you're actually realizing that you haven't done anything about this you haven't done that properly like we had we just had this uh on Facebook there was this group that I have on on like 8-bit computing like old Commodore 64 games and people did like AI slob images above and and uh and then were annoyed that people didn't clap and one of the answers was like it's like it's like being uh going in the 1800s going into a cafe and using a typewriter type A B C D and showing it to people and saying like look I'm a writer and you're like you know you you you don't uh if you don't get actually any um if you don't get any applause for AI generated stuff then because you don't deserve applause for it because you didn't do it the machine did and it's just uh fascinating the slot that's out there my favorite right now on Facebook is every two seconds there's a picture of uh uh of Shakira at the football world uh world cup with a different dog in her arm and she's like and it's like it's different dog groups like oh Shakira and her Irish setter Shakira and her puck Shakira and her and basically somebody generated millions of photos of Shakira holding a different dog and then posting them on all these groups and you're like this is such a waste of time and effort if for everybody involved including the poor dogs that don't exist it's just really bad the it's a good example you said with the typewriter um and the ABC that's like going back in time that's really really interesting um and yeah the the AI slop is just it I feel it is killing social media because it's just it's it's it's watering down the content I want to see and and it's just I spend less time on socials because of AI now.
• 31:20
  SPEAKER_01: And people like oh yeah I can grow my account because I can post a hundred times a day it's like I don't want to see that a hundred times a day.
• 31:26
  SPEAKER_01: So yeah it's it's it's it's tough.
• 31:30
  SPEAKER_00: It's definitely a mindful you're you're growing your real followers with it.
• 31:35
  SPEAKER_00: I mean I've been accused many many times that I post too much because I just did it myself people said like you have a basement full of children that type for you which is not that's not what they do.
• 31:45
  SPEAKER_00: But the uh the the the thing is like people were just like oh I can't follow you because you post too much and you're like yeah that makes sense but nowadays it's kind of different because you basically don't have time-based uh uh uh uh threads anymore in uh news threads anymore but you basically have like algorithm-based ones so it doesn't matter how much you throw out there you might get lucky that it gets in it's not like people see uh subscribe to you and see your stuff in one go anyways yes I love I love this one I love this one this is basically an app that uh analyzes everything you do on your machine and then at the end of the day you get a punch card printout uh basically showing how much you've used sim uh you've used uh which which app uh for how long utterly pointless but I find it kind of cute the animation was really good I liked how they were consistent with the punch card and the pricing it uses the same style of punch card I thought it was is very very clever beautifully done cool I wouldn't use I wouldn't use it unfortunately I just like you said it's it's it's you don't want to just you don't don't want to know that's true this is also really cool as well it's basically uh a a language to uh to automate a tool like automator on Mac can do so many things it was absolutely atrocious in terms of the language that it used it used to be uh Apple script which is taking programming in English and taking the worst of both and making a language out of it and now it's actually uh JavaScript as well but uh this one now allows you to actually say like I want to do this on my computer and then uh it actually generates an automated script for it to actually automate the thing and let it run on your machine.
• 33:30
  SPEAKER_00: I think that's a pretty cool uh way of doing it rather than just like set starting a new agent and having to feed it and this is for this is the ubiquitous computing thing for people who don't want to know that actually it makes kind of sense.
• 33:43
  SPEAKER_00: I don't know what trauma means or if it's trauma related but yeah it's uh um it's an interesting little tool.
• 33:51
  SPEAKER_00: And uh last but not least I love this one as well because it's a simple one it's basically color combination uh of two colors and then it actually shows you if different people with different color blindness issues or visual deficits can actually see those colors.
• 34:06
  SPEAKER_00: So that's something most accessibility tools have built in as well in terms of linting but this is actually a pretty nice little tool uh that actually uh shows you what what these colors look like and again I mean I spent a few years of my life into that so basically if you uh if you have edge and you want to see that as well you can always go to the simulation here as well and actually see the different color uh deficits as well so basically here you can say the preferred color schemes but here you can see also if people with different color blindnesses how they see the page that's built into the developer tools of the browser in this extra toolbar as well here.
• 34:45
  SPEAKER_00: So if you just want to do a test without reloading it we uh we spent three months making that happen and also blurred vision so basically you can blur the page and see like if you font is big enough or not.
• 34:56
  SPEAKER_01: That's a really good oh I didn't know that that's really nice.
• 35:00
  SPEAKER_00: Yeah it took us ages to get in so that's interesting to see but it's in the developer tools of Edge uh not Chrome but Edge.
• 35:06
  SPEAKER_00: Okay.
• 35:07
  SPEAKER_00: Did did you miss one?
• 35:09
  SPEAKER_00: You missed the first one the site yeah yeah I wanted to get to that one as well this is basically uh uh uh uh again kind of a gag app but it actually is like if your site project you haven't maintained it you can actually make a uh make a funeral here an online funeral that actually allows you to download the old code and basically say like it's not gonna happen anymore and I find that really really lovely like you can leave a rose you can light a candle and so on and so forth for all the projects that died and uh never went anywhere I think that's so interesting because most projects die either because we lose interest we move to something else or or nobody uses it.
• 35:48
  SPEAKER_01: So I was expecting this to be a lot busier um because I thought people like you know what I can get some traffic maybe from it or something to to my GitHub repo or to my website or whatever.
• 35:59
  SPEAKER_01: But so I thought it'd be busier but I I do like the idea it is very clever.
• 36:05
  SPEAKER_02: I don't say it's Dagabair is it Dagabare that's created this because I feel like this because it's got logology at the top yeah yeah yeah it's always got some interesting projects that are kind of in this sort of like half half funny or not you know so it's it's really nicely done.
• 36:19
  SPEAKER_00: Yeah and I think I think it's good I mean it's like I always talked about that when it came to enterprise companies and social media like when a project wasn't successful then they basically open sourced it and I'm like that's not open sourcing I called that farm sourcing you send it to the farm to live with the other unloved projects and basically somebody will maintain it and nobody ever maintains it.
• 36:41
  SPEAKER_00: So giving it a funeral and say like this is this is dead and here we go is probably the the more um honest way of dealing with it.
• 36:50
  SPEAKER_00: So the other thing we have is like we have uh a tool spotlight and that's also an article on the VR Developers magazine where you can read up all the bits and bobs that are not in the video.
• 37:02
  SPEAKER_00: One thing that puzzles me is that no matter how big a hard drive I buy it's always 70% full in no time.
• 37:09
  SPEAKER_00: The problem is that file tools that come with the operating system don't give you a simple view on sizes of folders and files.
• 37:15
  SPEAKER_00: Even worse in recent years operating systems started hiding files and want you to use app stores and assistants instead.
• 37:22
  SPEAKER_00: That doesn't help you though when the machine runs out of space.
• 37:25
  SPEAKER_00: A classic tool that fixes that on Macs is the commercial daisy disk.
• 37:29
  SPEAKER_00: It allows you to see hard drives as a sunburst graph and you can find large files that way and add them to a deletion queue.
• 37:36
  SPEAKER_00: Now there's also NeoDisc by Lucas Takayazu which is free and open source and has a few more ways to visualize your hard drives and file sizes.
• 37:44
  SPEAKER_00: It does not have a deletion queue but asks you to use Finder instead for deletion.
• 37:49
  SPEAKER_00: SpiralDisc by Adiileo Barone is another alternative and this one is cross-platform for Windows, Mac and Linux.
• 37:56
  SPEAKER_00: So go and find the Spacehawks on your hard drive and free those bytes.
• 38:00
  SPEAKER_00: That was the spotlight it's also on the magazine the three of them I didn't know about the other two like NeoDisc, something that came up DaisyDisc I've been using for years.
• 38:08
  SPEAKER_00: Ironically yesterday when I recorded the uh the social media part of that one the the uh the the the portrait one um my computer crashed because I was I I ran out of space so uh I should have run those tools before I actually recorded the video of them but uh I find it fascinating that that no operating system has cracked that actually some Linux distributions have a really good way of showing you where the files go and stuff but uh uh yeah both uh both Windows and Apple want you to use their cool fancy tools rather than think about files and as an engineer that always I found that uh annoying because I want to know where my data goes but yeah these are really good to find huge files and clean them up.
• 38:48
  SPEAKER_00: So we have headlines uh some of them are made up others I actually found so you have to tell us which ones are made up and which ones are fake.
• 38:56
  SPEAKER_00: I'm gonna be terrible at this oh my god the pressure I want to come last we'll see world first wind powered underwater data center starts operating in China wind powered underwater um but anyone doing some interesting things oh my god the fuck the fuck is ticking um we're gonna shoot uh I don't know if that's help or trick uh let's go let's go let's go fake let's go fake and oh it's news somehow it's news get a Waymo use how Berkeley used a robot taxi to flee the scene in a first of its kind San Francisco case news wow are you confident on that I am I don't know why we see a lot of Waymo in the headlines good and bad so yeah I'll go with you on that idea correct yep it's news scientists gave aggressive fish psychedelic drugs a breakthrough came next fake that's animal cruelty they can't do it it's news what yeah Google back researchers seek to transform retired phones into connected data centers I could see some truth in that but I could see it partly fake uh okay fake I'm going fake retired phones it's news and we we got click farms with retired phones so why not actually use them for data storage you know it's true wow Japanese energy firm uses drive with data of 10.9 million clients it's news it's true it does we see a lot of this don't we uh people USB stick on trains yeah and 10.9 very specific uh he's cutting on to your tricks here Chris you know it's these little red herring correct it is news yeah these are not just an AI music detector for other streaming services for other streaming services that's what's called me what do we think eight seven oh we haven't had any fake ones in a while so maybe fake yeah it's useful they actually offered it to them but they didn't want them Chinese drivers are using tiny plastic heads to fool Tesla's autopilot safeguards sounds real this is about this is about basically Tesla saying you're tired and you actually shouldn't be driving so they actually use like heads to feel alive and feel awake it's fake it's fake your heads in in the in in in the place right so it's number eight I'll get my coat this startup is training an AI model on skin that stays alive for a month come on give me a fake one it's fake give me a fake one please yeah it has to be I mean it's I mean it's very strange just such a strange headline though it could be real it is real oh my god to Ave I am coming last this is what I have to deal with every day like finding stuff like that and reading it and you're like really but yeah it's these are the real headlines Google Gemini quietly assigns uses a productivity potential score based on search history score report of the influences ad pricing used that's some wacky stuff that would happen for sure yeah evolve pricing right it could be let's have a look it's fake and next time let's just go for the opposite of whichever one we think yeah I think number 10 Chinese firm to send thousand robots into real workplaces to cook inspect and dispense medication Daniel Dan I'm leaving this for you like seriously my my my record is terrible um it's a reverse leaderboard so the more you get wrong the better you do probably um I would say this feels like news uh but I for that reason I want to say fake shall we go fake I trust you let's do it don't trust me big mistake this is so right so it's I think the new trick is don't choose the one we want and don't and then choose the opposite of the one we don't want right so it's like double negative that's why yeah I love that you're ganging up on me you're supposed to actually be in competition about this is fun as well right moral support I don't go on the leaderboard so I'm just moral support but I don't think I'm actually doing much support which is the problem sorry about that you're loads of support I appreciate it okay yeah come on last three we've got this eye of the tiger AI assistant translates ancient languages in segments true definitely and so we think it's fake but then that would be what Chris wants us
• 44:00
  SPEAKER_02: Things that we'll go for news.
• 44:01
  SPEAKER_02: Okay.
• 44:01
  SPEAKER_01: Exactly.
• 44:02
  SPEAKER_01: That's the right thing.
• 44:06
  SPEAKER_02: He got it to get in the great.
• 44:09
  SPEAKER_02: All right.
• 44:09
  SPEAKER_02: We've got one question left.
• 44:12
  SPEAKER_02: We're going to do it.
• 44:12
  SPEAKER_02: We're going to do it.
• 44:13
  SPEAKER_01: Okay, let's do it.
• 44:15
  SPEAKER_00: Amazon Web Services announces 15% price increase for GPU instances citing sustained AI demand.
• 44:21
  SPEAKER_01: It's got to be true, right?
• 44:23
  SPEAKER_00: It's got to be a good idea.
• 44:24
  SPEAKER_01: Price increase all these companies of AI and GPU, and it's increased, increase, increase.
• 44:34
  SPEAKER_01: Okay.
• 44:37
  SPEAKER_02: That was yeah.
• 44:38
  SPEAKER_02: Can we do another?
• 44:39
  SPEAKER_02: Let's do another.
• 44:39
  SPEAKER_02: Let's try again.
• 44:41
  SPEAKER_02: We're gonna have to keep that on.
• 44:42
  SPEAKER_02: Now what I will say is the beauty of this, Eddie, is when there's only three people on the leaderboard, you get yourself a bronze now.
• 44:48
  SPEAKER_02: You know, so it could be worse.
• 44:50
  SPEAKER_02: You're not that, you're bronze.
• 44:52
  SPEAKER_01: I like that.
• 44:53
  SPEAKER_01: Okay, I take it.
• 44:54
  SPEAKER_01: Positive attitude.
• 44:55
  SPEAKER_01: Love it.
• 44:56
  SPEAKER_02: Exactly.
• 44:57
  SPEAKER_00: Cool.
• 44:57
  SPEAKER_00: Good work.
• 44:58
  SPEAKER_00: Cool.
• 44:58
  SPEAKER_00: So that's bigger news for today.
• 45:00
  SPEAKER_00: So, Eddie, um, what do you have for us?
• 45:03
  SPEAKER_00: We we invited you for a reason.
• 45:05
  SPEAKER_00: I wasn't there, Dan invited you for a reason.
• 45:07
  SPEAKER_00: So, what was the reason that you want to uh to share with us that you basically want to woo us with?
• 45:13
  SPEAKER_01: I've got so many projects on the go, and I just thought I'll share a fun one for a few minutes and see what people think.
• 45:19
  SPEAKER_01: So let me um let me share my screen.
• 45:22
  SPEAKER_01: So, what I thought there's like I use carbon.sh like all the time, you know, creating um code blocks that look good in uh documentation, social posts, and all the rest.
• 45:33
  SPEAKER_01: And I just thought it would be fun to vibe code uh an animated one.
• 45:37
  SPEAKER_01: And so I just you know created this for a bit of fun as an open source project, almost at 50 stars, which is great.
• 45:43
  SPEAKER_01: And you can kind of configure it and see what you wanted to do.
• 45:47
  SPEAKER_01: Um and there's lots of configuration, I'm not gonna go for each one.
• 45:51
  SPEAKER_01: And then just exploring that a bit more, I wanted to see you know what we could do in terms of maybe more 3D.
• 45:57
  SPEAKER_01: I mean, if I just hit surprise me and just see this slightly different kind of 3D effect.
• 46:03
  SPEAKER_01: So I was having a bit of fun um with this.
• 46:07
  SPEAKER_01: Again, I don't know how many people would use this in documentation because maybe it's like you know, too much.
• 46:11
  SPEAKER_01: And it was great to see the community think, well, actually, you know, I want, I want the console, so I might want, sorry, the terminal, so I might want output, and then you'll see at the at the end, or if I skip to the end just to save time, you can see it at the bottom, which I think looks really good.
• 46:25
  SPEAKER_01: Let's go surprise me again and see what we can get.
• 46:28
  SPEAKER_01: Uh let's go something a bit simpler.
• 46:30
  SPEAKER_01: Maybe go clean.
• 46:31
  SPEAKER_01: Let's go simple for now, just so it's a bit easier to see.
• 46:34
  SPEAKER_01: And then other people came up with some ideas about annotating things as well, which I think was um, you know, really, really interesting.
• 46:40
  SPEAKER_01: So I thought, okay, that's pretty good.
• 46:42
  SPEAKER_01: So there was these annotations, so you could, you know, add comments, I don't know, test.
• 46:47
  SPEAKER_01: And uh I thought that was super interesting.
• 46:50
  SPEAKER_01: So it just grew from there.
• 46:51
  SPEAKER_01: It's a fun, you know, open source, open source product.
• 46:54
  SPEAKER_01: Um, and then someone said, Well, what about I want to save it to the cloud?
• 46:57
  SPEAKER_01: Because if I configure something and set it up, I don't want to uh have to then just you know put in the copy and pasting in the code, probably the easy part, but I don't want to have to then you know set up the console, set up the especially if you add more um annotations, for example, uh like this.
• 47:13
  SPEAKER_01: So I thought, okay, well, we can add a cloud version as well, you know, that'd be fun.
• 47:17
  SPEAKER_01: But I thought, right, you know, I work, I work for PayPal, so let's add some PayPal as well.
• 47:21
  SPEAKER_01: So I do sign in, so you can sign in um with with GitHub.
• 47:26
  SPEAKER_01: Uh, because at the moment in the public account, you can't save it.
• 47:28
  SPEAKER_01: So it was just a way for me to show what I've been working on with getting started guides with PayPal.
• 47:33
  SPEAKER_01: So when you should sign in, I've gone through the payment already.
• 47:36
  SPEAKER_01: So I've already gotten a pre-moon account.
• 47:37
  SPEAKER_01: It's just the PayPal button that everyone's familiar and uh used to.
• 47:41
  SPEAKER_01: And now you can kind of save and reload in the cloud.
• 47:43
  SPEAKER_01: So I thought that was uh a bit of fun, a great way to for me to um experiment to see because I've never done really any 3D animations before.
• 47:51
  SPEAKER_01: It's not something I do.
• 47:52
  SPEAKER_01: So I was trying to use AI to help me do that.
• 47:54
  SPEAKER_01: I was using my kind of experience in keeping things secure in terms of the you know, the database, and then using what I do as my day job with PayPal and bringing it all together.
• 48:04
  SPEAKER_01: This is why I was saying right at the beginning, where I like to kind of combine my side projects with my day job and and sprinkle in something a bit new in there, that's you know, for example, like a like a new technology or something like that.
• 48:16
  SPEAKER_01: Um, and for those of you who you know want to use this, please go ahead.
• 48:20
  SPEAKER_01: It's um open source on GitHub.
• 48:22
  SPEAKER_01: Uh and like I said, almost at 50 stars as well.
• 48:25
  SPEAKER_01: But the other thing I wanted to show people that I was actually going to make a video on on my full journey in making this and the challenges that I had, and actually got different models to make different versions of this at the beginning for the mock-up to see which ones I like more.
• 48:37
  SPEAKER_01: Um, but I also want to show people that uh PayPal has an AI toolkit where they've got uh MCP server and it's got skills.
• 48:44
  SPEAKER_01: So therefore, adding PayPal to this that people will see in the video is actually the shortest part of the project because it just kind of went ahead and did it and I just added my secret, uh my client ID and my secret.
• 48:56
  SPEAKER_01: And I want more open source projects to add payment to it because having been in open source for before there was Git and GitHub a long time, I mean, same with both of you.
• 49:07
  SPEAKER_01: I know Chris, um, you see a lot of your talks as well, you know, about a lot of open source things.
• 49:12
  SPEAKER_01: People work really hard and I want them to, I want it to be sustainable.
• 49:15
  SPEAKER_01: I really want open source to be sustainable.
• 49:17
  SPEAKER_01: So if anyone takes anything away from me today, just please add a payment to your open source projects, even donations uh or add some extra premium features.
• 49:26
  SPEAKER_01: Sometimes people do want to pay for those extra features and you and you can make a um a bit of money on the side so you can keep working on your project or work on your next project.
• 49:35
  SPEAKER_01: So, yes, I think that's uh super interesting.
• 49:38
  SPEAKER_01: I one thing that I've always failed on a lot of my kind of side projects, even ones where I wanted to do it as uh have a have a premium or freemium models, but for the premium part, I never added it on until like a year or two later, and it's just too late then, and that's one thing that I think I missed out on.
• 49:56
  SPEAKER_01: So definitely learn from my mistakes.
• 49:58
  SPEAKER_01: Any what are your thoughts and questions?
• 50:01
  SPEAKER_00: Well, I mean, I used to have a PayPal button in like 2001 on my blog, I think.
• 50:06
  SPEAKER_00: Uh, and like that back then they still worked.
• 50:08
  SPEAKER_00: So basically, say like donate with PayPal, and I think I made about$300 a month with it or something like that.
• 50:14
  SPEAKER_00: Now you got like Ko-Fi as well, or you got like Patreon, of course.
• 50:18
  SPEAKER_00: That's what most people use.
• 50:19
  SPEAKER_00: So I guess going back to that simplicity of just like saying putting it in there.
• 50:24
  SPEAKER_00: In terms of this tool, we had uh another guest on the show that actually does an animated um uh code displayer, but it's basically it allows you to show it step by step and make uh annotations at this as the thing, so it builds it like uh so it is a walkthrough thing, and he's been working on that for a year right now.
• 50:43
  SPEAKER_00: I mean, I love that this is basically making the the idea of a code display being readable and easy to take in completely impossible.
• 50:50
  SPEAKER_00: So that's an interesting way of of uh subverting the uh the expectations of your of your doc readers, but uh I mean it shows that you had fun playing with with with e-stuff.
• 51:00
  SPEAKER_00: So that's using WebGL or what's uh uh what's the the thing that you're using?
• 51:05
  SPEAKER_00: Are you using WebAssembly there?
• 51:07
  SPEAKER_01: It was WebGL, yes.
• 51:09
  SPEAKER_01: Yeah.
• 51:10
  SPEAKER_01: Um I experimented different things and it was um super interesting.
• 51:14
  SPEAKER_01: And some of them were really kind of resource hungry and and some weren't.
• 51:16
  SPEAKER_01: So it's just finding that balance between finding something that that that's fun, but also um didn't uh wasn't too complicated, let's say.
• 51:26
  SPEAKER_01: Um and doing it open source, it was great for people to make suggestions.
• 51:29
  SPEAKER_01: I mean, they wanted a uh a file uploader, so someone contributed so they didn't have to copy and paste in.
• 51:33
  SPEAKER_01: They could do a file uploader.
• 51:35
  SPEAKER_01: Some people suggested also having having steps, so therefore, as it goes through the steps, um maybe that's not let me just slide in, it's probably a bit easier, but going through the steps, it would actually um kind of look more animated as it would go through the steps, which I thought was uh quite interesting.
• 51:52
  SPEAKER_01: I should probably simplify this the animation a bit, but yeah, it was definitely good fun.
• 51:56
  SPEAKER_01: I have to have a look at the one that that you mentioned as well, because that sounds um yeah, really, really interesting.
• 52:03
  SPEAKER_00: And it also, I mean, it does the whole like optimization of the generated gif as well, like uh basically um uh making sure that it actually that it's as as small as possible because there was this big thing as well that people we always had the issue like how do we display source code?
• 52:20
  SPEAKER_00: We either display it in a pre-tag with like uh with like as source code, then it's kind of accessible, but then again, a screen reader user doesn't want to read every character of that code as well.
• 52:30
  SPEAKER_00: And then we had the problem that to color code it, we have to put spans around it like hell.
• 52:35
  SPEAKER_00: Now there is this range CSS thing as well, and Dave Rupert just released a thing yesterday last week that did the rounds heavily because it's actually not using any spans or any diffs or anything in it, but it just it just uses the CSS engine itself to colorize the different parts of it.
• 52:52
  SPEAKER_00: Then, of course, uh, I mean in slides I did screenshots and basically showed them, but then you have the problem that if you have a high resolution, like a 4K display 50 meters wide at our conference, it's blurry as hell.
• 53:04
  SPEAKER_00: Uh, and then you basically have that issue.
• 53:07
  SPEAKER_00: Then there was an SVG animation one that basically instead of using a GIF animation, you use an SVG that actually displays it bit by bit because SVG has built-in animation, and of course it uh it renders the font uh seamlessly, so basically it doesn't have the blurry problem.
• 53:22
  SPEAKER_00: But yeah, it's always the issue like how do you display source code?
• 53:25
  SPEAKER_00: But then again, uh I go back to when I wrote docs like Edge and these kind of things.
• 53:30
  SPEAKER_00: Like, if you have to show a ton of tech code in your docs, then probably your API is too complex, too complex.
• 53:37
  SPEAKER_00: So that's something to think about as well.
• 53:39
  SPEAKER_00: Like, if people have to go through like huge code things to just get started, then basically simplify the API and uh and bundle it up in something else, and then give people a source code repo to fork instead of just copy and paste tons of stuff from your documentation.
• 53:54
  SPEAKER_00: I remember when when brackets came out, the first text editor inside HTML, so you can actually start editing things in there, and that was basically uh then and now it's just a normal thing to take HTML, live HTML, project it onto Canvas, project it into WebGL, and do these fancy bits with it.
• 54:12
  SPEAKER_00: I mean, this is how people sold Keynote and PowerPoint in the past because these were the two, these were the effects they wanted to do.
• 54:20
  SPEAKER_00: Not they should do them because it's really not in, but they can.
• 54:25
  SPEAKER_01: Yes, it was when you see someone doing a presentation for the first time, they'll try and use like every single animation on in a different slide.
• 54:31
  SPEAKER_01: So they'll go through a whole list of animations, and it's just like you don't need to use everything.
• 54:35
  SPEAKER_01: Or when also when Bootstrap came about, I really like Bootstrap when it first came out, but everyone had to, including myself, we had to use like every possible component possible.
• 54:44
  SPEAKER_01: Um, it's like, no, you don't have to.
• 54:46
  SPEAKER_01: And I was like, Oh, yeah, you're right, you don't have to.
• 54:47
  SPEAKER_01: It's true.
• 54:48
  SPEAKER_02: It does go back to that taste thing though, doesn't it?
• 54:50
  SPEAKER_02: About how it's not actually just taste, it's like years and years of doing it and then getting feedback on it.
• 54:55
  SPEAKER_02: Because I I refer to it as that kind of like um uh confetti editing.
• 54:59
  SPEAKER_02: And it's funny when you see that confetti here on screen, it's that idea that you need to keep someone constantly engaged with you know these animations.
• 55:06
  SPEAKER_02: But when you are doing like a walkthrough for code, that can be a really great animation because it's kind of showing the like magical, you know, inclusion of a line that does all of the work.
• 55:14
  SPEAKER_02: And it um I know what you mean, Chris.
• 55:16
  SPEAKER_02: It's like you can't really kind of encourage taste because you're giving all these features, and then people will just use as many of them as they can.
• 55:22
  SPEAKER_00: And I think the big at the OS X thing when keynote came out with the fire animation to turn fire to have fire from ROM text to another, that was what everybody used, or the RGB one which I would turn it into three RGB panes and then turn the thing around, and you're like, if you have to wait for your animation to build and you're as a speaker stand there like twiddling your thumbs, that's probably not the right way of doing it.
• 55:45
  SPEAKER_01: If I go to PayPal uh dev, so if anyone, oh sorry, wrong one.
• 55:50
  SPEAKER_01: It's actually on the not developer on the actual PayPal account.
• 55:53
  SPEAKER_01: We do have the um uh AI toolkit if people do want to go check it out as well.
• 56:00
  SPEAKER_01: I'll just to AI should come up.
• 56:02
  SPEAKER_01: There we go.
• 56:03
  SPEAKER_01: So people can contribute to that project as well.
• 56:06
  SPEAKER_01: So it's uh yeah, it got in the clawed marketplace, which I know that was quite quite challenging to begin with to get in there, but it is uh quite interesting.
• 56:14
  SPEAKER_01: Um so yeah, but I I like what um you were saying, Chris, about uh I mean keeping APIs simple is important, but I guess some APIs have grown over the years and decades, and there's just a lot.
• 56:27
  SPEAKER_01: Um it contains a lot, so uh yeah, having skills and plugins do definitely really help, especially it comes to webhooks and um you know signatures and those sorts of things.
• 56:37
  SPEAKER_01: They are uh easy once they're working, but challenging sometimes during to debug and and and to build it in.
• 56:45
  SPEAKER_00: Cool.
• 56:46
  SPEAKER_00: So yeah, that's happening too.
• 56:48
  SPEAKER_00: So you can basically look that up and uh probably Eddie will post about it in the nearer future a lot, anyways.
• 56:54
  SPEAKER_00: So I think that's good to see.
• 56:56
  SPEAKER_00: But it's fun that you had a certain project running and you you you you did things.
• 56:59
  SPEAKER_00: You you quite forgot to name the project.
• 57:02
  SPEAKER_00: What was it called?
• 57:03
  SPEAKER_01: Oh, yes, good point.
• 57:04
  SPEAKER_01: You did code real.
• 57:07
  SPEAKER_00: Code real, okay.
• 57:09
  SPEAKER_01: AI came up with it.
• 57:10
  SPEAKER_01: I can't take credit.
• 57:11
  SPEAKER_01: We went through a few namings and uh yes, we we came up with that.
• 57:14
  SPEAKER_00: Uh AI came up with that, so it's much better than code fake, yeah.
• 57:19
  SPEAKER_00: Yeah, cool.
• 57:21
  SPEAKER_00: Customer difference nowadays.
• 57:24
  SPEAKER_00: Well, I guess that's uh then we're kind of at the end of the hour.
• 57:27
  SPEAKER_00: So uh that was We Are Developers Live with Eddie Joe.
• 57:30
  SPEAKER_00: It's it's always fun to have you around.
• 57:32
  SPEAKER_00: Um again, our big focus is right now what's scrolling on the bottom there in a marquee uh element, September 23 to 25th, uh 26 to 26, San Jose, California.
• 57:44
  SPEAKER_00: And we got rid of the AE right now.
• 57:46
  SPEAKER_00: I see that because we had this there at the beginning.
• 57:49
  SPEAKER_00: And yeah, we are developers.us, uh you get uh you can get your tickets.
• 57:54
  SPEAKER_00: Um Eddie, we're gonna give you a code so you can send it out to your followers as well to actually get more people uh in there, uh, so we can get you like a 50%, 40% off or something.
• 58:05
  SPEAKER_00: We'll see what we have.
• 58:06
  SPEAKER_00: And I think that's uh uh it's a good opportunity to meet us, good opportunity to meet people if you're in America and you see this.
• 58:12
  SPEAKER_00: Uh the of course it's not gonna be as cool as the Berlin event, but like I mean, the Silicon Valley finally gets something technical to talk about, so that's good to have as well.
• 58:22
  SPEAKER_00: So yeah.
• 58:23
  SPEAKER_00: Uh, but yeah, it's gonna be an interesting month for us ahead now for the next few weeks, and then we're gonna be yeah, getting all the travel tips that we need and packing tips that we need.
• 58:33
  SPEAKER_00: And hopefully, Fake One is gonna be on your uh on your schedule to learn more about this and find more, not get tricked into answering wrong things over and over again.
• 58:44
  SPEAKER_00: Thanks again, Eddie.
• 58:45
  SPEAKER_00: Any last words for you for people out there?
• 58:48
  SPEAKER_01: Uh no, just thank you all for watching and listening.
• 58:50
  SPEAKER_01: Thank you both.
• 58:51
  SPEAKER_01: Um, you've been awesome to have me on your show.
• 58:53
  SPEAKER_01: It's just great.
• 58:54
  SPEAKER_01: I love watching these, so to be part of it has been super, super exciting.
• 58:57
  SPEAKER_01: Um, everyone, just remember to keep building your side projects.
• 59:01
  SPEAKER_01: Make sure you've got a work-life balance, so do take some time to touch on grass.
• 59:04
  SPEAKER_01: And any questions, uh, you can find me on most social platforms that I'm very active on X.
• 59:09
  SPEAKER_01: And I look forward to seeing you both again in person because I love virtual stuff, but I also love being in person.
• 59:16
  SPEAKER_01: Um, it's always just great.
• 59:17
  SPEAKER_01: I think last time I saw Chris was uh Chris, I saw you in um in Croatia.
• 59:23
  SPEAKER_01: Croatia yes, yes, exactly.
• 59:26
  SPEAKER_01: So um, yeah, it's uh we get to meet people at uh random places all over the world.
• 59:30
  SPEAKER_01: It's always uh always good fun.
• 59:32
  SPEAKER_01: But thank you again both for having me.
• 59:33
  SPEAKER_01: Uh it was just amazing.
• 59:35
  SPEAKER_01: Cool.
• 59:35
  SPEAKER_01: Any last words from you, Dan, for people out there?
• 59:38
  SPEAKER_02: Yeah, I was I was hoping you're gonna say that.
• 59:40
  SPEAKER_02: First of all, I just want to say um Eddie, thanks for joining us, and we'll go back to what we said right before we kind of started recording.
• 59:45
  SPEAKER_02: It's been six years since we've been chatting, and I can now finally verify that you're not a catfish.
• 59:49
  SPEAKER_02: You are as advertised, or you're a very good deep fake, and we'll have to verify it in person.
• 59:54
  SPEAKER_02: So now that means you've got no excuse but to come to uh OER developers event.
• 59:57
  SPEAKER_01: So I'm definitely gonna try.
• 59:59
  SPEAKER_01: I would love to, that'd be awesome.
• 1:00:00
  SPEAKER_00: Well, then thanks for the thanks for watching.
• 1:00:02
  SPEAKER_00: We see you next Wednesday, and uh, there's lots of coffee with developers coming out because we've just recorded in the last few days.
• 1:00:08
  SPEAKER_00: Some really cool stuff coming your way, so have a peek at that as well.
• 1:00:11
  SPEAKER_00: And the tools that we showed today are now on the magazine, so you can read it up there as well, and you can click the links and support the uh the makers of those tools, maybe pay through PayPal or to any of the other payment systems that are not existing as a competition.
• 1:00:26
  SPEAKER_00: But, anyways, any of those that will work for you.
• 1:00:29
  SPEAKER_00: So I see you soon, and we see you next Wednesday.
• 1:00:31
  SPEAKER_00: Bye bye.