Leo [00:00:25]: And we're back after a slightly longer than usual hiatus. Yeah, there's a reason. How was Peru, Gary? Gary [00:00:32]: Oh, yeah, I had a great time. I did a kind of an adventure travel thing through Peru. So I saw various different parts of it. I hiked the Inca trail, which is the highlight for me, and you know, it ends at Machu Picchu and I got to see all sorts of cutting edge 15th century technology. Well, which is kind of true because, you know, the, the thing is their engineering, I mean that's what the, that's the takeaway from my perspective was the Inca civilization. They were incredible engineers, right. I mean the buildings they had earthquake proof, stood the test of time to where they're just there now. It's in fact, I had to stop, like find a better term here because you want to say, oh, here's some great Inca ruins. Gary [00:01:26]: The thing is, there weren't ruins, they were like complete. I mean, the only thing missing is the roofs, which were wood, which would need to be replaced on a regular basis, but the stonework that was all perfectly intact, including the Inca trail itself. So it's kind of interesting. And also the irrigation was incredible. I mean, I mean you can still walk in towns there that basically the streets and the irrigation channels around the streets were built by, you know, the Inca civilization 600 years ago. And it's still the same now. And I, you could see it when it rains, actually running off the water from the, the road that you're on. And it's kind of hard to believe that it's like, oh yeah, from like this level up it's an Irish pub. Gary [00:02:13]: But from like basically, you know, this level down, it's an Inca wall and an Inca street irrigation channel that's just been here for 600 years. Wow. Leo [00:02:22]: Yeah, I gotta, I gotta agree that the, the Incas, the Aztecs, the Mayans, all of them, they all had pretty amazing civilizations that I don't think we really give enough credit to when we're comparing against, you know, things like quote unquote, Western accomplishments, so. Gary [00:02:37]: Exactly, exactly. Yeah, it was, it was quite impressive to see all of that there. And some of the terrorists, the terrorist farms are still being farmed. I mean, they're actively being farmed now. They never stopped. You know, a lot of them are like World Heritage sites and things, but, but some of them are just still in use because there's so many of them that, you know, why build new terraced farms along in the mountains when you have these existing ones, perfectly good Leo [00:03:03]: 600 year old terraces yeah. So I got an interesting question. Gary [00:03:07]: Yeah. Leo [00:03:08]: The other day. It's one of those that, you know, you. You read it and you go, ha. Oh, yeah. It made me think. The question was this, how do I know that you're not AI? Gary [00:03:20]: Yeah. Leo [00:03:21]: And the problem is that the true answer is you don't. I. I cannot prove that I'm not an AI right. Now, granted, today's AIs are, I'll say, you know, limited. I guess there's somewhat obvious if you. Especially if you know what to look for. But as I keep saying in so many different places, today's AI is the worst AI you will ever experience. Meaning, of course, that everything's going to get better. Leo [00:03:57]: So the fact that today an AI video of someone doesn't have them doing things like scratching their nose or, you know, you, whatever, these random things that aren't really part of the presentation, but they're just part of what it means to be living in a human meat sack, they're going to start replicating that to the point where you won't be able to tell. And so that's one thing that I started to think about. It also falls into that category of you can't prove a negative, right? You can't prove that something doesn't exist. You can't prove that I'm not an AI. But it also sent me down a different path. And that is, does it matter? And I mean that in the sense that especially with what you and I do, we provide information, we provide education, we provide answers. If you're getting an answer from a source that is correct, is complete, is easily understandable, is presented in a way that works for you, does it matter? Does it matter that that source was AI generated or not? And I'm not going to say that it doesn't matter. There's reasons that I suppose it could, but I think that that's a really important question to start thinking about, because like I said, it's only going to get better. Leo [00:05:36]: We're all going to get fooled at some point, right? There are a couple of YouTube channels right now where I'm not convinced. Right. There is someone whose presentation is so slick that I could see it being a real person who just happens to behave that way. But I could also see it as being 100% AI generated. And I asked myself the question with that. If I didn't worry about that, if I didn't care about that, am I still getting value out of whatever's being presented? And the answer in that case was yes, right there's good stuff coming across that channel. So I don't know, it just, it was like I said, it was one of those questions that you get every once in a while that sent me down the philosophical rabbit hole. I'm kind of, kind of curious where you are on all of this. Gary [00:06:30]: Yeah, well, I've actually got a couple of related topics to that. And the first is, is that I may have already been fooled or at least I don't know, because. So, okay, both you and I, I'm sure, get this where we get pitches from people. We get an email or contact form Leo [00:06:45]: is filled out, single day, multiple. Yeah. Gary [00:06:48]: And they want to. They want to, oh, I have a new app out, you know, will you review it or feature it on your channel? That kind of thing. And it's annoying for me because I actually say in my form, it's like, I don't do that, you know, so they're ignoring that. They're just filling out the form or f found an email address for me somehow and they ask it. But one of my weird complaints for years was that they usually don't even bother to pretend. They just throw. Here's their RPR pitch. We've got a new app. Gary [00:07:17]: It would be great if you covered it. And I thought, hey, hey, butter me up a bit. Just first say, like, oh, I love MacMost. I love your channel. I've been watching your videos for years. They don't. They just go right into it. Which is fine. Gary [00:07:33]: I get that they're actually using an app or some other service where goes out to 100 or 1000 different people like me at once, which was weird. I recognized a few weeks ago I got one from somebody that actually the first paragraph was, hey, I've been a big fan of your channel for a while. It's helped me learn a lot about using my Mac. I particularly like this video. I was like, oh, how nice that they actually took the time. They seem to be a real fan of my work. And then they asked to review a product. So I probably delete the, you know, email because I still don't do it. Gary [00:08:09]: But it was nice. Then I felt it was weird because the one I got a couple days later also called out some things. At the beginning I was like, wow, okay. People are changing. It seems like after years of getting these more anonymous ones, I just happened to get two in a row that were from real people. Then a third one, then a fourth one. Then it's been for a few weeks now. And now it's like, well, wait A minute, something switched, something changed. Gary [00:08:32]: And I highly suspect, but I have no way to prove it, that there's some sort of tool or maybe a technique going around where basically, you know, if I just asked ChatGPT, Hey, I want Gary to feature my app, compose a message for me. It would probably. ChatGPT will come out with, hey, I'm a big fan of MacMost. I particularly like your video on whatever. And here's this new app I wanted to talk about something like that's got to be going on because it was a really abrupt switch. It was from almost never do I get those kinds of requests to. It seems like everyone now all of a sudden as of like a month ago. Leo [00:09:12]: What cracks me up is that like I said, I get them to multiple times a day, probably, I don't know, half a dozen or a dozen of these kinds of requests a day. And they're asking for links from the website or paid article posts or you YouTube sponsorship opportunities. There are a couple of classes of these that just crack me the heck up. One are indeed the, the AI generated versions. It's pretty clear that, you know, nobody's really done anything other than point AI at the site and the AI has selected like some 15 year old article to gush about, right? And you know, it's like, okay, well you're not paying attention. But the other are some of the, especially on the YouTube side, the product placement suggestions that I get are like completely off the wall. They have nothing to do with technology. They're like, it's, it's like, you know, we have these women's dresses that we think would be a great fit for your channel. Leo [00:10:16]: Yeah, no, no. So it's weird. Like I said, there's. And of course the reality is it's like there's. They're coming from multiple sources, right? Some sources are just doing the bare minimum, reaching out to everybody, like they've been doing it for the past 10 or 15 years. But some of them are on the edge, on the bleeding edge, so to speak. Using the technology to try and sound a little bit more appealing to folks like you and me, I, and like you on my website, I have statements that say, I don't do product placement, I don't do paid links, I don't do paid articles. You know, it's one of the things that I'm quote unquote proud of is that everything on Ask Leo was written by Leo. Leo [00:11:02]: So it just goes against, you know, everything that I stand for to even consider these kinds of things. And, and yet they keep trying and you know that it's all 100% automated. There's not a real person who suddenly saw something about your site that they liked. Gary [00:11:23]: Yep, indeed. So I can't be 100% sure that there has been an AI switch, but I suspect it. But then that just goes back to what you were saying before. It's like, is it a person? Is it AI? Is it, does it matter? It doesn't matter because the email gets deleted either way. And I just feel bad if it actually was somebody that was like, oh no, I do like your stuff and all this and I created this app and it's like, oh, great, that's wonderful. Still not going to mention it because I don't do that, but nice to hear from you, you know, that kind of thing. Leo [00:11:53]: Somebody slaved over a hot email for hours crafting it to get your attention. Gary [00:11:58]: Yeah, sure. Now, you know, with the topic of does it matter, I want to talk about a little bit about AI music, which has been in the news a lot the last few days, and how I recently used it for something. And this also ties into something you mentioned months ago and I reiterated at some point. When you use something like a photo, you know, you run into the problem of like, oh, somebody can claim copyright infringement, right? Leo [00:12:31]: Absolutely. Gary [00:12:32]: And we both run into a problem where we legitimately obtained photos and then you still get, you know, hit with it. Same thing goes with music. So you. Because obviously if you take a piece of music that's like from a known artist or, you know, whatever, right, it's. That's copyright infringement if you're using it. However, it still happens with Clipart, as has happened to me in the distant past, you get something from a legitimate source and then you still get hit with a copyright claim because either it's was licensed from somebody and there's like, you know, so there was the originator of it and then somebody took it and put it into a collection and claimed, right, that oh, you have the rights to do it, but you actually didn't. Or it's like they're, they're doing a trolling thing where they took these, this music, they copyright it, then they hit you with the copyright thing and sure, you could go and take them to court and win, but you know, it's like, you know, it's a problem. So to get away from all that, an easy way to do it is to use AI, which, you know, you suggested. Gary [00:13:37]: And I also do, you know, use AI to generate images when it's really not that Important. I do a monthly newsletter to just my patrons. And I had this idea to do a whimsical scene that has a Mac in it and just has something to do with the month. You know, if it's November, it's going to be a bunch of leaves lying around. If it's like summer, it's going to be a beach or whatever and I throw a Mac in there, somebody using a Mac and it's fun. But I could go and try to find artwork like that. It would be hard, harder than just typing in a prompt. I could go and pay somebody to create a photograph like that, but that would be just a huge luxury cost to go and have that done. Gary [00:14:18]: So instead of the alternative of having no image, I just have a little AI image. And I just found it whimsical. And it's just, you know, not, not meant to be anything serious. Same thing here for music. So you can go to sites like Suno and you can have it create music. And there's a lot of people talking about this, how artists are using it, how non artists are using it, how Spotify is filling up with AI music, that kind of thing. And how AI music has no soul because it's not created by a person and all of that. But I created a video from video clips I captured while on my trip to Peru. Gary [00:14:59]: I actually did a cool thing. This is worth mentioning. And I'm probably going to do a video on this at some point. In addition to taking regular photographs, I used a different video app. In this case it was Final Cut Camera, but it could be any video app as long as it doesn't put the videos in your camera roll. It has its own little library of videos. And I had a little button on my phone that, that just took me right there and allowed me to start filming. And I filmed many things a day. Gary [00:15:27]: Usually not like the things you would take pictures of, but stuff like walking down the street, walking through an airport, panning from left to right when there's something visually stunning in front of you. Tons of movement oriented things that were like 2 to 10 seconds long. And I just kept doing it out of habit, just go, go, go. And when I was done, I had about 400 clips nicely in order that I shot them. I took those 400 clips and I dumped them all into a video editor and I cut them all to two second intervals. In other words, most of them I just cut down to two seconds. Two seconds. Two seconds. Gary [00:16:08]: Well, this one's more interesting. Four seconds. Two, two, two, six, two, two Four, you know, and all this just to create an even pace to it. And I got it down to exactly 10 minutes and it was a really nice flowing video, always moving of the entire trip from when I left to all the different things I did. Lots of panning around like tables where people were eating, lots of stuff like that. And, and it was, it's nice. It really gives you the. And I, I made it mostly for myself. Gary [00:16:39]: I'm going to enjoy watching this thing in a few years, right. And in, in retirement I'm probably going to watch it more than a few times as well. But the weird thing is, is the sound was all over the place because most of the time there was no sound associated and a lot of times it was just background noise. So silence, the whole thing. Now I have a silent video which just seemed weird to watch it in silence. Right. Be nice if I had some music to match. So what do I do? If I use real music, then I run into issues. Gary [00:17:14]: If I put it on YouTube and let YouTube put music onto it. Well, now I can only use that video at YouTube. It just exists as a YouTube video. So I went to Suno and I had Suno generate some music. Frustratingly, no matter how many times you ask Suno to give you something 10 minutes long, it says here's your 10 minute song. That's 4 minutes and 32 seconds or whatever. But I, eventually I was like, well these are all fine. I could fade one into the other. Gary [00:17:37]: And I created like three in fade of the moon. So I've got this music and I've got several compliments from people on the music. I'll include the video, a link to the video here we can, you know, in the show notes. Leo [00:17:48]: Sure. Gary [00:17:49]: People want to see it and it, it does make the video. I mean the video goes from being just this weird silent thing to actually having some life to it. But it's generic AI Like I asked it for like folksy rock, Peruvian South American music for traveling. You know, I was like descriptions like that and I varied them and that's how I got the three songs. And sometimes it actually hits stuff hits on the beat because the randomness of that is life. You know, if you have enough 2 second clips and you have enough music every once in a while like the music changes tone right at the right time and you're like, oh, I planned that. But you know, it's an illusion. But the thing is, it's like what? You know, there was no way for this little thing I was going to spend any Money having music generated. Gary [00:18:40]: I don't have the talent to generate the music. And so what was the alternative? I mean, sure, I do agree that AI music has no soul to it, but I'm not claiming that this is like an album and people are going to want to buy it. This is just background music. Sometimes music's background music, sometimes art, you know, images are just a little extra splash of color to something you're doing. I, and I think too often in the debate that gets overlooked. You know, there's a place for it and there's a place not for it. And most of the time people are talking about this stuff. I agree that AI music, it's like I don't want to see it. Gary [00:19:22]: I don't really want to see it on Spotify or Apple Music. That should be a place where you get human music made by humans and you listen to it. If you want AI music, you should be able to get that too. Like for instance, if I want something that lasts for two hours and has a background music to help me concentrate, well, maybe AI music is good for that. But if I actually want to hear what artists create, then I am like, I don't want to see that on Spotify or Apple Music or any of the other services. I want it to be very separate kinds of things. Leo [00:20:01]: Yeah. To me the most important thing about all of this is transparency. Gary [00:20:06]: Yeah. Leo [00:20:06]: Somebody trying to pawn off AI generated. Anything as human generated, that's a non starter. But being AI and upfront about it. Absolutely. So you said that SUNO will now do four minutes because the last time I looked at it they would only do 30 seconds, so. Gary [00:20:25]: Oh yeah, no, it definitely does. And I had to pay, so I had to pay 10 bucks for a one month subscription because it's very adamant that you don't own the copyright to any of this and you can't use it except for just to hear unless you're a subscriber. But as soon as you go to the ten dollar level then, okay, now you've got, you're creating stuff and you own the copyright and you could do stuff with it. So I wanted that. And 10 bucks was an easy sell. It was like I could be, I could be the rest of the day trying to figure this out or I could, I have these three songs, ten bucks. Now I have the three songs, I put them on the music and you know, it's like two minutes later I'm rendering out the video and you've got Leo [00:21:01]: 29 more days to make more decisions. Gary [00:21:05]: Yeah, well, yeah, Believe me, I've thought of that. It's like, oh, you know, but then I don't want it to be like, oh, I've spent $10. And now I'm also going to spend a bunch of my time trying to think of what I would use it for. It's like, just let the $10 go. Like if somebody came to me and said, here's 10, exactly 10 minutes of music for your video, pay me 10 bucks, I'd be like, done. So, you know, I just have to just take it at that. Leo [00:21:29]: So couple things as you were talking. One is, don't know if you've noticed, but on Ask Leo, I've done the most of my article images now include a corgi just because. Right. I think I love. It was whimsical. Right. Just something to, to have a little bit of fun, you know, to get people to smile before they start reading my articles. And again, I think I mentioned it before. Leo [00:22:00]: I do all of my AI images in technically, I call it Pixar style, it's 3D animation style. So whatever Gemini comes up associated with that is that it's not an attempt to fool anybody into thinking it's anything other than AI generated. And of course, all the images are credited underneath. But what I stumbled onto, I think last week one of my friends shared a video of some music on, on Facebook. And I still need to look up the, the YouTube URL, because that's probably a more useful one. And it was, you know, clearly country western music with a, a message. Right. It was something that this person really resonated with and the alarms were going off. Leo [00:22:58]: Right. I was looking at this and thinking, okay, that sounds really good, maybe too good. It looks really good, maybe too good. And I ended up doing a little bit of research. And no, they are not upfront about whether or not this is a real person or an AI generated person. And if you take a look at the discussion in various places on the web, there are two themes that come across. People are absolutely claiming this is AI and then there's another group of people that are absolutely in love with the music. Right. Leo [00:23:48]: So this goes back to the. Well, does it matter? Yeah, it's. I just, I just found that so fascinating that, that something which is clearly cutting edge AI kind of stuff is going to. Is generating that kind of a positive response for what's been generated. Gary [00:24:11]: Yeah. Especially if, I mean. Yeah. I mean, it doesn't matter if, even if you're a human generating human art. I mean, if you get 10,000 people liking your stuff, that's all you need. The other 8 billion can be either they don't know who you are or they could actively hate you. It doesn't matter because only the 10,000 that like you, right, matter. So if somebody generates, you know, AI stuff and is profiting from it in some way, and they can ignore the rest, you know, ignore all that and. Leo [00:24:50]: I don't know, here's a question for you. Is the. Has the talent required not really changed as much as. I'm sorry, not really been taken away as much as shifted up a notch? Because what I will say is that let's assume that the singer's videos are. Is. He's an AI singer, right? He's completely AI generated. The people that created him are incredibly talented. They're just talented at creating AI. Leo [00:25:31]: I've mentioned this before elsewhere as well, is that one of the things that AI is forcing a lot of people to do is think through their questions before they ask the AI because they. Gary [00:25:42]: Right. Leo [00:25:43]: Basically, the better a question you can ask, the better an answer you're going to get. And that too. It's not a search engine, it's. It's something else. But it's, It's a different level of skill that ends up getting used, but it's still a skill. Gary [00:26:00]: Yeah, I saw somebody post recently in a reaction to, you know. Well, the first person posted, sometimes when I think through how to prompt AI for something, just the act of thinking it through brings me the answer. And then somebody actually posts posting a response to that, congratulations, you've just reinvented thinking. And I mean, it's true, but I don't, I don't know if it's unique to AI because I think it's been, it's sometimes, you know, calling customer support. You know, I'm sure every customer support person that's worked, worked long enough has had the, okay, explain to me your problem. And as the person's explaining the problem, they're like, oh, wait, hold on. And now I see it myself. Leo [00:26:43]: I mean, I've, I've gone to the point of writing up a nice long post for, I don't know, a support forum or, you know, or, you know, any of the other, any of the other support forums that are out there. And by the time I'm done writing my question, it's like, oh, that's what the problem is. And I just delete what I've written because I now thought of the answer. Yeah. So in the AI theme, continuing the AI theme today, one of the Things that happened while you were away. Is anthropic announced Mythos or depending on who you believe, Mythos, which kind of cracks me up. I've heard, I've heard about it in several places, but one of the podcasts that I listen to originates in the UK and he was adamant that it was Mythos. So he actually asked Claude for the pronunciation. Leo [00:27:45]: And of course Claude said Mythos for whatever reason. And yet, you know, you go to the dictionary or whatever and it's Mythos. So controversy, yay. Supposedly Mythos is extra good at identifying vulnerabilities in software, including existing all major operating systems, many major applications, and so forth. To quote their press release, we have identified thousands of additional high and critical severity vulnerabilities that we are working on responsibly disclosing to open source maintainers and closed sourced vendors. Thousands is the thing that jumps out to me there. And they've started a project called Glasswing, which is essentially them allowing a select few, a lot of these major players like Microsoft and Google and Facebook and whomever else, access to the Mythos AI model before it is released publicly so that they have an opportunity to deal with those thousands of identified vulnerabilities before they get made public. The question I have, of course, is, is Mythos just as good at fixing these things as it is at finding them? That's not a flip question. Leo [00:29:08]: It is very possible that the same level of intelligence that allowed it to define these things and apparently string multiple vulnerabilities into, you know, into an access path, it could be just as good at fixing things. The thing that I thought was most fascinating about this, there's a lot of things that are fascinating about Mythos, but the one that really got me was that this was not intentional. This was in no way intentional. They just made the next really good large language model. And as it turned out, one of the things that it's really good at, among many other things, I'm sure, is finding vulnerabilities. So, okay, great. Does that mean that the next model is going to be even better? And the model after that, and the model after that? You know, at some point Mythos is going to get out. And when I say get out, that means that either anthropic will decide that, you know, they've given everybody else enough time, the public needs to, you know, should have access to it, or Mythos is in the hands of like several dozen companies. Leo [00:30:28]: What are the chances that it's not going to get into the hands of somebody that shouldn't have it. Gary [00:30:32]: Right. Leo [00:30:34]: Like I said, is this the future of AI model releases? So I've got a couple of articles that I'm going to link to in the show Notes. One is Anthropic's own press release which is fairly lengthy. It's an interesting read. Assessing Claude Mythos previews cybersecurity capabilities. And then there's an article in Hacker News which I think was a slightly more, I'll just say objective approach to it. Anthropic's Claude Mythos finds thousands of zero days flaws across major systems. It's fun times out here on the bleeding edge. Gary [00:31:12]: Yeah, I mean, I think, yeah, this is a, I mean finding flaws in unreleased software is going to be like a major use for this, which isn't talked about much. But you know, just making your stuff more bulletproof, that's all you could do is make it more bulletproof. Right. You can never make it 100% right. But. But it will certainly make it harder for humans to find vulnerabilities. If an AI can go through and pass. Leo [00:31:42]: That's the thing, right? Depending on humans don't have to find the vulnerabilities anymore. Gary [00:31:48]: Yeah, well, yeah, that's true, but so you just have. So there'll be like a war of like. No, you, your, your vulnerability finding AI has to be better than the bad guys vulnerability finding AI. And technically it should be because it should have more information and it should have it earlier. Leo [00:32:07]: In theory. Gary [00:32:08]: In theory. Leo [00:32:09]: So honestly, we've just described the, what I've been calling the race for years, right? Yeah, the race between people creating malware and the race between people patching or detecting malware. And the bad guys are always ever so slightly ahead in that particular race. In this one, you're right. Theoretically the people who have the software, who are creating the software are able to give, you know, their AI more information sooner. But I think what's scariest about this is that this is finding vulnerabilities in existing software, software that's already out there. And that's why this project, Glasswing exists, so that all of these creators, all of these vendors for all this existing software have the opportunity to do something about it. It's been rumored that we should be expecting lots of patches over the course of the next few weeks as the result of this research. Gary [00:33:06]: Well, and so I have this thought about it is, you know, there, there are flaws in software that people could exploit for gain. Right. It's always got to be for gain. Right? Gaining access to money, gaining access to influence, gaining access to information. That's why you want these flaws. A long time ago, it was the easiest thing if you wanted to gain, to get us find a software flaw, you know, create a computer virus, for instance, 20 years ago or whatever. And then at some point, it started to get easier just to scam people. You know, phishing. Gary [00:33:44]: It was just easier just to go and basically ask people, what's your password? But in a way that they would actually give it up instead of trying to design software that would somehow steal their password or steal access to it. And so we crossed kind of this. This thing where, yeah, there are people still trying to do software vulnerabilities, but the human vulnerabilities were easier to get to and. And we never turn back. Right? The human stupidity never gets fixed, and the. But the software gets more and more secure. So now we're getting into this realm where more and more obscure vulnerabilities are what it takes to compromise software. And it just becomes even more obvious that if you want to profit from, you know, getting information, somehow, it's the human vulnerabilities that it takes, you know, to do it. Gary [00:34:41]: Like I've always said, it's like nobody's doing a huge, you know, scheme to try to steal your credit card online. When you just hand your credit card over to a waiter and, you know, and they take it in the back, Something is still done in the United States. As somebody who just traveled internationally and did not once take the credit card out of their wallet and use the phone 100% of the time. Leo [00:35:05]: Right. Gary [00:35:08]: It was shocking to come back and within 48 hours, having taken the credit card out of my wallet so many times just to eat at restaurants and things like that. I was just at a. Just at an office, too, where I had to pay somebody something, and they. They not. They couldn't even swipe the card. They actually needed the numbers on my card to actually. For me to pay them to do something wild and wild. And it was so wild that I actually went and said, at this point, you have an own. Gary [00:35:40]: You have a way for me to pay you online, don't you? And they're like, yeah. And then I sat down and I did that on my phone, and they said, oh, the payment just came through. I was like, yeah, it just was incredible. But the thing is, it's, you know, these human vulnerabilities, they just become more obvious that that's the way it should be used. So. So, yeah. So I don't know. I don't know. Gary [00:36:02]: It's obviously we need to patch all the things that need to be patched. But at some point, it's just like normal people never have to even worry about this stuff anymore. They need to worry about the much more obvious, you know, oh, this is your bank calling. We have a problem. Please confir Social Security number. Leo [00:36:20]: I believe that, that you're correct. The average person probably doesn't need to worry about this too much. Gary [00:36:27]: Yeah. Leo [00:36:27]: Where I think that this is probably a little bit scarier. Gary [00:36:32]: Yeah. Leo [00:36:33]: Are in what I'll just refer to as corporate espionage. Sure. And nation state activity. Gary [00:36:40]: Yep. Leo [00:36:41]: Right. Gary [00:36:41]: Spicy. Leo [00:36:41]: These are the kinds of things that, you know, if somebody can breach the government somewhere, I mean, yes, they can do that by asking a person, I need your login credentials. I'm from it. Right. But in reality, especially in the government and you know, some corporations at least, you know, people do get trained to a slightly higher level of paranoia than the average consumer does. Which makes, I think, these kinds of identified vulnerabilities at least valuable or useful for, you know, the North Korean hackers trying to get into whatever they're trying to get into. Gary [00:37:22]: Yep. So speaking of security. Leo [00:37:26]: Yes, please do Gary [00:37:29]: so. Longtime listeners will know that there's a pattern here. The pattern is that one or both of us go on a trip and we come back with a story of how frustrating it is to use VPN on said trip. This will be the last time I do that because I'm done. I'm done with using vpn. It has stolen my last hour of vacation time on this trip. Every time you travel, the VPN is very nicely designed for staying at the nice Hilton with the nice WI fi and your VPN will protect you there and all of that. It is not well designed for staying in many different places, connecting to many different kinds of WI fi's all using different ways to connect. Gary [00:38:12]: It consistently fails all the time with playing the whole game of I can't sign on to the hotel's WI fi because the VPN is protecting me because I'm not on the WI fi yet. All this stuff. And I still kept running into that still. And I was only using my iPhone on this trip. And I still kept running into the thing where I would have to struggle to connect. And sometimes just switching off the VPN was the only way to connect once I was connected, turn the VPN back on. And it usually worked, but thanks. Now that's like 10 minutes I won't get back from my trip of, you know, in frustration and it even got worse. Gary [00:38:56]: I usually, I like to find out why things didn't work. In this case I wasn't able to. But in the middle of my trip I stopped being able to use iMessage. Now, iMessage is Apple's messaging system that works completely through the Internet, right? Does not rely on SMS, RTs, any of that. It's just an Internet technology. You can get an iPad with Wi Fi, there's no mobile phone connection and you do it on your Mac. It's just Internet based, right? But despite that, in the middle of the trip I stopped being able to communicate with people over imessage even though Leo [00:39:34]: I had VPN turned on, right? Gary [00:39:36]: That was with VPN turned on and off, right? So it was turn off. And I was able to load web pages perfectly fine. I was able to check email perfectly fine. Imessage failed to work. So troubleshooting that for a little while, I came across some possible explanations, none of which I was able to confirm because partially because I was short on time, I just needed to connect and move on, right? And VPN kept coming up. And the way I think things went down was that when you use imessage One of the IDs you have is your phone number. You could say, you could set up imessage and say people can contact me with and you can list a number of email addresses, your Apple ID email address, but also others. And you can list your phone number which is like default on the phone. Gary [00:40:33]: So people can go use imessage with only your phone number. And Apple's smart enough to say, hey, you're trying to talk to this person with this phone number using imessage. We have them in the system. We're not going to go and send this sms, we're going to send it full imessage. So I wasn't using my mobile phone provider. Of course I was in another country that is not supported. It's expensive to go and say, yeah, extend, give me international roaming, right? It's like it's 12 bucks a day, so AT&T charges. I just got an esim. Gary [00:41:06]: I've done that in the past. You get an esim, you could buy it with an app. $24. I had 10 gigs, twice the amount I get with AT&T for 24 bucks. I know, it was like, okay, easy. And it worked. Leo [00:41:19]: That just gave you another phone number? Gary [00:41:22]: Well, no, I bought a data only esim. Leo [00:41:25]: Oh, okay. Gary [00:41:26]: If I had bought a full esim, I would have gotten another phone number, a Peruvian phone number that would have been mine temporarily, but that was of no use to me because nobody had that number and you know, whatever. So I just said I just want a data esim. It's che, it's all I want and for the first week, no problem, I've used it before in far away places, no problem. Then in the middle I stopped being able to use imessage and I kept finding references to oh, you're using an esim. Oh, you haven't been connected to your sms, your real SMS number in a while and now imessage is getting confused because of these things and your phone number is no longer associated. People who were just using imessage with your email address, that's fine. But people that were sending you imessage through your phone number, that was what's getting confused and you don't really have any control like you could, somebody could just choose one, as my wife probably did a long time ago or whatever. And it just basically got all confused. Gary [00:42:34]: I actually had to go to Google Voice, which I still have a free grandfathered account at Google Voice and use the SMS there to send a message with our safety password. To be like, this really is me and I can't for some reason connect. And then I don't know how I fixed it. One of the many things I did was I connected using AT&T for a second for which they proudly went and said welcome to international roaming, 12 bucks, you know, and then I turned it off immediately and also I changed networks a few times. I went completely WI fi all this and it kind of got restored and then a day later everything seemed to be working fine. Leo [00:43:17]: So you, you paid for one day of AT&T. Gary [00:43:19]: Yeah, and I don't know if that even fixed it. Like I don't know if that was what it was, switching to another WI FI network, whatever it was, I don't know what actually fixed it, but all I do know is in trying to research it, VPN kept coming up over and over again using iMessage over an eSIM for a long period of time and sometimes people reported problems anyway. The whole thing really made me regret the almost non existent security I got through vpn because of course I'm using SSL for every single thing I'm doing. Leo [00:43:54]: Sure. Gary [00:43:54]: I'm not connecting to anything that were the DNS would have been an issue. I didn't do any online banking, you know, I didn't do any retirement planning, I didn't do any secure corporate anything. You know, it was just, there was nothing that you could have gotten from my DNS. So the Whole thing seemed like, oh, I was being overly cautious using vpn. And what I paid for was probably an hour or two of my vacation time over the two weeks for trouble connecting, and then probably an hour or two dealing with this issue, which I don't know if it was caused by it, but I decided I'm done. Coincidentally, right around the time I was coming home, I got a renewal notice from my VPN provider. And with this fresh emotion, I hit the cancel button. Leo [00:44:43]: Thanks, but no thanks. Gary [00:44:44]: Yeah. And I was like, I'm just done with this. I'm not going to do it. Plus the fact that, and we've talked about this before, you get all sorts of other stuff. Like you. If you use VPN and you just do a Google search, Google's going to hit you with, hey, confirm you're human. It's like, oh, f you four times. Yeah, yeah. Gary [00:45:03]: And then other places I go to all the time are using the CDN stuff, you know, where it's Cloudflare. Right, right. And they hit you all the time with confirmed. You mean if you're on a VPN simply because they have a list of all of the VPN servers and they're saying, oh, you're on a VPN, let's be extra cautious because sometimes bad actors use VPNs to hide who they are. Here I'm trying to be like the secure person by using a VPN and you're wasting more of my time. People think, oh, it's a big deal. A Cloudflare box appeared on your screen. You click the little check mark button and you were through. Gary [00:45:45]: Yeah, Try doing that over edge network out while you're on a bus in the middle of Peru and you just want to check your email or whatever it is you know something you're doing, and it's not as fast as it appears. You hit a checkbox and you go. It's like a few minutes of like struggling with the web page. You just display the little thing I wanted to see. Leo [00:46:09]: So, yeah, so my vp, the VPN I use is protons. And since I'm in the proton ecosystem, it's just part of the package, so I've got it available. My VPN would not work when I was in the hospital. Of all the places. Gary [00:46:28]: Yeah. Leo [00:46:29]: And now I wasn't quite as time constrained as you were. In fact, I hadn't perhaps more time than I would have wished for, but I could not get that thing to work. And I ended up coming to the same conclusion. I just said at some point you Know what? Forget it. Turn off the vpn. Everything's HTTPs anyway. You know, I'm spending most of my time just writing my own blog or dealing with email. It's not a big deal. Leo [00:46:54]: But yeah, I, I get that it can be frustrating for sure. Gary [00:46:58]: Cool. Well, let me, let's finish on a up note, because this is the tech enthusiast hour. Let's be enthusiastic about something related to the travel. I once again, on this trip, I took my big camera, which is a Sony A7CR. The A7CR itself isn't particularly big, but the lens I took was pretty big and heavy and I took it with. And this trip, it was interesting because the group of people I was traveling with, I was the only one with a big camera. There were one or two other people that had small compact cameras and everybody else was on their phones all the time. And to be fair, I use my iPhone for plenty of pictures, but I took my big one. Gary [00:47:42]: Now, you know, the funny thing was, in our group and all around me, I was maybe, I think it was the oldest one in our group, hiking the Inca Trail, which is like a 9,000 foot elevation gain and a lot of distance at very high altitude. And my pack was maybe the heaviest of everybody because I was carrying this big lens with a camera the entire time in addition to my phone. And everybody just else just had their phone. So I. But I brought it with because it's like some of the most beautiful views you can possibly imagine on a hike or on the Inca Trail. You see amazing Inca cities the entire way that you have almost completely to yourself. And then you have the one that's better than all of those at the end, Machu Picchu. You know, I didn't want to get there four days and then take a picture of Machu Picchu with my iPhone. Gary [00:48:31]: I wanted to take it with my Sony using the, you know, Sony G class lens. That's what I wanted. So I carried this thing with me. And a lot of times I would use both cameras to take pictures of things and over and over again looking at them. When I was done, it was always the Sony, the big camera that won in so many ways. And I really tried to figure out like, okay, why, why exactly is it winning? And obviously, one way, one reason is because it's a big lens. I mean, the glass in it is like the weight of the iPhone itself, right? It's a big, big lens. The other is the sensor, right? It's a full frame sensor. Gary [00:49:13]: And yes, it's 60 megapixels, but that's not what was doing it. If it was a 20 megapixel full frame sensor, it still would have been really good. It's just a really big surface area with much bigger pixels than the 48 megapixel tiny sensor in the iPhone. And so it won for that a lot. And it won in the fact that whenever I would look at two photos that were about the same, oh, there's a McCaw, I'm going to take a picture of it, and then I'm going to take another picture with my iPhone, just so I have both and I look at them. And the bird was at the same size in the photo when I zoomed in. There's the muddiness you get with the small sensors when you zoom in, when you look, when you crop it to that. And that just wasn't there with the Sony. Gary [00:50:02]: Even though the pixel difference wasn't that that much. It was like the pixels themselves were much clearer as to what that color was supposed to be in that pixel when I really looked at it. So I ended up with a lot of photos that the cropping with the iPhone, even though it was a beautiful photo as a landscape, cropping it down just was too muddy to be useful. Like, I was like, okay, I'm not going to look at that ever again. It's just a muddy. But the Sony cropped it down. I was like, that still looks good. I mean, I could tell it's not super high resolution, but that's still an acceptable photo, you know, so that was good. Gary [00:50:40]: But also, you know, the viewfinder, it's underrated. You know, the Sony's got a viewfinder and I still use it on all the time with the viewfinder looking through it. Partially because of battery life too, because without using the big LCD screen, I'm not recharging the battery as often. But also there's just something about looking through that viewfinder and framing that picture, focusing on what's really there that I felt gave me better results each time. And in addition to that, just the mindfulness of it, the act of like, oh, look at this. Get the camera out, or even if it's around my neck, lift it up, switch it on. Which mode am I going to be in? You know, what am I going to do here? Let me look through it. Oh, do I need to adjust the exposure a little bit here because it's a little dark or it's a little light or whatever? What do I see in the viewfinder? What am I taking a picture of? Is there something moving like an animal in this. Gary [00:51:36]: When's the best time to actually press that button? There's something about that that I really just enjoyed the whole time, and I think gave me better results than simply holding my iPhone up and just tapping that little virtual button on the screen, you know? So I really like that. And. And as, like, one final note on that. I took amazing pictures and had amazing weather the entire trip, except for the one moment where it really mattered. At the end of the Inca Trail. The Inca Trail really ends with you coming to an amazing view of Machu Picchu. Leo [00:52:17]: Right. Gary [00:52:17]: And everybody has seen this view of Machu Picchu. Leo [00:52:21]: Sure. Gary [00:52:22]: It's a photo. Leo [00:52:23]: Reproducing it right now. Gary [00:52:25]: Yeah, Vision it right now. That's when it was raining. It was rainy and cloudy. And when we got to the viewpoint, it was kind of like, take everybody's word for it, because all I see is a wall of white. But there's no rush because these clouds move fast. In the mountains in the Andes, there just wait, wait. Let's see what happens. Every once in a while, the clouds would thin out and you could see the shadow. Gary [00:52:51]: You could see that picture. Your mind could fill in what was there, but it was still hazy. It was still like, you know, you could see parts, you could not see other parts, whatever. So you take a shot, okay, this might be as good as it gets. I don't know. And then it would cloud up again. And then five minutes later, oh, wait, it's lightening up again. You take a couple other shots, and it gets cloudy again. Gary [00:53:10]: And you keep doing that. And then after an hour, it's time to move on. And we never, never seem to get that view that everybody sees. Just bad luck, right? And that's fine. So went in, and by the time we got actually into Machu Picchu, the clouds had lifted. Beautiful views. Amazing stuff. Amazing selfies in Machu Picchu. Gary [00:53:33]: So it was all good. Then I get home and I look at those kind of hazy pictures that I took of Machu Picchu, and I look at them and say, well, you know, I did shoot raw. If I zoom in, I can actually see a lot of details. But if I zoom back out, it's all hazy. Let's see. Open up Photoshop. Photoshop Raw processing. Hey, there's a slider here that says dehazed. Gary [00:54:06]: That's what I want. And I slid that sucker all the way over, and wow. Boom. It looked amazing. Leo [00:54:14]: Cool. Gary [00:54:15]: I actually, since I use pixelmator Pro more than Photoshop, different controls, but there are actually Controls like that, clarity and all this other stuff. Different result, but very interesting. Like, oh, they look different, but they both are much better than what I saw. And it makes sense because, as I said, if I actually zoom in, I could see detail. It was zooming out that the white pixels, the ones that were reflecting off of the clouds, they were in the way. And dehaze and Clarity settings and pixelmator Pro and all were actually saying, oh, okay, we'll get rid of that, and we'll emphasize the detail that was behind it. And so I was able to save the shot of Machu Picchu because I didn't take it with just my iPhone. I didn't. Gary [00:55:02]: You know, I went. And I just shot it anyway, RAW with all the correct settings, as if I could see it well. And. And it's not the. The last time RAW has saved me. I shot my old Canon, the solar eclipse. In 19, 2017, I shot the Solar Eclipse, and I forgot to take off the filter during totality. Leo [00:55:26]: Oh, right. Gary [00:55:27]: And so the. The filter I was using before totality was still on it. So I had great shots up to totality. And then it said blackness. And because the filter was filtering out too much light. And the thing is, I just needed to take it off and just take a regular photo. But because I shot raw, the black there actually had the light there just very, very low. And I was able to save my Eclipse photo. Gary [00:55:52]: So shooting RAW with a nice big lens and a nice big sensor sometimes makes the whole thing worth it. The five pounds of that camera for four days carrying it, and I got myself a really nice photo that I just would not have. Leo [00:56:07]: The direction I thought you were headed with processing that photo was ever so slightly different. You've probably heard of a technique to generate HDR photos where you take three shots, one under, one over, and one exposed, and then you use the tool to basically layer them on top of each other. You know, adjusting exposures and so forth, you end up with a much richer photo. There are ways to simulate doing that, even if you haven't actually taken the shots one under, one over. And that just seems like another interesting technique to potentially have tried with your photo to see if it would have basically, you know, beefed it up a little. Gary [00:56:56]: Well, that was actually my intention. When I opened Photoshop up, I was like, let me see how I can do this. And then I saw the Dehaze, and I was blown away by the results. One of the problems with doing it, the layers, is the clouds are moving so fast. I mean, you could see them drifting. Leo [00:57:12]: Right. Gary [00:57:13]: So it was. So there would have been a lot of like, figuring out what parts of each photo to use. Lots of stuff. And I had planned on doing that and spending the time, but dehaze just worked. Leo [00:57:24]: You could. Sure. That's. Which is great. I mean, that's fantastic. You can do the HDR thing with a single photo. You basically take three copies of the same photo and then adjust the exposure on one to be under. On its layer, the other one to be over on its layer, and then you merge them all together. Leo [00:57:43]: But. And it gets, I mean, it's, it's not as ideal as actually having three separate shots. Gary [00:57:50]: Yeah. Leo [00:57:50]: But it actually does a pretty good job of, of, of getting you there. Very cool. Well, I'm looking forward to watching your video, listening to your video, and hopefully seeing some of these photos at some point. Yep. Let's see. So what's cool this week? This is one of those ain't it cools. That is not for everybody. Have you watched the Boys? Gary [00:58:18]: Everything except the new season. Leo [00:58:20]: Okay, well, we're in the new season. We're up to date, not quite done yet. I think there's like two or three episodes left. And like I said, it is definitely not for the faint of heart. There is much violence, much, much of everything. What I truly appreciate this season, though, are the subtle and not so subtle digs against current events. Things that are happening. You find, you know, he's making statement, you know, the, the, the, the antagonist in the show is making statements that map almost identically to things that are happening in the real world. Leo [00:58:59]: For the record, Daredevil, which would be my number two, ain't it cool for this week, if I were allowed to have to, would have been the same. They're doing their. They just finished their, their season and there were a number of situations where you're looking at what's going on and saying, oh yeah, they're making a statement about this that's happening in the real world, which I did. Again, I appreciate that kind of messaging within these shows and I enjoyed both shows just a lot. So anyway, the boys on prime and Daredevil I think is on Disney this year. Gary [00:59:36]: Cool. I will also feature something I haven't quite finished yet, but obviously I'm reading the book Apple the first 50 years. So Apple celebrated its 50 year anniversary on April 1st. One of the many things they did to celebrate that, in addition to having Paul McCartney play at the Apple campus, was a new book called Apple the First 50 Years by David Pogue. Who's a well known, you know, tech journalist and author. They, I guess I don't, I don't. I never looked up to see if they hired him to write it or he wrote it independently or there's some sort of thing. But Apple certainly pushed the book. Gary [01:00:15]: And I'm listening to the audiobook now and it's interesting because I've read so many books on the history of Apple. It's interesting to hear yet another one because the other ones, like I've read the Steve Jobs Walter Isaacson book, which goes into extreme detail, but it's all centered on Jobs. Of course I've read. I was written by Steve Wozniak, but of course that's centered on him. And I've read other Silicon Valley stories and all this stuff. So to hear this kind of told from the perspective of like a biography of the company is really interesting. It seems to move so fast because there's a lot to tell. Leo [01:00:57]: Yeah. Gary [01:00:58]: And so he moves over things that are like, there's, I know a lot more detail about this particular item than he seems to spend like three pages and then he's moving on. But the book is huge. So I guess at some point they have to stop this from being like 4050 years. Yeah. And there's some neat asides. Some of the best stuff is constant asides. So instead of skipping the little fun stories that aren't that important, he puts them as a sides, little sidebars where you, you know, he just talks about this little product that really wasn't worth mentioning, but it's kind of fun. And you hear about, you know, these little things or pranks that, you know, Steve Wozniak used to pull and stuff like that. Gary [01:01:37]: He puts them all in these neat asides, which are like almost the best, best parts of the book. And even though I'm doing an audiobook, I really am probably going to get the printed book just because it'll be fun to have all these little asides as, like, things I can easily flip through and find to hear these little stories over and over again. So anyway, so so far, yeah, recommend it. It's a lot of fun. Leo [01:02:03]: Let's see. Self promotion. How do I know you're not AI? The article that I was talking about at the beginning of this episode as we're recording this, it publishes tomorrow. So by the time you hear this, it will be out. It's askleo.com191783 Like I said, it's an interesting reflection of what you can and can't assume about the content you're going to be consuming over the course of the next 10 or 15 years. Gary [01:02:35]: Cool. I'll point to a video called 10 Weird macOS Features that Are Somewhat useful. I did a video like this years ago but found that like half of those weren't in macOS anymore. Leo [01:02:46]: Oh wow. Gary [01:02:47]: So I wanted to do it again, talk about the ones that were still there and then add some new ones to it. So I've got that. And also I want to add one last thing, right, there's one more thing as Steve Jobs, because you mentioned you were looking forward to seeing my photos and I from the trip and I thought, well Leo, you could have looked at them. I sent you the link. Leo [01:03:10]: No you didn't. Gary [01:03:12]: I didn't. I did though. Here's the thing, you're not on imessage. You use SMS or RCS more likely. So guess what happened when I sent the link out to you while in Peru? I only had an EIM that was data only, so it never got delivered. I looked back. Leo [01:03:33]: You didn't. Gary [01:03:34]: Yeah, I looked back and there it is not delivered. And this happened to another friend of mine as well. But it never occurred to me to go and say well maybe I should audit and look at who is getting SMS messages. I sent it to you again, but yeah, I wasn't able to. I didn't send it to you. And probably that other not delivered message circumseed yours and that's why I didn't see yours. So you never got the link. It never got delivered. Gary [01:03:58]: And that brings up a whole other thing of the sms. It's like what if I always think what if I need a two factor code while I'm gone? That's sms. Not only do I have no way of getting it, but it's not even like it goes somewhere else. It's not even like I could have called home if I wasn't able to get my email. I could have called my wife and said can you check my email for me and see if this important message arrived, there was nothing there. There's no SMS forwarding, right? At least not with AT&T. There's no SMS Vacation hold. There's nothing. Gary [01:04:32]: If you want to go and get an esim, whether it's data only or not, your SMS phone number is dead. Leo [01:04:40]: I have multiple articles on the various trials and tribulations of people dealing with two factor while traveling and my single biggest piece of advice is don't rely on sms. If you can use totp Slash Google authenticator compatible. Because that's not network. That doesn't require a network. Gary [01:05:00]: And much more secure, of course. Leo [01:05:02]: Among other things. Gary [01:05:03]: Yeah, among other things. But it does, you know, and one of the things I thought is at some point in the future, because I've been with AT&T too long anyway, it's, I mean, well, getting an Esim for 24 bucks for 10 gigs kind of reminds you how much more you're paying for less at home. And then you look and you say, well, E Sims for. Oh, you can get E Sims in the usa. I could be paying so much less. So the thing is, is that I've also talked about on the show how I've got a. For my old home landline number that's now through voip. And the cool thing is, is that I pay almost nothing for that. Leo [01:05:39]: Right. Gary [01:05:40]: It rings through to a VoIP line that I pay almost nothing for. But I also have so much control over that number now. Leo [01:05:49]: Do you get text messages, though? Gary [01:05:51]: I could, I could pay like. Yes. Leo [01:05:55]: The problem is that I've been down this path, as you can imagine. Gary [01:05:59]: Yeah. Leo [01:05:59]: And as it turns out, Google Voice, for example. Gary [01:06:04]: Yeah. Leo [01:06:05]: Which I think you mentioned you have. Gary [01:06:07]: Yeah. It's a separate phone number. Yeah. Leo [01:06:10]: That does text messaging. Great. Except it's not accepted as a text message destination for security purposes for many online services. Gary [01:06:22]: Interesting. Leo [01:06:23]: Well, I mean, explicitly disallow voip sms. Gary [01:06:28]: Interesting. I don't know what happens. All I know is that since it was a landline previously, I never used it for sms. So I didn't turn it on, even though it was like pennies. It would cost me pennies to do it. But it's like, well, I'm not going to use. I don't want another number that, you know, is being used for sms. But it did make me think that at some point I want to take my longtime mobile number, move it over to a VoIP service where now I have control over it. Leo [01:06:54]: Right. Gary [01:06:55]: And then I could do many things, including using many VoIP apps to actually get phone calls and presumably SMS messages of some kind wherever I want. And then maybe going back, even if I have a valid SMS number on my phone, making that more of a like, I don't give that number out kind of thing. It's for specific purposes. The number I give out is this one. And I have control and I could do a million different things. And when I travel, I could have it forward calls, I could have it forward sms. I could have it with a message that says to leave a message, press 1. To forward a message to me, press 2. Gary [01:07:34]: To reach this person instead, press 3. I mean, you know, it's incredible what you could do for pennies a month you can't do for, like, $100 a month. Leo [01:07:45]: You could have it forward to your Peruvian voice number. Gary [01:07:48]: Well, yeah, I could. I could have paid for a real sms, real esim with a number and forwarded all of that to it easily. Leo [01:07:57]: Fun times. Gary [01:07:57]: Yeah. Anyway, anyway, that's maybe something we'll talk about again at some other point. Leo [01:08:05]: Well, very good. Yes. After that. That pretty much does wrap us up for this. As always, thank you, everyone for listening, and we will see you here again real soon. Take care. Bye. Bye. Gary [01:08:19]: Bye.