Who Let the Blogs Out? - DTNS 4911  ## Metadata - Author: [[Daily Tech News Show]] - Full Title: Who Let the Blogs Out? - DTNS 4911 - Category: #podcasts - URL: https://share.snipd.com/episode/614f31fa-43f2-4528-909b-23a2c0bc182b ## Document Notes ## Highlights - **Quantum Computing Breakthrough** - Google's quantum computer, Willow, surpasses the QEC threshold, reducing error correction needs. - This breakthrough may lead to practical business applications by the end of the decade. Transcript: Tom Merritt Exclusions apply. Details at DSW.com. Google announced that its 105-qubit Willow quantum chip breaks the QEC threshold, the quantum error correction threshold, meaning the need for error correction reduces as the number Of qubits increase. It can solve a benchmark algorithm problem in five minutes that would take the frontier supercomputer, which up until recently was the fastest, but still currently the second fastest Supercomputer on the planet. It would take that frontier supercomputer 10 septillion years to solve. Nate, I don't think I will be around to see if a classical computer were to finish that. Nate Lanxon No, I doubt I will be either. But I think the significant thing here is that this was stable for like an hour and a lot of these qubits they are not stable for a very very long time that's that seems to be the big breakthrough Here like this theoretical total processing power is one thing but the stability of the qubit itself is actually pretty good i see our technique that leads on this as well. Tom Merritt The 105 qubits basically make one stable qubit, right? Because they all error correct for each other. Nate Lanxon Yeah, which is impressive. But also, there is no application for this right now. This is all pure theory. You know, it's a big number. It's a bit of a PR thing, but it is genuinely, I think, from what I've read about this, a step forward and a significant one. Tom Merritt Yeah, there are some practical uses, even of the previous version, Sycamore, that did not have this level of error correction. But the problem was those practical uses could also be done by classical computers. They might be done a little faster by the quantum computer, but you really didn't need the quantum computer to do them. So it's one of those things where you're like, well, the quantum computer isn't useless. It can do things, but it can't necessarily do things better yet. This passing the QEC threshold means that Google thinks the end of the decade, it can actually have a computer that would be able to be used in a way that would be perfectly error corrected And therefore far outpace. It would be fault tolerant. So it would far outpace what classical computers could do. And they even think they could probably get it to do a benchmark that does useful output by the end of next year. Nate Lanxon I mean, by the end of the decade, I mean, that's really not that far away. That is a depressingly close milestone in terms of being the end of a decade. Five years. I mean, I could see it, I suppose, but that's pretty incredible. Tom Merritt What it tells me is that we need to actually really start preparing for a world with quantum computers. Quantum computers are not good at the same things as classical computers. So we're not talking about the fact like, oh, I just need to replace all my computers with quantum computers. Quantum computers are good at very specific things like, for instance, cryptography. And so, you know, NIST, the National Institutes of Standards and Technology here in the US, put out quantum-resistant encryption standards recently because anything that is encrypted With current encryption standards that aren't quantum resistant would be able to be broken easily once quantum computers become available. Nate Lanxon Yeah. And I think, I mean, at the moment, most of our security methods are based on being prohibitively costly or energy intensive in order to crack in a reasonable amount or useful amount Of time. And so a lot of the theory here is, well, if you remove the limit of computing power, then suddenly everything we've got is basically very crackable, in very layman's terms. As I know you like to point out, there'll be people going, ah, but, but, but, but, and I'm sure there are those people. But that is like the very basic version of the situation. So it's also very hard to run these things, and they're going to be very, very expensive for a very, very long time. So there's kind of a natural inhibition that a lot of these technologies are going to have, which is that, yes, they could technically maybe do this, but then who's got them? And how did they get them? In fact, that's going to be a lot further away. Tom Merritt That was one of the things that Google quantum AI was talking about was they actually, because they reached this quantum error correction threshold, think it's finally worth thinking About finance. It's finally worth thinking about, you know, practical costs and bringing these to a position where businesses could afford them. That certainly isn't going to make them cheap for all of us to use, but it could make it so that businesses could be able to take advantage of these. Up until now, it's kind of been scientific and academic endeavors that have taken any advantage of what quantum computers can do, simply because it wasn't cost effective. So it's more a handshake agreement of like, we're going to let you use this to prove what it's good at, which is good for Google. But they're getting to the point where they think, oh, we might actually be able to create a small business around this. And by small, I don't mean a small business. I mean, a business that doesn't generate a whole lot of revenue. Nate Lanxon Yeah, which is a proof of concept in and of itself in a way but if it is at that point that is a pretty remarkable step forward for the technology that i think even in our lifetimes you know End of the decade stuff notwithstanding even in our lifetimes we will look back on with the same way we look back on the early era of blogging slight prediction of what we're going to talk About later exactly it's like wow that was so old and so archaic. And would we have imagined then what we have now? No, we'd have thought it was ridiculous and very, very hard to imagine. So never bet against the future. Tom Merritt Yeah. No, no, no. Very, you know, I often think like in the 60s, a classical computer required you to be a huge business and get a mainframe and maybe even rent the mainframe. 20 years later, people had them on their desktops. 20 years after that, people carry them around in their pockets. And the phones we carry around our pockets vastly outperform what you had to put in a mainframe back in the 60s. So, yeah, we're at that stage of the mainframe with the quantum computer, maybe even the 50s, not the 60s era. ([Time 0:11:59](https://share.snipd.com/snip/2731ad8f-dece-49c6-857a-5dca269b514a)) - **Reddit Answers** - Reddit Answers uses a generative model to summarize relevant Reddit threads, helping users find information. - It focuses on summarizing discussions within Reddit, unlike broader chatbots. Transcript: Tom Merritt Reddit is testing Reddit Answers, a chatbot that lets you ask questions and then uses a generative model to return a summary of relevant threads that address your question. But it's just Reddit. It's saying, based on what people on Reddit are saying, here is a summary of an answer to your question. It's not even trying to be factual. It's saying, this is what folks on Reddit say. Below that summary are quotes from actual user posts. So you can see like, oh, how did they actually phrase it? What did they actually say? As well as links to the full conversations that the chatbot used in its answer. So it's citing where it got it from. To me, this is better than using Google to search Reddit because it's just on Reddit. So it's meant to be the best at finding things on Reddit. Although it's a little more narrow than asking Claude or ChatGPT because it can't go find information elsewhere. Features rolling out to a limited number of US users to start and then will expand from there. Uh, I know a lot of people just go to Google and say Reddit colon blah, blah, blah, blah, blah, blah, blah, uh, or, or format it correctly, but, just search Reddit, Google, I only want to Find answers there. Do you ever do that, Nate? Yeah, absolutely. Nate Lanxon It's usually the Reddit is the suffix on the query, you know, how to stop a something Reddit. That is usually a thing. Usually it's my wife, Kate, that finds most of these things, you know, how to fix the boiler. It's usually her. But I mean, this seems like a logical step forward for Reddit. I did go back and look at their original press release announcing this, and they did characterize it as an evolution of the search technology, which actually is what this feels much More like. You know, I think comparing it to ChatGPT, Claude, any of these perplexity, I suppose. It's not really in that same league, but it's also not trying to be. It's trying to be a quite narrow but good tool, which, if anything, is kind of what a lot of the best, or rather what we think some of the best uses for AI and AI agents in future are going to Be. Very narrowly focused, specialized domain expertise, it gets called. And in this case, the domain expertise is Reddit and Reddit's users. So it makes sense, but I would struggle to call it a particularly smart bot, he says, having not actually used it, just read two stories in a press release, but you know what I'm saying? Tom Merritt No, I do. And I think that's fair, where your expectations may be like, well, the Reddit chatbot didn't know the answer or the Reddit chatbot, you know, got this part wrong. But where that if this is implemented the way it looks like it's implemented, like you say, I haven't tried it yet either because it hasn't rolled out to me. But it is just telling you what Reddit says and Reddit can be wrong. But I think it's easier to wrap your head around that than ChatGPT, which just sounds like it knows everything when all it knows is what the internet was trained on and the internet can Be wrong. I think in the context of Reddit, we're more used to that. We're more used to saying like, well, I am on Reddit, so I will always keep that in mind. But there's some good advice on here and some good facts. Nate Lanxon That's very true. I think part of it is an interface thing as well. Searching Reddit on Reddit is not a particularly fun experience. Searching Reddit from Google is often an extremely good way of finding answers on Reddit. So I wonder if this is sort of somewhere to serve the people who want something in a little bit of both. You know, they're hopping mad. They want something in the middle, to quote Alan Partridge. So that's a niche reference for American audiences. I apologize. But I think that that is the reality here is that it's a means to an end and a very good means. Yeah. Tom Merritt I don't think I saw a few articles out there, you know, comparing it as a competitor. I don't think it's a competitor to any of these other chatbots. I think any more so than Reddit's previous site search was a competitor to Google. It wasn't. It was a way for you to be like, oh, I'm on Reddit. Let me find something on Reddit versus trying to find something on the entire web. Now, Reddit has become somewhat of a substitute for what's on the entire web, right? But that's not what Reddit search was designed for. It was just designed to make you search Reddit. And I think, you know, to go back to what you said earlier, Nate, that's what this is for. This is for helping you find something on Reddit, but it's a little more effective than just a search because it can summarize, which I think might send you into a thread more often than If it wasn't summarized. Because a lot of times I look at threads and I'm like, eh, that one might be good, but I don't have time to investigate. If I've got a good summary, that may convince me, okay, I want to go into the thread and see the nuances of what people said about this. ([Time 0:18:46](https://share.snipd.com/snip/af179738-2122-421e-bbe7-d944fabdbda7)) - **Evolution of Blogging** - Merriam-Webster declared "blog" its word of the year 20 years ago, but blogging has evolved significantly. - Nate Lanxon reflects on the shift from personal online journals to modern platforms like Substack and vlogs. Transcript: Tom Merritt 20 years ago this month, Merriam-Webster, not a person, but the company that makes the dictionary, made blog its word of the year. They defined it as online journal where the writer presents a record of activities, thoughts, or beliefs. Now, since that time, blogs have changed a lot. It's hard to tell what's a blog and what's just a website, what's just a newsletter. Nate has been looking into this without a little bit of a rabbit hole that brought it to your attention. So first of all, tell us how you came upon realizing that we're at the 20th anniversary of this. Nate Lanxon Yeah, I was reading up on some tech history for my podcast after I turned it from being a news show to a nerd nostalgia, some sort of thing. And I'd spotted this old blog reporting that the word blog was Merriam-Webster's Dictionary's Word of the Year for 2004. That's 20 years ago. It made me feel pretty old. I was already feeling quite old because I'm 40 in six days. So I went digging a little bit further down that rabbit hole you mentioned, there was then a New York magazine story, I was going to say New York Times, it was a New York magazine story that Accredited the first actual blog as being launched or written in 1994 by this guy called Justin. His page is still online, actually. I looked it up. It's still there. It's pretty dull. So I wondered if there was a more interesting first blog. And I kept on searching. And then I found the Encyclopedia Britannica's assertion that the first blog was actually even earlier than the one in 1994. It was in 92. But that was written by Tim Berners-Lee, who I sort of thought he doesn't get the right to have the first blog he invented the whole thing like let's not count that um so there's a bit of Ambiguity there um but what he did get me thinking was kind of how if you if you did if you did take 2004 the year it was word of the year as being at least that kind of we considered things to Be peak blog at least by a dictionary how different was blogging then to what it is now which is what started getting me interested in all this because it seems like kind of an alien concept That's been um i suppose usurped in a way with other platforms you mentioned newsletters um or substacks uh just now because you're writing or you were writing something that is primarily Words. Maybe you have a few pictures. You're manually adding tags and links and then publishing on the, excuse me, on the hope that someone's going to subscribe to it or is already subscribed. Get an email notification or navigate to it from one of their web browser bookmarks or something, which is just not the case anymore for the vast majority of people. I'm actually still a big RSS user. I know you are as well, Tom. And I looked through my feed list. I'm just looking at it again now here. There are only two subscriptions that I'd say are blogs, as in they are written by a person or by a person for a company that is something that person is is doing or um or looking at or sitting On or eating or something and the rest are basically descendants of what blogs um what what they were so that's things like i would say engadget or gizmodo like they did start out as blogs And they've become, you know, huge enterprises in their own right. But they're basically just news sites that still have RSS and they might be based on a blogging platform like WordPress or something. So it's like bloggers in 2004, they are now, I think, as you said, they are writing sub stacks or they're monologuing into YouTube or I suppose monologuing into a podcast or doing lengthy Threads on X or threads, I suppose. And I guess, you know, I want to say is blogging dead? I learned from years writing for Wired that it's never a smart thing to proclaim that something is dead. And I'm quite sure there are still millions of people writing travel blogs and personal accounts of things that are important to them. But I think the motivation and the reason for doing so, if you're going to do it as a blog, it has to be much more profound than it ever was in the past because we've now been conditioned to Think that the only way something is valuable is if it's reaching the largest number of people, which has to be social media in some form. You can't just assume that someone's going to subscribe to your RSS feed or bookmark your homepage, although I would and do. So that's kind of the background to this. ([Time 0:25:20](https://share.snipd.com/snip/81a49053-c547-4d53-b410-e1d7f19ae026)) - **Vlogs: The New Blogs?** - While traditional blogs may seem less relevant, vlogs have emerged as a spiritual successor. - Influencers often fulfill the original blogging purpose of sharing personal experiences and thoughts. Transcript: Tom Merritt Yeah, yeah. It really is interesting because by this Webster definition, Engadget is not a blog. Google's blog is not a blog. Microsoft's blog is not a blog. They are not a journal where the writer presents a record of activities, thoughts, or beliefs. There are still a few out there. The one that came to mind immediately was John Scalzi's Whatever blog, where he very much just talks about daily stuff, some of which is his writing activities because he's an author, But some of which is just his thoughts on whatever's happening that day or what his daughter or his wife is doing. And that is a blog. But I think blog came to mean sort of a journal of updates, in which case I would say that the Google blogs and the Microsoft blogs count as blogs, they're corporate blogs, but they are Different than just a list of press releases, right? Like they have a more personal tone. They are updates on what's happening in that company or that division of the company. But really, as you were talking, what I realized is the spiritual successor of blogs is the vlog, Where the idea of blogging really survives the strongest, I think, is the vlog where You have very successful YouTube series that's just a person filming their everyday life. Here's where I went. We call them influencers now, but if they were writing this stuff back in the day, they probably would have been called bloggers. Nate Lanxon It's funny. I was thinking about this. I'm sorry for the TMI, but I was thinking about this in the shower earlier on. I thought, you know, back in the day, if you worked for a large company and somebody called you a blogger, you would correct them and say, no, sir. No, I'm a journalist. I write stories. You want a blogger who does posts. They're over there in the corner. And, you know, it's very self-righteous, but a lot of people did it, myself included. And I think now there is this kind of weird stigmatism as well, where you don't want to be called an influencer. You might be one, but some people wouldn't want to. And then the people that really do want to be called an influencer, a lot of people judge them for being vacuous and just wanting to be famous, which is a shame because both back then with Blogging and today, influencers are both legitimate forms of, well, I don't say reporting as an influencer, but they are legitimate things that some people are very, very good at, Make good money from and have a place in the world. Um, but I think there's stigmatism that comes with the word that has definitely changed over the years. Um, and like right now, if someone said you are a blogger, I mean, I'm not, but if someone said I was, I'd be like, okay, whatever you're, you know, you stink. Tom Merritt It really doesn't matter anymore. Yeah. You need a shower where you will have a revelation on the meaning of blogs. Trust me. Um, yeah. Yeah. I, I, I think back now that you mentioned that, uh, when Engadget launched, it called itself a blog. So it was clearly not a, not a, uh, an insult. I remember a point in CNET where it was referred to, I think by Molly, but I don't know for sure as a blog and they got upset as if we were you know at cnet uh dismissing them as just a blog uh And nowadays if you call and gadget a blog i think they'd just be confused and be like well no not really like i guess we used to be you know but it's just a news site now um and i think blog has Come to mean a series of semi-regular daily posts, right? Like maybe they're not every day, but they're a post of the day, possibly multiple times a day. But the blog at a company is, here's the record of everything that's happening in this part of the company or smaller companies, the company itself. Daily Tech News Show uses WordPress to post its website and it's in blog format because, you know, we post our episodes daily, but I don't consider it a blog. A blog is more of a diary or journal of some sort that's documenting something. I think that's where I settle on it now. That's what a blog is. It's a tool. It's a piece. It can be its own thing or it can be a section of another thing. Nate Lanxon And it's something that I'm now proactively seeking out. And there's one guy that I used to follow on Instagram who has a blog called Tokyo Times, which if you like street photography from Tokyo, it's worth pulling up. And just subscribe to his blog as a blog. And it feels quite nice. You know, there's very little evil in a world when you're reading a blog in an RSS reader. It's a bit of escapism, which was never blogging's goal. And no one had a goal really for blogging. But for me, subscribing to some now is a bit of a respite from the busyness of the world. Tom Merritt Well, I think we'd be remiss if we didn't touch on the fact that a lot of what made blogging in its earliest days became what people did on Facebook, what people did on Twitter, you know, And was diffused into social media. And so some of it has come back, like we've touched on a couple of times now in the form of Substack. But I think there's still a place for that good old fashioned personal blog. So yeah, maybe it's poised to make a comeback. ([Time 0:30:00](https://share.snipd.com/snip/985362e8-0bd2-4476-887f-e4dfae3dba70))