WEBVTT 00:00:00.160 --> 00:00:03.940 I think you should ask an LLM to come up with a testing strategy for this setup. 00:00:04.560 --> 00:00:07.380 Oh, that's that's so simple. Even I could do it. 00:00:07.840 --> 00:00:09.860 What's the mission of the testing here? 00:00:10.160 --> 00:00:12.580 So that we don't have to record the same thing twice. 00:00:13.915 --> 00:00:14.975 Well, that's a risk. 00:00:16.155 --> 00:00:19.675 Always a risk. Always a risk. Well, a 00:00:19.675 --> 00:00:22.495 risk that we've turned to reality before as well. 00:00:22.795 --> 00:00:30.570 We have. Yeah. In our early episodes, we we quite often well, more than once recorded things twice, 00:00:30.570 --> 00:00:33.390 didn't we? Absolutely. But they got better with the retelling, 00:00:34.490 --> 00:00:39.955 probably. Yeah. We had more rehearsed opinions in those days. Blimey. 00:00:40.575 --> 00:00:41.555 Blimey indeed. 00:00:42.735 --> 00:00:43.955 I wrote some things. 00:00:44.895 --> 00:00:45.775 Interval things. 00:00:45.775 --> 00:00:50.995 Interval things. But I just want to do the quickest one first because I find it hilarious. 00:00:51.135 --> 00:00:51.935 Okay. Go. 00:00:51.935 --> 00:00:54.820 So Hewlett Packard. It's just 00:00:54.880 --> 00:00:55.860 already hilarious. 00:00:56.560 --> 00:01:07.685 It's sad. A once great company. They actually changed their customer service system so that if you rang them about your printer, you would have to wait fifteen minutes minimum. 00:01:07.745 --> 00:01:08.225 Okay. 00:01:08.225 --> 00:01:12.705 So even if all the operators were free, you still had to wait fifteen minutes 00:01:12.865 --> 00:01:13.365 Okay. 00:01:13.505 --> 00:01:15.825 To put you off, ringing them. 00:01:15.825 --> 00:01:16.485 I see. 00:01:16.865 --> 00:01:32.800 And you just think, that's just terrible. And then this was an article in the register, and then right next to it was another article about HP changing their mind about doing this. Because when somebody said it out loud or wrote it out loud in an article 00:01:33.365 --> 00:01:34.085 Then it seemed 00:01:34.325 --> 00:01:35.285 It didn't seem 00:01:35.285 --> 00:01:36.805 didn't seem like a great idea. 00:01:36.805 --> 00:01:40.105 No. Seemed a bit, some might say, mean spirited. 00:01:40.645 --> 00:01:44.825 Yeah. Yeah. You do wonder about the the group think that got them there. 00:01:45.045 --> 00:01:46.085 Yes. Yes. You know, 00:01:46.085 --> 00:01:51.960 all the total inability to challenge someone's ridiculous opinion about how things should work. 00:01:52.020 --> 00:01:55.860 I don't know what you mean. Yeah. That. 00:01:55.860 --> 00:01:59.240 Yeah. It's probably that, isn't it? But, yeah, that that's pretty good. 00:01:59.300 --> 00:02:00.580 I did find that amusing. 00:02:00.580 --> 00:02:08.895 So My first job at university, I did a good service desk stuff, and I would have liked that fifteen minute, relaxation time. 00:02:10.235 --> 00:02:16.155 And I don't get the impression that they will have, done it to make the lives of their customer service representatives happy. 00:02:16.155 --> 00:02:17.135 No. That's true. 00:02:17.240 --> 00:02:35.835 I feel as though it probably wasn't for that. Yeah. Maybe something to do with costs. I'm just gonna put it out there. After listening to 15 of I think the messages were something like, oh, it would be much quicker if you just went on our website to find this support. 00:02:36.215 --> 00:02:37.415 Use our new chatbot. 00:02:37.415 --> 00:02:48.840 It will be much quicker if you went on our website. Do you not feel like you're wasting your life away on this on hold when you could be on our website finding the answer to your question? Or 00:02:49.320 --> 00:02:55.020 So maybe it's just like a reliable fifteen minutes of relaxation time on the phone for the customer. 00:02:55.240 --> 00:02:59.560 It's slightly opposite of mindfulness, isn't it? Yeah. Mindlessness. Yeah. 00:02:59.560 --> 00:03:08.675 Although, calling the HP helpline for a printing issue would probably be the very last thing that I would do. 00:03:08.975 --> 00:03:27.470 That is beyond beyond the last thing I would do. In fact, I forestall it by never having HP printers. Yeah. My my way of buying printers is actually quite interesting, well, to me anyway Yeah. In that I feel as though you have to look at what is the cost per sheet. 00:03:27.665 --> 00:03:58.995 And it turns out the cheapest way to buy a sheet printed sheet of paper is to get an enormous work group printer that comes with so if you look long enough, you'll find one that comes with 12,000 sheets or something in the in the existing toner. Yeah. And then you go and you use it for a bit, and then the really awful thing happens. It runs out of toner, and you discover it will cost more than you paid for the printer by a factor of double or three times to buy the toner to put in it. Right. 00:03:58.995 --> 00:04:02.455 And then you buy another printer. Yeah. All very green. 00:04:03.155 --> 00:04:11.670 No. No. I I don't believe that they were originally manufactured as disposable single use objects. 00:04:11.810 --> 00:04:21.750 But they've made them economically. That's the only way to go, really. Yeah. At least if you're a human, I guess if you're a mega corp, you can probably just pay for the toner. 00:04:22.475 --> 00:04:24.315 Yeah. Mega corp. 00:04:24.315 --> 00:04:27.615 But to me, £800 is a lot of money for toner. 00:04:27.915 --> 00:04:28.415 Definitely. 00:04:28.875 --> 00:04:38.610 Even for untold numbers of thousands of sheets. I'd rather just print with no red. Yeah. Well, okay. With no cyan, yellow, or magenta. 00:04:38.910 --> 00:04:40.930 I think you need the black. That's what you need. 00:04:40.990 --> 00:04:57.835 Yeah. Yeah. I specifically got a very basic printer, which you couldn't do anything special with. So it was only for the the only reason I really use it is to print out things to sign for the business. They're real things that you need to sign like accounts, not just things to sign. 00:04:57.835 --> 00:04:58.415 No. No. 00:05:00.395 --> 00:05:07.290 Yeah. Signing is good fun, of course, but but be ecologically friendly about it. 00:05:07.370 --> 00:05:12.830 Made up certificates and things like that. Yeah. There you go. That's for you. 00:05:12.890 --> 00:05:14.890 Actually, the, The what 00:05:14.890 --> 00:05:16.030 a lot of things award 00:05:16.090 --> 00:05:34.470 for you. Apple Pages has got some really good pretend certificates templates so that you can just knock you know all that curly q kind of design to to make it unforgeable Yeah. Where you can just print, like, 10,000 copies of it. But, yeah, it looks looks pretty good Alright. 00:05:34.470 --> 00:05:34.710 Okay. 00:05:34.790 --> 00:05:37.770 If you if you want a Yeah. Amusing certificate. 00:05:40.550 --> 00:05:43.690 So HP, the worst offenders then? 00:05:44.085 --> 00:06:18.415 The idea that you would deliberately force people to waste fifteen minutes of their fifteen minutes of their life just gratuitously to penalize them for trying to get their thing that you sold them to work, It's just I mean, it's not surprising that they you turned on it because when you say it out loud, it just it sounds terrible. I mean, what meeting what did that what what I I can't even what was the meeting like where they decided to do it? No. Are they all just well, yes. Do we set it as three hours? 00:06:18.635 --> 00:06:24.840 Oh, no. That's too long. Oh, what about four minutes? Oh, no. They they expect that. 00:06:24.840 --> 00:06:26.860 It has to be long enough to hurt. 00:06:27.880 --> 00:06:32.840 But not so long as to be utterly ridiculous. Although fifteen minutes is pretty 00:06:32.920 --> 00:06:37.365 Their their judgment of what is utterly ridiculous is Yeah. Is is in question, I would say. 00:06:37.365 --> 00:06:51.070 I imagine the meeting was, very senior person suggests the thing, the fifteen minute thing, and then everyone else goes very quiet. Yeah. And then it gets implemented. 00:06:51.210 --> 00:07:01.470 But the meeting would have been called something like strategies or moving support requests to the web Yeah. Or something like that. Mhmm. And there would have been one. Yeah. 00:07:01.530 --> 00:07:06.105 Punish them for doing the not make the web really good. 00:07:06.105 --> 00:07:06.745 Well, yeah. 00:07:06.745 --> 00:07:09.245 But but punish them for going on the phone. 00:07:09.865 --> 00:07:15.785 Yeah. And then still have your web support as a maze of ancient PDF files for 00:07:16.025 --> 00:07:17.325 All alike. Yeah. 00:07:19.100 --> 00:07:24.000 With a baffling array of of printed names, which all look incredibly similar. 00:07:24.700 --> 00:07:31.700 Baffling array of printer names. Yeah. I think you're right. I think that's probably what what the strategy was. 00:07:31.820 --> 00:07:32.640 Been there before. 00:07:32.700 --> 00:07:37.465 Yeah. Yeah. So we're doing something different, aren't we? 00:07:37.925 --> 00:07:38.425 Today? 00:07:38.485 --> 00:07:41.845 We're surrounded by gadgets. Surrounded. Additional gadgets. 00:07:41.845 --> 00:07:50.880 In in yeah. For Ian's office, there is often gadgets, but there is a a wider array now. Another baffling array. 00:07:51.500 --> 00:08:10.745 You see, when Ash let the side down by moving to Android Oh, yeah. And thus removing himself forever from well, not forever, possibly. He could change back. Well, yeah. But, removing himself from the pool of people with available iPhones to do recording using iPad multi cam sorry. 00:08:10.965 --> 00:08:18.265 Final Cut Pro multi cam with multiple iPhones. I have stepped into the breach with a spare iPhone. 00:08:18.800 --> 00:08:25.600 Yeah. And to be fair, I only had an iPhone seven. So it probably I don't know if if this even would have worked. 00:08:25.600 --> 00:08:55.100 Or I'm just going to say put my thing well, I can't get my fingers in my ears because there's literally enormous headphones anyway. And those of you that actually see the fruits of this, exercise might actually be able to appreciate that visual joke. Because for about half of the last episode, I was pointing at this and saying, oh, well, that's something something and then pointing at something else. Mhmm. And, I felt as though it was kind of amusing, but also a bit difficult to follow for listeners. 00:08:55.240 --> 00:09:04.865 What what will we do without the the that's not a great thing to do on a podcast type jokes about pointing at things? 00:09:05.245 --> 00:09:31.005 Well, there's an extremely high probability that this experiment will prove that we are going to remain an audio only podcast. But for just this episode, we are recording ourselves using Final Cut Pro multicam. And Ash looks quite sensible, and he's got nice whiteboard and post it notes in his background. Yeah. And I've got a door and a chair. 00:09:31.005 --> 00:09:39.950 So it's it's I like to think I control the environment of this podcast with a rod of iron, you know, with my special buttons. But 00:09:44.170 --> 00:09:46.670 you haven't controlled the physical environment. 00:09:46.890 --> 00:09:51.850 But I haven't successfully controlled the physical environment. Yeah. I will. 00:09:51.850 --> 00:09:52.590 Oh, well. 00:09:52.935 --> 00:10:00.375 You've I'll just keep what I'll do is the video can entirely be Ash, and the audio can entirely be me. 00:10:00.375 --> 00:10:05.435 Oh, yeah. I'd like that. The the the chaotic part of me likes that. 00:10:05.655 --> 00:10:10.590 Yes. There's something there's something to that, isn't there? Yeah. However 00:10:11.530 --> 00:10:15.150 However, we're here. What are we here for? Things. 00:10:15.450 --> 00:10:22.590 Things. When do we want them? Now. Soon. But do we want things? 00:10:23.025 --> 00:10:26.625 Yes. Oh, dear. Yes. I suppose we are here for things. 00:10:26.625 --> 00:10:33.925 Yep. It's it's me to go first. Because last time, you told me that testing was dead, and that was first. 00:10:33.985 --> 00:10:40.590 It was it was first. Yeah. I think we had a little bump in listenership through our evil use of clickbait for that. 00:10:40.970 --> 00:10:42.910 Your evil use of clickbait. 00:10:45.610 --> 00:10:52.965 Ash would like to disassociate himself from any suggestion that testing has ever been may or not or may be dead. 00:10:53.425 --> 00:10:54.405 That's that's correct. 00:10:55.025 --> 00:10:57.025 You see, I did say may not in there. 00:10:57.025 --> 00:10:58.485 Mhmm. Well 00:10:58.865 --> 00:11:05.630 So you have just disass by saying that is correct, you've disassociated yourself also from the notion that testing is alive. 00:11:07.530 --> 00:11:09.470 Testing is Schrodinger like. 00:11:09.770 --> 00:11:11.050 Yes. Depending 00:11:11.050 --> 00:11:12.090 on who is looking. 00:11:12.090 --> 00:11:19.005 It's in a quantum intermediate state. Yeah. Its qubits are being rotated through a quantum gate. 00:11:19.865 --> 00:11:21.085 I still don't understand. 00:11:22.185 --> 00:11:26.825 Well, in the end, I think we all collapse or something like that. 00:11:26.825 --> 00:11:28.205 Or something like that. Yeah. 00:11:30.380 --> 00:11:32.160 Alright. I have to concede. 00:11:32.860 --> 00:11:34.780 Concede to the beginning of things. 00:11:34.780 --> 00:11:42.480 To going first Okay. And beginning of things. So, Ash, what is your thing? 00:11:42.540 --> 00:11:43.500 What is my thing? 00:11:43.500 --> 00:11:46.465 What? Oh, fuck it. I've done 00:11:46.465 --> 00:11:48.065 that for a while, so I think you you Yeah. 00:11:48.065 --> 00:11:49.685 Well, is it allowed occasionally? 00:11:49.905 --> 00:11:50.805 Yeah. Of course. 00:11:50.945 --> 00:11:51.585 Brilliant. No. 00:11:51.585 --> 00:11:59.105 It's just when it becomes a a solid fixture of the podcast, and then we get a review of the podcast done. And 00:12:00.110 --> 00:12:01.810 Somebody notices. Somebody says, 00:12:01.870 --> 00:12:03.570 why do you keep doing the same jokes? 00:12:03.870 --> 00:12:05.790 We're never gonna have another review done, are we? No. 00:12:05.790 --> 00:12:06.270 No. That's true. 00:12:06.270 --> 00:12:09.790 So we just we've done that. That's that chapter is That 00:12:09.790 --> 00:12:12.590 chapter is gone. Well, it's not gone. It's still with us. 00:12:12.590 --> 00:12:21.145 It is still with us. Yep. As we address our one listener and other things. And we've got a better a better cover art as well, haven't we? 00:12:21.145 --> 00:12:24.685 Yeah. No. That's true. So what is my thing? 00:12:24.905 --> 00:12:26.605 What is your thing? 00:12:31.000 --> 00:12:39.735 So my thing is called user story splitting or just the chopping up of work in general on the development team. 00:12:39.815 --> 00:12:40.315 Dismemberment. 00:12:40.695 --> 00:13:03.780 Dismemberment of work in general on the development team. So the thing is so if there's if there is ever a skill that teams need but rarely have in the either to the refined state or the, or at all that is required in order to do this well, it's splitting down work into deliverable and valuable tasks. 00:13:04.195 --> 00:13:07.895 So effectively, when you're given something gigantic. 00:13:08.275 --> 00:13:17.250 Yeah. Yeah. Which, you know, even in our agile world, we still get given big post, I was gonna say 00:13:17.490 --> 00:13:21.750 Big post it now. Big briefs. Big briefs. It's big, big bloopers. 00:13:22.530 --> 00:13:30.150 We still get given big chunks of work to do, don't we? Because Yes. It's often, articulated in that way. 00:13:30.665 --> 00:14:06.505 So is there a difference between taking a piece of work and breaking it down in the sense of deciding how to approach it and a a sort of is there a more grand thing of breaking it into two similar pieces of work that then you know, for example, you may say, well, actually, I'm gonna go back to the business who've given me this enormous thing. Yeah. Enormous piece of work because thing is a reserved word in this podcast. If they've given you an enormous piece of work, you might say, well, can we think of it as these two pieces of work? Yeah. 00:14:06.505 --> 00:14:17.980 Yeah. But equally, you might have a piece of work that's maybe not too enormous. So you might say, now I want to break it down to find out what the tasks are. Yeah. Ah, that's that's what I mean, isn't it? 00:14:17.980 --> 00:14:24.515 Yeah. It's about splitting it down into tasks or splitting it into two, I'm trying not to say project. 00:14:24.655 --> 00:14:26.895 Yeah. Yeah. Yeah. I know what you mean. Yeah. 00:14:26.895 --> 00:14:52.345 Yeah. Well, there's there's kind of, like, a few things, isn't there? So it's like you've got, like, the behavior when the piece of work is is first sort of introduced to the team and the size of it and thinking about, like, how it might be split or, you know, having the wherewithal to say, actually, that's really big. So can we think about this differently and split it down a little? And then you've got, like, once you've this is probably like there's, like, a couple of orders of this. 00:14:52.405 --> 00:15:26.225 There's that part when you first, you know, introduced to whatever the big piece of work is. How do you react there? And then there's also, once you've taken it into the team and you understand it a bit better, how do you then split it up in order to begin to work on it? Because I think often I've seen, like, the first reaction is just to start working on it, which has some value, but the splitting behavior doesn't seem to come naturally. It's not like the first thought, you know, to say, well, how do we break this down, once you've got, like, some understanding of it? 00:15:26.285 --> 00:15:47.825 So if I describe this in, like, how it's it's manifested itself in my career. So, basically, a big requirement comes into the team, and then everyone is just like, oh oh, dear. This is gonna take, like, like, years or whatever it is. So, but no one's willing to say it. So they just say, okay. 00:15:47.885 --> 00:15:51.505 And then just just go off and and be nervous for a while. 00:15:51.885 --> 00:15:58.740 So it's size, isn't it? It is size. Yeah. The the there's no the motivation for splitting it is the size of it. 00:15:58.900 --> 00:16:08.920 Yeah. Right. And also sometimes to understand it. But sometimes Yeah. With smaller sizes, you get more understanding as well because you kind of you get to see how things sort of fit together. 00:16:08.980 --> 00:16:19.025 Yeah. Usually, what happens is people start to raise some, like, initial tasks, and they tend to be quite technical in nature. It's like, oh, well, let's have a database, and let's have a this, and let's have a that. 00:16:19.405 --> 00:16:22.945 Let's do the things we know we're gonna need to do, whatever happens. 00:16:23.565 --> 00:16:41.665 Yeah. Yeah. But I I'm always like, it's probably a bit too early for that as well. And then everyone, like, everywhere, kind of all at once tries to work on it and tries to do something with this giant thing and, like, nibble around the edges of it. And then I say, can we stop and think about how to break this down? 00:16:41.805 --> 00:16:48.365 And then no one stops because they're already, like, scared by the size of the thing that they're trying to work on. 00:16:48.365 --> 00:16:49.485 With a small t. 00:16:49.485 --> 00:17:11.675 Yeah. With a small t. And then once I get to a certain point when I can't resist anymore, after leaving the team to to to see what they do after, you know, a little bit of time, I would say, right. Okay. Let's get together and and think about how we can split this work and what where, like, the seams are of this work. 00:17:12.135 --> 00:17:17.355 But that has been repeated, like, through my career, like, ad infinitum. 00:17:18.620 --> 00:17:19.280 Ad nausea. 00:17:19.340 --> 00:17:43.055 Ad nausea. So to me, it it seems like it comes quite naturally. I quite enjoy picking up a piece of a larger piece of work and thinking about how it could be it could be broken down. But it doesn't seem that way on, you know, for for other people on that I work with on development teams. So I'm just wondering, like, is it I think it's a really hard thing to do, isn't it? 00:17:43.220 --> 00:17:45.700 I think it is. I mean, I had a hot take on this 00:17:45.940 --> 00:17:46.340 Okay. 00:17:46.980 --> 00:18:10.035 Which which was, is it because it's hard to split it if you don't have both the knowledge from the that the business has about the context Yeah. And the knowledge from the technology side about what's easy and hard in that environment. And it's hard to do because it's hard to find people within both of those domains. Yeah. Because I'm kind of going somewhere with that. 00:18:10.540 --> 00:18:21.200 I when I thought about it some more, I was kind of thinking about actually, I kind of get why it's you. Yeah. Because it's testers, other people with their feet in both domains. 00:18:21.260 --> 00:18:22.080 Well, yes. 00:18:22.140 --> 00:18:43.150 And actually, serious question, this is to make up for saying testing implying testing might be dead last time. Maybe this is something that actually testers are well suited to lead on doing Yeah. Because it is it's actually they're the ones perhaps with the most feet in both domains. Yeah. Most feet. 00:18:44.330 --> 00:18:44.830 Okay. 00:18:45.290 --> 00:18:46.170 They're the ones with 00:18:46.170 --> 00:18:47.850 a clear foot in each domain. 00:18:47.850 --> 00:18:59.675 Yeah. Yeah. I go through, like, I go through stages with this, though, because when when when these scenarios, like, come about, I'm like, I don't I don't get paid enough to do that kind of work. 00:19:00.215 --> 00:19:01.595 Yeah. But what does that mean? 00:19:02.135 --> 00:19:21.540 I don't know. Maybe it's just a slight rebellion on my part, you know, to say, well, I have this skill, and it's probably something to do with my background as a tester. But I'm like, well, until I can't resist it anymore, I'd still resist it on those grounds that it's like, well, this is kind of higher order work. 00:19:21.875 --> 00:19:42.570 You know? But is it something I mean, that's why well, kind of my question. It doesn't it make sense actually for testers to play play a leading role in that? And in fact, wouldn't it be a good idea if you made it part of the process? So for everything that came in, you said, is it is this too big, too small, or just right? 00:19:42.570 --> 00:19:42.970 Well, you 00:19:42.970 --> 00:19:46.570 see Is it a Goldilocks task? See, the thing is that asking. 00:19:46.650 --> 00:19:53.095 Is there's some, like, I I agree with you, and then reality kicks in. And no one 00:19:53.335 --> 00:19:55.355 I liked it when you just agreed with me. 00:19:56.215 --> 00:20:11.470 So so yes. Ideally. But in my again, I'm kinda going back to my experience. I then need to say it nine or 10 times until the message, like, sinks in and the splitting actually happens. 00:20:11.470 --> 00:20:26.365 But that's part that's the making it part of the process half of it, isn't it? Yeah. You're looking you're asking for every task. Is Is it I'm gonna say it again because I didn't think it had gotten off of a a laugh last time. Is it a Goldilocks task that's just right? 00:20:28.025 --> 00:20:29.865 Is it too big or too small? 00:20:29.865 --> 00:20:30.905 Yeah. Yeah. 00:20:30.905 --> 00:20:32.160 Or is it just right? 00:20:32.560 --> 00:20:33.840 What? The size of the task? 00:20:33.840 --> 00:20:34.880 The size of the task. Yeah. 00:20:34.880 --> 00:20:37.040 Well, you want Just right, don't you? 00:20:37.040 --> 00:20:48.785 Is that belaboured enough, do you think? Yeah. Just just right. But if the team asked that about every task, then instead of flapping about trying to do enormous tasks Yeah. Yeah. 00:20:48.865 --> 00:21:03.850 And it was in the process to, not do it. And it was led by someone, may well be a tester Mhmm. Who's got a foot in both camps and can talk about how we play this back. I mean, tasks coming from the product manager. Right? 00:21:03.850 --> 00:21:04.670 Yeah. Yeah. 00:21:04.810 --> 00:21:17.725 And so they can play back to the project map product. Blimey. They can play back to the product manager and say, hello, mister or miss product manager. Yep. This task is really big. 00:21:17.865 --> 00:21:30.205 What if we did it like that? Yeah. As these two tasks, then immediately you would be able to report progress because one of them will be finished quicker. I don't know. You'd have to find a way to to sell it to them. 00:21:30.570 --> 00:21:49.115 One of the interesting things in the, in the articles that you linked about this, which we will obviously put in the show notes for everybody to read. I didn't say that last time. I was gonna say the second article because I can't actually remember. One of them was by Mike Cohn. Yeah. 00:21:49.495 --> 00:21:54.535 And he talked about splitting it by considering the path through it 00:21:54.855 --> 00:21:55.335 Yeah. 00:21:55.415 --> 00:22:04.520 Which I thought was really interesting. So and then and and often you he he was sort of saying it's a bit of an anti pattern, always just automatically trying to split it on technology. 00:22:05.220 --> 00:22:06.040 Yeah. Yeah. 00:22:06.340 --> 00:22:22.565 I thought that was interesting. There's a bit of a bit of a digression from the main point I was trying to make, which is but it's it's how you split it and and how you sell it and how you can propose things back to the product manager or the business that make sense. 00:22:23.025 --> 00:22:51.885 Yeah. So my my brain goes through it applies a bunch of patterns to something, like, pretty much as soon as I see it. Like, the the the big requirement might how could we split this? So would it be by so you could do it by path through the system. You could do it by, sort of saying, like, what's the simplest thing we can build and then add the more complex items as we go along or whatever the pattern is. 00:22:51.885 --> 00:23:00.120 And I kind of go through a bit of a process in my head in order to start to come up with options for for splitting work down. 00:23:00.820 --> 00:23:03.700 And, presumably, the path through it is one of those that you 00:23:03.700 --> 00:23:16.875 can see. Yeah. Yeah. And I often arrive at, like, a list of, like, capabilities that whatever it is needs to have. And then I I play with that list, put it in different orders, and group things together. 00:23:17.175 --> 00:23:23.035 But this is what's going on in my head and, well, in a markdown file in front of me as well. 00:23:23.300 --> 00:23:27.300 So if you were gonna guide a team through it, it might involve post it notes and moving them around. 00:23:27.300 --> 00:23:28.520 Yeah. Yeah. Yeah. Absolutely. 00:23:28.580 --> 00:23:35.000 You're doing some mental version of that. I mean, mental in in in terms of inside your mind rather than crazy. 00:23:35.140 --> 00:23:54.315 Yeah. Yeah. Because it's the kind of puzzle it's the kind of puzzle that I like. But I think and I think I I have a bit of a a skill for it, and I recognize when it needs to happen as well. Whereas sometimes, it takes a little while for teams to recognize when that needs to happen. 00:23:54.315 --> 00:23:59.790 Yeah. So I think there's there's something there as well. So maybe I could make it a bit more, like, explicit. 00:24:00.570 --> 00:24:03.530 What you get for being old? I mean, experienced, don't you? 00:24:03.530 --> 00:24:05.450 Yeah. No. I think I think it is as well. 00:24:05.450 --> 00:24:10.615 I can't call you old without calling myself ancient because I'm probably, not ten years older than you or something. 00:24:10.615 --> 00:24:22.280 Yeah. Absolutely. Oh, lord. Yeah. And the other thing that I often see is, someone always says, well, should we have an analysis task, a development task, and a testing task for this thing? 00:24:22.580 --> 00:24:39.645 And I'm like, no. Please, no. But I think my Mike Cohn mentions that in in the story splitting mistakes blog as well, where it's like, don't split it along expertise lines. Don't have oh, no. It's it's more like, don't have a spike for every task. 00:24:39.705 --> 00:24:43.065 Because that just means you you've got an analysis task to find out how 00:24:43.065 --> 00:24:44.105 to do it. Mhmm. 00:24:44.105 --> 00:24:45.705 Although, I've met a couple of people Sorry. 00:24:45.705 --> 00:24:47.305 Did he say do have a spike for 00:24:47.305 --> 00:24:48.970 every task? Don't have a spike for every task. 00:24:48.970 --> 00:24:51.610 I read that backwards, but I didn't read that bit very much. 00:24:51.610 --> 00:25:05.745 Oh, right. Okay. See, it's like yeah. Don't have a spike for every task because otherwise, that's just an analysis task, isn't it? The story is not independent of the analysis, which I do find is kind of another habit as well Oh. 00:25:05.825 --> 00:25:21.880 Is to wait for the analysis to be done as part of something separate and then start the story. It's like supposed to discover this stuff together. Yeah. It's kind of the togetherness, which is the point. So I always find that that sort of trend really interesting as well. 00:25:21.880 --> 00:25:30.940 Someone very proudly said to me a little while ago that they always have the lead developer does all the analysis for all the tasks, and then we just go away and implement them. 00:25:31.295 --> 00:25:34.735 That sounds like a great way to Is that why? Not not think about it very much. 00:25:34.735 --> 00:25:36.255 Yeah. Absolutely. Absolutely. 00:25:36.255 --> 00:25:39.955 This is like, oh, well, we've got all these people's brains, but let's not use any of them. 00:25:40.975 --> 00:25:42.515 Let's just have the lead developer. 00:25:42.815 --> 00:25:46.860 Maybe they can maybe they can just watch as I perform my brilliance. 00:25:47.000 --> 00:25:51.900 Yeah. And I'll try and keep this brief, but can we just return to Figma? 00:25:52.040 --> 00:25:53.500 Are we gonna need the music? 00:25:54.120 --> 00:25:55.660 Is there music for Figma? 00:25:55.720 --> 00:25:58.300 Well, the music is for Ash Rants, I've discovered. 00:25:58.600 --> 00:26:09.875 Alright. Go on then. Is this some relaxing music in order to temper the rant a little bit? Go on. Shoot. 00:26:10.655 --> 00:26:56.885 So once a set of Figma designs designs have been presented to the team, the team then has their brains cannot split them down because they just see the entirety of the designs. Once again, I'm not blaming Figma the tool. I'm blaming Figma as it's used. So, in terms of breaking things down, Figma and how it has been deployed has destroyed stories for tea as a viable technique. Right. 00:26:56.885 --> 00:26:57.045 I'm 00:26:57.045 --> 00:27:06.185 done. I'm fading out for you. Well, I hope you all enjoyed that. Mhmm. I did. 00:27:06.245 --> 00:27:07.225 Yeah. Of course. 00:27:07.550 --> 00:27:14.510 I also think having front end, back end, and database teams kind of dictates how you're gonna split the tasks down as well. 00:27:14.510 --> 00:27:16.690 Well, Conway's law does come into this. 00:27:16.830 --> 00:27:22.095 Yeah. Definitely. It's very hard to imagine everybody working together on something when you've got such 00:27:22.215 --> 00:27:24.415 It's not natural. Goes against the laws of nature. 00:27:24.415 --> 00:27:31.715 Well, yeah, exactly. When you've got such heavily delineated teams Yes. That have a specific set of skills. 00:27:32.895 --> 00:27:36.995 Particularly when one team is asking the other team when the real developers are gonna come along. 00:27:37.055 --> 00:27:45.770 Okay. Yeah. Exactly. So I I do find that that is a is a thing as well, but with a small tip, we've got stopped in. 00:27:49.475 --> 00:27:54.355 Just gonna gradually entice you into this, weird It's a strange world. Strange world. Strange world. 00:27:54.355 --> 00:27:56.675 Capital two things, small two things. 00:27:56.675 --> 00:28:00.995 Yes. Fortunately, it's bounded by this this podcast. You can say 00:28:01.235 --> 00:28:02.275 Well, yeah. That's true. 00:28:02.275 --> 00:28:09.050 I mean, you say it in supermarkets and stuff, don't you? Yeah. And it it's even though you say, what, a lot of things 00:28:09.210 --> 00:28:09.890 It's just natural. 00:28:10.170 --> 00:28:24.415 Just it's just it's just natural. Yes. One of our favorite things was mentioned in one of your articles that you, you brought up. You You know what I'm talking about? Rock button. 00:28:25.515 --> 00:28:28.095 Now it all set up, and then I have to change it around. 00:28:30.315 --> 00:28:32.015 Tell me what you're talking about. 00:28:32.075 --> 00:28:35.300 What I'm talking about is the next one. 00:28:38.880 --> 00:28:40.180 That wasn't so smooth. 00:28:40.480 --> 00:29:07.040 It would have gone a lot smoother if I hadn't switched the buttons over. Maybe we need a button cam as well. We need a third camera to look down at the buttons so that, the the full joy of them can be appreciated. Yeah. He it was there was a guideline in there about if it's, if it's chaos, then stop trying to to do do this task splitting or whatever you're doing and just put the damn fire out. 00:29:09.100 --> 00:29:14.895 And then it was something like, when it's complex, don't try and list all the tasks, but just find the relevant ones. 00:29:15.055 --> 00:29:15.455 Okay. 00:29:15.455 --> 00:29:23.055 And do something. And then when it's complicated, it was, find all the tasks and then do the most important one. 00:29:23.055 --> 00:29:26.015 Okay. Well, that's that's pretty solid advice. That isn't it, I think. 00:29:26.015 --> 00:29:28.035 Yes. I thought it was quite a good use of 00:29:32.060 --> 00:29:36.060 conefing. Oh, we'll have to do conefing as a thing again soon, won't we? 00:29:36.060 --> 00:29:42.700 Yeah. If only for the, to justify the enormous cost of the Roadcaster Pro two. Yeah. 00:29:42.700 --> 00:29:58.710 Exactly. Good afternoon. For the tenth time, I want a lot of things, everyone. Good afternoon. I think that's an excellent time for me to say that was my thing. 00:30:01.570 --> 00:30:07.990 I enjoyed that thing a lot. I think I even said all the things I wrote down to say about it. 00:30:09.730 --> 00:30:11.910 So it's just rude in that. 00:30:12.435 --> 00:30:18.755 I know. I know. Well, I don't know what possessed me. I did I I decided to write something out, look. But there's 00:30:18.915 --> 00:30:19.075 I'm 00:30:19.075 --> 00:30:35.630 just pointing to the camera. You can see that that's me writing the thing down. Actually, I was trying to hold the thing up so the little image of me on my iPad could read it rather than the camera from which the little image was going, oh, it's not quite working. Oh, dear lord. Yeah. 00:30:35.630 --> 00:30:37.330 Let's send that one on, can Evan. 00:30:39.550 --> 00:30:40.430 Yes. Which, you 00:30:40.430 --> 00:30:41.085 know Let's. 00:30:41.325 --> 00:30:41.965 Is the end of 00:30:41.965 --> 00:30:43.485 all conversations as we know. 00:30:43.485 --> 00:30:49.105 I'm resisting the temptation Yep. To press that button again Yep. Because I'd only get it wrong. 00:30:53.780 --> 00:30:55.560 So let's have a let's have an interlude. 00:30:55.620 --> 00:30:58.360 We could argue that clanger is kind of saying interlude. 00:30:58.580 --> 00:31:02.120 Yeah. That's true, actually. Very chirpy interlude. 00:31:03.700 --> 00:31:12.825 On the button, it says whatever. Well. But, you know, interlude. I think that's better, isn't it? 00:31:13.125 --> 00:31:14.585 Have you got any interlude things? 00:31:14.725 --> 00:31:19.285 Oh, there were some interlude things. Yeah. And I had them open on my laptop, but now I've forgotten them all again. 00:31:19.285 --> 00:31:21.065 Firstly, we're on blue sky. 00:31:21.600 --> 00:31:24.420 We are on Blue Sky. Big news. 00:31:24.480 --> 00:31:30.640 Big news. And Ash has joined Blue Sky? No. I haven't, actually. Still haven't. 00:31:30.640 --> 00:31:31.300 I can't. 00:31:31.440 --> 00:31:32.820 You've busted on first. 00:31:33.920 --> 00:31:42.105 Just a bit over social media, to be honest, apart from, well, now LinkedIn. I don't know what LinkedIn is, but it's something else. 00:31:42.245 --> 00:31:52.940 It's hard to argue that one should not be over social media. Yeah. And I find blue blue sky a bit ambiguous because it's still owned by a billionaire, so it can still be subject to intitification. 00:31:53.880 --> 00:31:57.000 Subject to right wing intitification at any time. 00:31:57.000 --> 00:32:08.145 But a mastodon is therefore wholly better for being decentralized Yeah. And much more niche user controlled. Mhmm. But we are on Blue Sky because Transistor provides a Blue Sky integration. 00:32:08.285 --> 00:32:08.925 Alright. Cool. 00:32:08.925 --> 00:32:13.085 So when our most recent episode came out, it was posted on Blue Sky. 00:32:13.085 --> 00:32:13.585 Lovely. 00:32:13.805 --> 00:32:20.700 And as I'm the only person that follows our blue sky account, I had to retweet it. Re blue sky it. 00:32:20.700 --> 00:32:21.900 Re blue sky it. 00:32:21.900 --> 00:32:22.940 Repost it. 00:32:22.940 --> 00:32:23.840 Alright. Okay. 00:32:23.900 --> 00:32:39.695 Some some term that doesn't involve tweet. Yep. So I did. So we have now auto posting on blue sky. But one thing that's nice about the way they've done it is if you reply to the post on Blue Sky, it appears on the episodes page on our website as a reply. 00:32:39.995 --> 00:32:54.220 So you can actually talk about the episode underneath the actual episode on its very own page on whatalotofthings.com. Nice. That's our website, whatalotofthings.com. So, yes, we're on Blue Sky. 00:32:54.360 --> 00:32:56.060 But when will we be billionaires? 00:32:57.000 --> 00:33:00.345 Well, we've just gotta get our circulation up a little bit. 00:33:00.345 --> 00:33:02.525 Okay. We can do that. 00:33:02.585 --> 00:33:03.405 More listeners. 00:33:03.465 --> 00:33:04.125 More listeners. 00:33:04.185 --> 00:33:22.230 What's it even more? We have to come up with we'll ask an AI to do it just to make it worse, but we'll say, here are the titles of this week's things. Please come up with the most clickbaity episode title that you can. And then then it will say, they talked about these two things. You won't believe number two or something like that. 00:33:22.230 --> 00:33:24.630 Oh, yeah. Yeah. That's quite a common pattern, isn't it? 00:33:24.630 --> 00:33:31.825 It is. Although because we've only got two things. It's not very it's a bit less compelling than Well than it is when you've got a list of 10 things. 00:33:32.445 --> 00:33:35.985 Yeah. So what what else have we got on the list? Claude Code. 00:33:36.765 --> 00:33:47.780 Oh, Claude Code. Yes. I we we did this episode. And my thing in the last episode, as I know you all very well remember You, 00:33:47.780 --> 00:33:48.360 the listener. 00:33:48.420 --> 00:34:08.935 You, the listener, will very well remember. Will. And, I talked about how I was quite pleased with slurping up a repository and piping it into a command and finding out finding out things about it and using it to write test strategies to save me having to talk to any actual testers. They don't wanna do that. They'll just they'll just tell 00:34:08.935 --> 00:34:10.215 you the strategy's rubbish. 00:34:10.215 --> 00:34:11.515 They go on and on. 00:34:13.095 --> 00:34:13.650 You need to 00:34:15.250 --> 00:34:16.250 split your stories down. Split your stories down. 00:34:16.290 --> 00:34:23.250 What's the mission of the testing? No. I need to decide what level of testing you're gonna do. Check. What's the risk? 00:34:23.250 --> 00:34:24.550 Where's the value? 00:34:24.850 --> 00:34:32.835 I need the music again. On and on. Are we to do the music again? No. You could just repeat that with the music. 00:34:32.835 --> 00:34:45.730 That would be better. Yeah. So I was quite happy with that. And then, basically, yesterday, Our time machine. The day before. 00:34:46.910 --> 00:34:47.390 No. 00:34:47.390 --> 00:34:48.610 The day before yesterday. 00:34:48.670 --> 00:34:56.675 The day after The day. Or the time. The day after the episode. Oh my god. The day before Recently. 00:34:56.855 --> 00:35:22.670 The day you're not helping. The the day before the episode came out, Claude Code was released by by Anthropic along with Claude three point seven, which is their newest model. And everybody on various sites like Reddit and so on are all raving about how Claude three point seven is the best coding LLM. It's amazing. 00:35:23.045 --> 00:35:27.305 So you're gonna revolutionize how we use LLMs. 00:35:27.605 --> 00:35:28.645 Testing will be dead. 00:35:28.645 --> 00:35:33.765 Yeah. I enjoyed so I watched the video, which was the release, and it was too very fresh faced. 00:35:33.765 --> 00:35:34.565 It was It was. Yes. 00:35:34.645 --> 00:35:54.335 A developer and a product manager. Yes. And I enjoyed that Basically, they did a load of of stuff, like adding buttons to things and, which was, you know, apparently, the most basic thing that I've ever seen. And then at the end, they were like, can you, like, add some tests or something? And Claude was like, yeah. 00:35:54.335 --> 00:36:03.635 Sure. And then it just iterated over tests until they all passed. And then the two very fresh faced people were like, brilliant. Everything's amazing. I was like, yep. 00:36:03.855 --> 00:36:09.340 It did fail to ask about the risk associated with the testing. But the 00:36:10.280 --> 00:36:23.375 So Yeah. It, you know, it it it continued its unblemished record for, paying very little attention to what the LLO is actually testing. So, you know. Nonetheless Nonetheless. 00:36:23.595 --> 00:36:26.175 They at least added tests, and they thought about them. 00:36:26.315 --> 00:36:29.195 They did. They thought about Brief at the highest level possible Brief 00:36:29.435 --> 00:36:29.675 for a 00:36:29.675 --> 00:36:30.735 very brief moment. 00:36:30.875 --> 00:36:48.350 They thought some people watching this will be interested in adding tests. We better do we better do that. So I thought, well, I need to have a go at Claude Cove. Okay. So I wanted I've been wanting so maybe I should put the music off for me for this. 00:36:48.490 --> 00:37:14.430 We've been working on the Ilkki Live website Yeah. And we've it's converted to TypeScript now. And we've had to take some decisions about for so the biggest the biggest challenges about doing that was that it uses Notion as a database to store the performances. I think I talked about this on the episode for which our scars are still healing, where we talked about the the the 00:37:14.750 --> 00:37:17.965 I was just I'm I'm, like, winding up to begin again. 00:37:17.965 --> 00:37:19.165 I know you are. You said take a 00:37:19.405 --> 00:37:25.085 this was hard to do, and I was like, it's because there weren't any tests. It's because there weren't any tests. 00:37:25.085 --> 00:37:55.685 It's not because there weren't any tests. Although it didn't help. What was what was hard to do about it was that, basically, Notion has very, very complicated type system, which is probably alright reflected in TypeScript, but it's not it's a bit of a mystery. And so one of the things was, do I make this whole Notion thing type safe, or do I just say to myself, I'm replacing it with a database? I'm good. 00:37:55.685 --> 00:38:06.040 Let's just make the its interface type safe. Yep. And so, you know, the stuff comes back from Notion. We put it into an object, and then we zod the object. 00:38:06.180 --> 00:38:06.820 Right. Okay. 00:38:06.820 --> 00:38:19.155 Yep. We parse the object with zod. Just zod ing the object just seems like something you shouldn't do. That's how we make it make sure it's safe and it's got a proper typing. So we just do it through through that. 00:38:19.455 --> 00:38:40.860 And so one thing I've been wanting to do is say, okay. We'll need to replace Notion with a database. Mhmm. And I need to have an admin function in the app in the Ilkki Live website so that administrators can, in fact, manage Yeah. The things in the way Notion used to let them manage them. 00:38:41.115 --> 00:38:50.175 And so I went to Claude Code, and I thought, well, this is gonna be a pain to do. So let's see how Claude Code does that it. 00:38:50.475 --> 00:38:54.655 I actually think that's a really good use of an NLM for coding. 00:38:55.310 --> 00:38:56.530 It was great, actually. 00:38:57.070 --> 00:38:58.990 The actual It's 00:38:58.990 --> 00:39:00.130 within its competencies. 00:39:00.430 --> 00:39:10.695 Yes. Exactly. It's a really good example where it's like, you know, building a, I don't know, some kind of complex interface to communicate complex information. 00:39:12.915 --> 00:39:14.435 That that you yourself don't understand. 00:39:14.435 --> 00:39:23.080 Yeah. Exactly. It would be a really bad use of of an LLM, whereas an administrative tool like that sounds like a a really good way to to use it. 00:39:23.080 --> 00:39:32.280 Well, it it the thing was it saved me a lot of time. Yeah. And it it did a reasonable decent job of it. So, like, you know, as usual, you argue with it, don't you? Yeah. 00:39:32.280 --> 00:39:41.905 And it says, I'm just gonna do this, and I'm just gonna do that. And you and you've got three choices. You've got yes. You could do that. Yes. 00:39:42.045 --> 00:39:54.090 And you don't need to ask me to do that in the future. And no. I'm going to explain to you why not. And so I did use the no. I'm going to explain to you why not button a number of times. 00:39:54.090 --> 00:40:05.945 Right. It's not really a button because the other thing is it runs in a terminal, which is quite surprising for something like that. Mhmm. And it was it was very good. Yeah. 00:40:05.945 --> 00:40:31.920 It did a good job of it. And I think if I need to make changes, even quite sweeping changes, but maybe not all that challenging Yeah. To this code, It's going to it's really gonna make it a lot easier. And so, yes, it moved it even wrote a script to copy all the entries out of Notion and put them into Postgres Yeah. Which I just thought was a nice Yeah. 00:40:32.000 --> 00:40:32.980 Optional extra. 00:40:33.040 --> 00:40:44.475 What so yeah. So things like, simple administrative functions, non complex data migrations sound like good things to use in LM. 00:40:44.935 --> 00:40:46.215 Yeah. I mean, it's like As 00:40:46.215 --> 00:40:47.115 a coding assistant. 00:40:47.255 --> 00:40:55.830 All automation, you kind of you want to use it for the boring stuff Yeah. So that you can do the interesting stuff. 00:40:55.830 --> 00:40:56.790 Yeah. And 00:40:56.790 --> 00:41:12.865 lo and behold. Although, I still remember that quote from that can't remember who said it. But she said something we've linked to it before. She said something like, I wanted artificial intelligence to do the cleaning and the washing up Yeah. So I can do creative things Yeah. 00:41:12.945 --> 00:41:14.645 Not the other way around. Yeah. 00:41:16.070 --> 00:41:24.090 Yeah. It does feel like sometimes, you know, the general aim is for us to do the washing up and the, an AI to create the art. 00:41:24.150 --> 00:41:32.665 Yes. Yeah. But this did make me quite happy. I was quite delighted with the how that was able to to work. Yeah. 00:41:32.665 --> 00:41:32.905 Yeah. 00:41:32.905 --> 00:41:33.785 I said I think it's 00:41:33.785 --> 00:41:47.490 a good use of it. So yes. But it was just a bit annoying that the day after our episode about AI and programming Yeah. Came out, it all changed. But that's the nature of AI stuff, isn't it? 00:41:47.490 --> 00:41:57.030 Yep. It is. So I have one final Interval thing. Interval thing. So there's a website called podstatus.com. 00:41:57.170 --> 00:42:10.215 Pod status. And on the other day, recently, we received an email to ianandash@whatalotofthings.com. 00:42:10.215 --> 00:42:11.575 That's our email address. 00:42:11.575 --> 00:42:19.890 It is our email address. And, very nicely inquired, hello. How's it going? Pretty good. Thank you. 00:42:20.030 --> 00:42:29.810 But carlos@podstatus.com had some cool information that he thought might interest us. It did. Right. It did. Very interesting indeed. 00:42:30.110 --> 00:42:39.185 So in the category of technology on, Apple Podcast rankings in the Cote D'Ivoire 00:42:39.565 --> 00:42:48.280 You see, Ash, you're saying it wrong. You've got to start off with the category of podcast rankings, we were number something, and then only then say 00:42:50.580 --> 00:42:54.040 So in in the Cote D'Ivoire 00:42:54.740 --> 00:42:57.000 No. That that's made it even more fast. 00:42:57.060 --> 00:42:59.160 We are in the top 10. 00:42:59.785 --> 00:43:01.165 Technology? Technology podcasts. 00:43:02.025 --> 00:43:08.285 And then in Cameroon, we are in the top 100 of technology podcasts. Seventy eighth. 00:43:08.745 --> 00:43:10.505 But in our home country of Britain 00:43:10.665 --> 00:43:12.025 Well, yes. We're not on the list. 00:43:12.025 --> 00:43:13.495 We're not on the list. 00:43:15.250 --> 00:43:24.790 So I just wanna say thank you to all our listeners. I wanna say thank you to Podsteads, and thank you to, all our listeners, especially our listeners in the cam in Cameroon and Cote D'Ivoire. 00:43:24.930 --> 00:43:29.190 Yes. We will include links so you can verify these Yeah. These statements. 00:43:29.250 --> 00:43:32.605 Yeah. And that was my favorite interlude thing. 00:43:35.225 --> 00:43:36.105 That's very good. 00:43:36.105 --> 00:43:42.525 Favourite self aggrandizing interlude thing. So should we move on to, things with a capital t? 00:43:43.150 --> 00:43:45.070 Well, only one thing with a capital t. 00:43:45.070 --> 00:43:51.010 Thing. That's a lot of thing. Yes. Yes. Given we've only got one thing left. 00:43:51.230 --> 00:43:54.210 So, Ian, tell me what your thing is. 00:43:57.230 --> 00:44:00.675 So I enjoy titling my things. 00:44:01.695 --> 00:44:03.475 Put a lot of effort into the title. 00:44:04.095 --> 00:44:13.800 Yeah. So the title for this was Apple is gravely disappointed. And, actually Did 00:44:13.800 --> 00:44:14.760 they do a big sigh? 00:44:14.760 --> 00:44:19.420 That's a bit of an exaggeration because what they said was they're deeply disappointed. 00:44:19.480 --> 00:44:19.980 Alright. 00:44:20.200 --> 00:44:24.540 I don't know if gravely disappointed is more or less disappointed than deeply disappointed. 00:44:25.395 --> 00:44:31.495 I don't know, actually. I would say gravely is more disappointed. Gravely is quite, it's quite the word. It's weighty. 00:44:32.195 --> 00:44:36.855 Yeah. It feels no. I think so. Apple is only deeply disappointed. 00:44:36.915 --> 00:44:38.215 So they could be more disappointed? 00:44:38.275 --> 00:44:39.815 Not all that disappointed, really. 00:44:42.040 --> 00:44:43.260 Oh, okay. Cool. 00:44:44.040 --> 00:44:53.080 Yes. So that was my thing. Monumental thing. Monumental. Yes. 00:44:53.080 --> 00:44:53.475 Yep. 00:44:53.635 --> 00:44:54.515 Yes. One for the valley. 00:44:54.515 --> 00:45:07.095 One for the for the valley. Are you gonna oblige me by asking me what the source of their deep disappointment might be? 00:45:07.560 --> 00:45:11.740 Yes. Why is Apple gravely deeply disappointed? 00:45:12.600 --> 00:45:28.225 Because their customers in The UK will no longer have the option to enable advanced data protection. And they say, especially given the continuing rise of data breaches and other threats to customer privacy. 00:45:28.685 --> 00:45:29.185 Right. 00:45:29.565 --> 00:45:42.920 And Apple remains committed to offering their users the highest level of security for their personal data, and they are hopeful that they will be able to do so in the future in The United Kingdom. 00:45:43.380 --> 00:45:45.640 So what is advanced data protection? 00:45:46.900 --> 00:45:49.800 Ah, well, therein lies a story. 00:45:52.305 --> 00:45:53.265 Gather round, everyone. 00:45:53.265 --> 00:45:55.365 Once upon a time, there were three bears. 00:45:55.745 --> 00:45:56.645 Stay a while. 00:45:59.025 --> 00:46:00.565 So once upon a time 00:46:01.425 --> 00:46:02.425 Oh, there is a story. 00:46:02.425 --> 00:46:29.355 There is a story. Once upon a time, everyone was very happy that you had end to end encryption in your Apple messages, which you might call iMessages, but I think they'd probably not call that anymore. Okay. When you send a message to someone using Apple messages Yep. It would arrive with them, and at no point during the transit was Apple or anybody else able to read it because the encryption was end to end. 00:46:29.355 --> 00:46:41.820 Yep. So end to end means that if you're in the middle and to of an end to end encryption transaction or messaging or whatever, you can't read the messages. Yep. Fantastic. Mhmm. 00:46:41.820 --> 00:46:53.795 And end to end encryption is really a key technology for having any privacy Yep. In the this age that we live in where the Internet is how we have to do almost everything. 00:46:53.795 --> 00:46:57.395 Yep. But government's less keen on end to end encryption, aren't you? 00:46:57.395 --> 00:47:31.465 They are less keen on it. And they in in the case of Apple's end to end encryption, there was a hilarious backdoor, which probably wasn't hilarious if the authorities are kicking in your front door and arresting you. And and the way that the authorities could gain access to your Apple messages was that your backups on iCloud were not encrypted. Mhmm. So while your messages were end to end encrypted, if you made a backup on iCloud of your phone, the decrypted messages on your phone will be included in the backup. 00:47:31.605 --> 00:47:42.320 Right. And Apple had the ability to have access to the backup. And so the authorities would demand Apple give them information from people's phone backups. Mhmm. So 00:47:42.320 --> 00:47:52.075 that wasn't just like a fun oversight, you know, that they forgot to, re encrypt when taking backup. That was a genuine product decision. 00:47:52.295 --> 00:48:14.500 Well, at some point, it was. Yes. Yeah. And, of course, ever since they made that decision at the dawn of iCloud, they've been trying to find ways of of fixing it because I think as a company, and I know some people like to be sarcastic and cynical about this, but they they have made privacy of their users a selling point, and they do actually care about it. 00:48:14.500 --> 00:48:15.000 Okay. 00:48:15.285 --> 00:48:30.520 So what they eventually did was that they made something this ADP, advanced data protection. Mhmm. The biggest impact of turning it on was that if you lost your access to your iCloud account, you were scuppered. Right. Apple couldn't get it back for you. 00:48:30.520 --> 00:48:47.435 Yep. And in order to turn it on, all the devices on your iCloud account had to be up to a certain level of iOS and macOS and everything. So, if you had an old device that was on your iCloud account, you wouldn't be able to turn it on. And when you turn it on, it was full of dire warnings about you can lose access to your account. Yep. 00:48:47.975 --> 00:48:54.155 Okay. So and I think that's probably why I suspect not all that many people turned it on. 00:48:54.375 --> 00:48:56.795 Yeah. Because it was a bit scary when you go to do it. 00:48:57.150 --> 00:49:24.675 But journalists, for example, who whose whistleblowers they don't want people to be able to find out about Yeah. Who they are, you know, those kinds of people for whom it's very important to have good security will have turned it on. Yeah. So Apple are normally quite transparent about this sort of thing. And in fact, they did write, as we have said many times before, we have never built a backdoor or master key to any of our products or services, and we never will. 00:49:24.735 --> 00:49:38.600 But what they haven't been transparent about is the reason why they why they they've taken this measure. Yeah. Because it's only customers in The UK. So if if you've got it switched on in The UK, they actually can't turn it off. Yeah. 00:49:38.600 --> 00:49:43.480 But you have a you'll have a deadline. You have to turn it off yourself or lose access to your iCloud account. 00:49:43.480 --> 00:49:43.875 Okay. 00:49:43.955 --> 00:50:17.245 And new users can't turn it on. So they've not explained in any sentence of this article why it was that they've done this to users in The UK. Okay. And the reason for that, I think, lies in a article from the Washington Post in which it was reported that the UK government has provided a technical capability notice to Apple demanding that they let Apple that Apple lets them have access to these things that are encrypted by ADP. Yeah. 00:50:17.465 --> 00:50:27.410 And in fact, I think the scope of the notice as reported in the Washington Post was it would enable them to have access to the global Apple information. 00:50:27.630 --> 00:50:32.050 Right. So so it wouldn't be particularly possible to separate the two? 00:50:32.510 --> 00:50:46.275 No. Yeah. So Apple's response to this has been to just enable it for UK users. So, effectively, what that means is the UK government have simply made our security worse for just UK users. 00:50:46.575 --> 00:50:48.035 That's interesting, isn't it? 00:50:48.415 --> 00:51:20.425 There's a few things sort of surrounding this. So one of them is and I was sort of thinking, well, it's it's gonna be like they'll need a warrant for any particular data access. Sure. But actually, Matthew Green, who's an industry expert about this, has written a a sort of explanation of it. He quoted the Washington Post, I think, to say, the British government's undisclosed order issued last month requires blanket capability to view fully encrypted material, not merely assistance in cracking a specific account, and this has no known precedent in major democracies. 00:51:20.965 --> 00:51:29.870 So it's basically ours in China. Because I originally thought, oh, it will be, you know, warrant access, but it's not. It's it's unfettered access to everything. 00:51:29.870 --> 00:51:30.270 Okay. 00:51:30.270 --> 00:51:43.185 You do it allows them to do things like speculative trolling to find wrongdoing. Yeah. Sure. They'll just run, you know, code that goes to everybody's phones and finds catches you out doing something Yeah. Talking about something. 00:51:43.185 --> 00:51:46.325 Yeah. So even if you've, like, not been accused of anything. 00:51:46.945 --> 00:51:48.225 Exactly so. Yes. 00:51:48.225 --> 00:51:52.165 Yeah. So did did did Apple sort of fight this then? 00:51:52.640 --> 00:51:54.180 I think they still are. 00:51:54.240 --> 00:51:54.640 Right. 00:51:54.640 --> 00:52:06.165 Okay. Because one of the the premises of the technical capability notice is that Apple are forbidden to talk about it. They're not allowed to Okay. They're not allowed to tell the people they're doing it to. Yeah. 00:52:06.165 --> 00:52:15.545 Hence the very peculiar omission in their post there about the in the post about saying they're doing this, not why. 00:52:16.005 --> 00:52:21.600 Yeah. Yeah. The the the language was quite sort of, specific and studied, wasn't it? You know? 00:52:21.600 --> 00:52:22.400 It was. Yes. 00:52:22.400 --> 00:52:26.080 In order to to explicitly not mention lots of things. 00:52:26.080 --> 00:52:30.420 You know that lawyers had their wrote it, basically. 00:52:30.800 --> 00:52:40.095 Yeah. Yeah. So do you think this opens a bit of a bit of a box? And so it's I didn't know about technical what was it? Technical capability notices. 00:52:40.475 --> 00:52:59.690 Yes. So I guess this is sorry. I'm I'm asking a few things here. So so does that in sort of does that infer that if you don't meet this technical capability notice, then you probably your ability to offer it in in The UK will be I don't know. There there'll be some consequences for it, basically. 00:52:59.750 --> 00:53:12.675 Well, I think Apple complies with the law. Yeah. Its policy is to comply with the law in local jurisdictions. Yeah. So iCloud in China is all hosted on Chinese servers, and the government has the key. 00:53:13.135 --> 00:53:23.130 But this this technical capability notices were part of the the the is a 02/2016 act of parliament that's colloquially described as the snooper's charter. 00:53:23.130 --> 00:53:24.270 Oh, right. Okay. Yeah. 00:53:24.650 --> 00:53:37.475 Yeah. So this was sort of seen as overreaching in 2016 when it was just we're gonna give ourselves the power to do this. Yeah. But now it's they're actually doing it. I mean, who knows how many times they've done it? 00:53:37.615 --> 00:53:49.530 I mean, Apple, it's kind of come out. It's not clear where from, but it's come out. But, you know, who knows if other companies have already complied with these these types of things. 00:53:49.530 --> 00:53:50.590 Oh, right. Okay. 00:53:51.050 --> 00:54:07.185 But it's it makes me cross because, effectively, we're now less I'm an Apple user. Yeah. I have I use iCloud backups. I didn't have ADP turned on. But it's basically means that we're now less secure Yeah. 00:54:07.385 --> 00:54:15.470 Than it's it's not that security pointing at criminals is one thing and security pointing at everybody else is another thing. 00:54:15.470 --> 00:54:37.025 Yeah. You worry for whistleblowers and Yeah. Journalists. So there's a whole, like, sort of raft of of of groups, which have a very compelling reason for, end to end encryption and encryption at rest as well. So I think there's there's, like, different orders of of impact, isn't there? 00:54:37.240 --> 00:54:38.520 There there certainly is. 00:54:38.520 --> 00:54:51.000 Yeah. So for, you know, so for a lot of people, it would just be, like, you know, who cares? But I see that, the whole I've got nothing to hide, question comes up here as well, doesn't it? 00:54:51.000 --> 00:55:16.650 That's just the worst, most pointless argument against personal privacy. Is you know, I mean, back in in in the nineteen seventies, '19 eighties or whatever, before the Internet really came along, if the government wanted to tap your phone, they could, but they need to get a warrant. They needed to put in place a technical infrastructure. Yeah. And so they did it not all that much, but when it really mattered. 00:55:17.110 --> 00:55:42.420 And now because they've got this potential of just having this massive dragnet of data where instead of solving crimes the traditional manner like the five find outers Yeah. My favorite Enid Blyton grouping. Not the secret salad. I was never bothered about them. And the famous five are good, but, you know, I I still find myself a fan of the five find outers. 00:55:42.420 --> 00:55:45.080 Five find outers. It's it's a good name as well, isn't it? 00:55:45.220 --> 00:55:58.765 Well, it's clear. It's on the tip it says on the tip what they do. But, you know, that it used to be that you would tackle you if you you if you wanted to solve a crime, then you would do detective work. 00:55:58.765 --> 00:55:59.265 Yeah. 00:55:59.885 --> 00:56:24.835 And this sort of thing of and and then at the other end of the scale, you've got, like, minority report, the film minority report where they had, people who who had they had special precognition people who could look in the future, and then you could be arrested for a crime you were only going to commit. Yeah. You haven't done it yet. Yeah. And then somewhere in the middle, there's we look through everyone's data all the time to detect crimes. 00:56:25.535 --> 00:57:01.120 Yeah. Yeah. Because there's, like, when I when I kinda read this, like, I was thinking about, well, is there any other examples where you kind of have that level of surveillance, which is then, like, abused and abused quite badly? So you've got, obviously, like, undercover police officers going into certain certain sort of left wing groups, if you like, in the eighties, nineties, and probably beyond as well, who just going in to surveil and then end up doing some fairly dark things while they're there as well under sort of false pretenses. Yeah. 00:57:01.260 --> 00:57:11.280 So I'm like, well, you know, if you give that level of power, then it starts to be used in ways that you didn't intend. 00:57:11.885 --> 00:57:20.485 Well, you gotta just think back to the Edward Snowden revelations back when that that happened. They still haven't let him off that. They're very vindictive. He still has to live in Moscow. 00:57:20.685 --> 00:57:21.745 Think they ever will. 00:57:22.285 --> 00:57:31.280 And they pardoned some extremely Trump pardoned, some very dodgy characters. Yeah. But, yes, I don't think they ever will either. 00:57:31.280 --> 00:57:35.060 No. No. He's that's it for life for Edward Snowden, I think. 00:57:35.360 --> 00:57:42.615 But he, the revelations that he uncovered included, you know, people looking at their neighbors' stuff. 00:57:42.755 --> 00:57:55.020 Yeah. Yeah. Exactly. So it's like the the building of these, systems and and all all ways to look into data. It's like Uber's Uber's famous god mode, isn't it? 00:57:55.020 --> 00:57:55.740 Oh, yes. 00:57:55.740 --> 00:58:23.355 Which was always, like, a good example of extremely nefarious consequences and actions being taken with access to wide ranging data about people and their location and what they're up to. And I think it's just like, you know, those types of tools, I I struggle to to think of justifications for them. Yeah. Because, like, it's it's untargeted. It's not like someone is suspected of doing something. 00:58:23.580 --> 00:58:40.395 It's just like you're just opening the the, like, the door to a world of data just in case you might need it. But once you've opened it, then I think there's always the chance that someone will then abuse that power. And I think history has shown that that will probably happen. 00:58:40.695 --> 00:59:05.420 Yeah. In undoubtedly. I'm just, reading from Matthew Green's article, which, I would say there's a lot of interesting stuff around about this, but I would go to that as a Yeah. Probably number one source. He's writing in the worst case interpretation of the law, which I now find I can remember, is the Investigatory Powers Act. 00:59:05.705 --> 00:59:19.465 Yep. Knowing a k a the Snooper's Charter. The most worst case interpretation of the law, The UK might now be the arbiter of all cybersecurity defense measures globally. His majesty's government, he's written her majesty's government Right. 00:59:19.465 --> 00:59:20.025 Well, lots of other 00:59:20.835 --> 00:59:38.665 that. Could effectively cap the amount of digital security that customers anywhere in the world could depend on without users even knowing the cap is in place. This could expose vast amounts of data to state sponsored attackers such as the one who recently compromised the entire US telecom industry. Yeah. We know about that. 00:59:38.665 --> 00:59:51.660 Yeah. Remember that remember that the hell think about Huawei, five g, hardware. Anyway, it's it's kind of upsetting to me that they they'll do this. 00:59:52.360 --> 01:00:30.040 Yeah. I think it's like like a reflection of, like, not mood, but, like, situation, you know, a kind of extra extra paranoia and, you know, more powers to places where they shouldn't be going to over, you know, an individual's individual's life. Because, you know, as part of privacy, we we accept that there will be people who are doing bad things and using encrypted means of facilitating those thing. But then that's part of privacy's got two sides to it, doesn't it? You know? 01:00:30.040 --> 01:00:43.355 It's like, everyone's entitled to it. Some people will use it for nefarious means. The vast majority of us, you know, just want our data to be left alone. Yeah. But it's like, it's always the way. 01:00:43.355 --> 01:00:52.810 So, basically, in order to counter the very few people who will be using it for nefarious means, you implement something that's so fantastically wide ranging. 01:00:52.870 --> 01:00:53.690 And disproportionately. 01:00:54.070 --> 01:01:00.425 Yeah. Yeah. Exactly. Exactly. So it's the literal nuclear warhead to to for the nail, isn't it? 01:01:02.405 --> 01:01:16.700 Yes. We could go back to Benjamin Franklin. He would he who would sacrifice a little bit of freedom for a little bit of safety just doesn't deserve either the freedom or safety. That's it, I think. That's that's there or thereabouts, I think. 01:01:16.700 --> 01:01:18.640 Took a while, didn't it, that it got there? 01:01:19.020 --> 01:01:22.880 He also said, love your enemies, for they tell you your faults. 01:01:23.420 --> 01:01:28.880 I don't need enemies for that. Oh, okay. In the particular domain of faults regarding testing. 01:01:30.685 --> 01:01:31.245 I found 01:01:31.245 --> 01:01:38.785 a fault. Yes. Yes. Yes. So I'm not sure that was more of a ramp than a thing. 01:01:39.085 --> 01:01:50.190 Well, I guess it it I hope it's not part of a a slope or a further slope into less privacy for the individual. 01:01:50.330 --> 01:02:03.435 It feels like it might be Yeah. Because what government will be tempted by that Yeah. Now that it's Apple has done it for The United Kingdom. Mhmm. Why not America or New Zealand or Australia? 01:02:03.975 --> 01:02:13.690 Yeah. Then imagine if you trained some kind of artificial intelligence on that data to pick out things that you might be interested in. 01:02:14.150 --> 01:02:17.910 Well, I'm sure that definitely isn't happening. No. There's no way that we hear. 01:02:17.910 --> 01:02:20.810 There's no way that they would use that technology to do that. 01:02:20.870 --> 01:02:27.375 Maybe they could just dispatch policemen to your house to arrest you. Just I'm sure there'll be 01:02:27.495 --> 01:02:29.595 automated dispatch of policemen. Hello? 01:02:29.815 --> 01:02:30.535 One one of the 01:02:30.695 --> 01:02:36.395 Hello. Hello. Hello. What what am I being arrested for, officer? I don't know. 01:02:36.455 --> 01:02:38.880 Yes. The LLM told me to come and pick you up. 01:02:38.960 --> 01:02:40.980 Computer said Nick. Computer said, 01:02:42.480 --> 01:02:44.980 fill that chap's collar. Job done. 01:02:45.280 --> 01:02:48.400 Yes. Dear. 01:02:48.400 --> 01:02:52.145 We'll sort out what he's being accused of later. Just bring him in. 01:02:52.145 --> 01:02:54.805 Yes. That'll be us. Yep. 01:02:59.025 --> 01:03:00.325 A profound thing. 01:03:00.385 --> 01:03:01.345 A profound thing. 01:03:01.345 --> 01:03:02.405 A profound thing. 01:03:03.360 --> 01:03:08.160 That's more deep. The monuments tend to go upwards, though. Yeah. Exactly. Profound is Profound. 01:03:08.240 --> 01:03:09.040 Is deep. 01:03:09.040 --> 01:03:12.260 I was trying to think of something that goes with gravely as well. But 01:03:12.560 --> 01:03:17.625 Yes. So I am also disappointed. Apple is disappointed. 01:03:17.625 --> 01:03:18.825 I was gonna say in general. 01:03:18.825 --> 01:03:26.025 I'm I'm just linking back to the title just so that I don't even know why. It just feels like that you should finish where you 01:03:26.025 --> 01:03:28.285 began. You're gravely disappointed. 01:03:28.585 --> 01:03:29.910 Yes. I'm not angry. 01:03:30.370 --> 01:03:33.670 Yeah. I'm I'm disappointed. You know, mildly disappointed. 01:03:34.130 --> 01:03:35.190 I'm a bit angry. 01:03:35.650 --> 01:03:38.390 I haven't reached apoplectic fury. 01:03:42.535 --> 01:03:45.355 Nobody's tried to set a deadline on you, have they? 01:03:45.415 --> 01:03:49.835 No. No. That's true, actually. That's true, actually. No one's done that for a little while. 01:03:50.375 --> 01:03:52.075 Have they been listening to this podcast? 01:03:52.135 --> 01:04:05.940 I don't know. I don't I'm I try and be consistent with my messages around deadlines and estimates And that I don't believe in them. Do you want me to say that right. Okay. 01:04:06.640 --> 01:04:08.020 So the only Look 01:04:09.360 --> 01:04:11.220 alert. Look. I look interested. 01:04:11.280 --> 01:04:11.940 Look interested. 01:04:12.320 --> 01:04:40.525 So the only reason team's estimate. There is no that it is the team's estimate. There is no test estimate, and testing can only ever stop. It can never finish. And I can stop whenever you like. 01:04:46.745 --> 01:04:47.565 That's it. 01:04:52.390 --> 01:04:54.390 Oh, dear. Do you feel better now? 01:04:54.390 --> 01:05:01.690 I do. I do. We'll have to put together a a greatest hits compilation, won't we? We can just do all of my runs. 01:05:01.830 --> 01:05:08.105 Well, yeah. And you could, make put them on a soundboard app, and then you could use them in meetings instead of going to them. 01:05:08.105 --> 01:05:20.040 Yeah. It's like, oh, I've said I've said this a million times before. What would I say? I I can just have, some kind of, AI AI assistant, which is me Yes. Basically. 01:05:20.040 --> 01:05:25.580 And whenever it hears the word test estimate in a meeting, it just starts playing that clip. 01:05:25.640 --> 01:05:38.535 Well, we could actually we could we could make an AI we could make an AI that knew all of the things we've ever said on episodes of what's a lot of things. 01:05:38.535 --> 01:05:39.195 Yeah. True. 01:05:39.495 --> 01:05:41.035 And then you could ask it. 01:05:41.175 --> 01:05:45.970 What does Ash think about test estimates? Press the button. 01:05:46.110 --> 01:05:59.945 Yeah. Yeah. It will be very quick. Oh, dear. Well, anyway, thank you for enjoying that thing. 01:06:00.245 --> 01:06:03.225 What? We don't say that. 01:06:04.405 --> 01:06:05.945 So how do you get in touch, Ian? 01:06:06.485 --> 01:06:08.665 Well, I think we've already covered it, haven't we? 01:06:08.725 --> 01:06:15.420 Okay. Because Carlos Carlos, indeed, got in touch. From pod Status. Status. 01:06:16.040 --> 01:06:17.420 Yep. Got in touch. 01:06:17.640 --> 01:06:20.780 So, basically, if you get in touch, you pretty much guaranteed a mansion. 01:06:22.360 --> 01:06:27.215 Yes. And and then, you know, with that piece of glory, you can You 01:06:27.215 --> 01:06:30.415 can dine out for You can dine out for months. For months on that. 01:06:30.415 --> 01:06:41.490 Yeah. Well So are there any closing thoughts? Maybe we ought to do like Jerry's do you remember Jerry Springer? Yeah. So I must admit. 01:06:42.670 --> 01:06:54.525 My I only watched a couple of them ever. And, in them, people would go and say stuff to each other, and then some other people would have a fight. And then people in black would come and separate them. 01:06:54.525 --> 01:06:58.045 And then then there'd be like a paternity or maternity test. 01:06:58.045 --> 01:06:59.645 Yes. And then there'd more fighting 01:06:59.805 --> 01:07:00.325 Then more fighting. 01:07:00.445 --> 01:07:15.860 And shouting. Yeah. But the thing that always happened at the end was that he would provide the audience with his final thoughts, but they never were because he always did it the next time. He he's I suppose there's a last episode somewhere with his actual 01:07:15.860 --> 01:07:19.400 final thoughts. Yeah. That's true actually. Yeah. But Would have got there eventually. 01:07:19.780 --> 01:07:45.110 But he he would summarize what he felt Mhmm. His audience ought to have learned from from watching the show. Maybe we could come up with a final thought about what we think people should have picked up from the glorious opportunity to listen to us for the last blimey. At least we're not to an hour and a half yet. 01:07:45.650 --> 01:07:54.705 So I think it's that's a good time to not add any final thoughts. Alright. So, dear listener 01:07:55.745 --> 01:07:56.645 Dear listener 01:07:57.265 --> 01:07:58.405 thank you for listening. 01:07:58.625 --> 01:08:01.365 Thank you for listening. The end.