605: Jim Nielsen on Subversive URLs, Blogging + AI, and Design Engineers
Jim Nielsen joins us to about URLs and linking as the new subversive way to maintain the web, paying for news in Canada, should content creators be worried about AI, the case for design engineers, RSS in HTML, and the state of state and UI.
Time Jump Links
Links
- About - Jim Nielsen’s Blog
- “Wherever you get your podcasts” is a radical statement - Anil Dash
- The Subversive Hyperlink - Jim Nielsen’s Blog
- More Files Please - Jim Nielsen’s Blog
- Canada to keep pressure on Facebook to pay for news, Trudeau says
- Cite Your Sources, AI - Jim Nielsen’s Blog
- For the first few decades of the web, the tacit agreement wa...
- The Case For Design Engineers, Pt. II - Jim Nielsen’s Blog
- RSS in HTML: A Follow-Up - Jim Nielsen’s Blog
- Style your RSS feed
- UI is a Function of Your Organization - Jim Nielsen’s Blog
- UI daverupert.com
- Notes from “Why Can’t We Make Simple Software?” By Peter van Hardenberg - Jim Nielsen’s Blog
- I Staked Out My Local Domino’s to See Just How Accurate Its Pizza Tracker Is
- The Benevolent Deception: When Should a Doctor Lie to Patients? - The Atlantic
Episode Sponsors 🧡
Elicit (23:41)
Elicit’s goal is to radically increase high-quality reasoning in science and beyond. As early as 2017, they pioneered process supervision, an approach to breaking down complex work for advanced machine learning systems, so that it remains transparent and controllable. Today they use language models to help more than 200,000 researchers each month. They just raised a $9 million seed round and are looking for exceptional engineers across frontend, backend, and ML. If you're an exceptional front-end engineer looking to build the next generation of AI interfaces with a modern tech stack (Next, Tailwind, Chakra), join them!
Transcript
[Banjo music]
MANTRA: Just Build Websites!
Dave Rupert: Hey there, Shop-o-maniacs. You're listening to another episode of the ShopTalk Show, a podcast all about front-end Web design and development. I'm Dave Rupert and with me is Chris Coyier. Hey, Chris!
Chris Coyier: Hey!
Dave: Hey.
Chris: Thanks for having me, Dave.
Dave: Oh, thank you.
Chris: Always a pleasure. ShopTalk Show.
Dave: Thank you for paying for this Riverside FM. [Laughter]
Chris: Sure.
Dave: So, Chris, who do we have in the studio today?
Chris: Yeah, I forget who pays for this.
Jim Nielsen is here! What's up, Jim?
Jim Nielsen: Hello, everyone.
Chris: A friend of the show. Past member, I believe.
Dave: Alumni and, yeah, you were in the D-d-d-d-discord for a while.
Jim: A long-time listener, Discord member, yeah.
Chris: Yeah. Right on. Yeah. Jim, if you don't know him, is a bit of a thinker on the Web. Really one of those must subscribes; one of those good RSS feeds. Although, Jim, I've got to tell you. Intervention here. You don't make it easy. You know?
You go to JimNielsen.com. There's no link to the blog. You've got to just know it's a subdomain. You know? Then Jim has this incredible blog. He has the sync blog and then he has, like, what I'm reading with thoughts that are just equally great blogs. You should subscribe. Really, I'm not blowing smoke. They're good.
But you can't find... You have to go to the About page to find the reading blog. You've got to just smash them together or something. I mean you do you, but it's too hard to find.
Anyway, subscribe to both. You should, everybody. Great stuff. And I thought we'd just pick out a few favorites.
Jim, you're quick to jump on the, like, let's blog back and forth thing. Old school style. Got to love that. That's the most fun.
Jim: That's my whole content strategy, to be honest. I'm just echoing what I read from other people.
Chris: [Laughter]
Dave: It's working. It's working.
Chris: But you kind of make connections in an interesting way and then call out, usually, kind of the juiciest bits. And that's not easy to do, so cheers.
Are you ready? Do you have any news to share? How's life? How are the kids?
Jim: Good. You know it's Leap Day, Leap Day of Leap Year today.
Chris: Yeah. How's work going?
Jim: And I couldn't be happier than to just spend my extra... What better way to spend the extra day that you get than to just be on the ShopTalk Show and talk about Web stuff?
Chris: Yeah, indeed. Well, thanks. Why don't we just jump into some topics? The people out there, they want to know.
I love this one recently. We'll just kind of go in reverse chronological order, like any good blog would. I'm sure you saw recently, Anil has got a gift of putting his finger on hot-button stuff. He had a great one recently about the subversiveness of podcasts. Dave and I mentioned it on a podcast not long ago how when a podcaster, which a lot of them very naturally do - we do - say, "Listen to us in your podcatcher of choice." That's a symptom of - what did he call it - a radical choice or a subversive thing or something. God, I'm sorry, Anil. I read the post. I just can't remember your exact choice of words.
But it's cool, and he was calling out open technology for doing that and how kind of rare it is these days. And it's the fact that podcasts are this RSS-based thing. Although, Shopify would do what they can to destroy that if they could. But that because they are, it's this open technology that anybody can do. And that's just kind of this amazing thing.
I don't know if you pointed to that necessarily, but your post was that even just links are subversive, in a way, and that so many companies try to push it down - in a way. Like famously Instagram is like only one link in your bio and no links in the descriptions. And Apple is like, you can link to stuff sometimes but not if it's payment information or if that link has any information about how you would pay elsewhere or something. It's just funny that way.
Do you remember? Can you expand on your thoughts on how just the humble HTML link is subversive?
Jim: I mean I think I had another post maybe a year or two ago that was kind of on the same idea. I just thought it was interesting how Apple announced they would finally allow developers to link outside of the app, and that was a big no-no. You could just get your app removed from the app store if you did that.
It was just basically like all you're doing is adding a link. It's such a common thing that you can do for free any time you want to anywhere that you want if you have a website. But if you have an app, we're talking about million-dollar lawsuits that are going on about whether you can add a single link in an app that goes to your website, which just kind of sounds crazy, and I think really contrast the differences between an app in the app store and a website on the Web and the differences between the two.
And to sort of go into a little bit more, have that lead into what Anil was saying, I think I wrote another post, too, that was on the same idea. I think it's interesting that if you follow the money, people are actually paying money to put advertisements in podcasts where they're not actually saying anything like, "Go to the app store and download our app." They're saying, "Find us in your podcatcher of choice." They're not pointing to any sort of centralized entity. They're literally paying money to say that, which I think is a win for the Web because if you can get people to be spending money to direct people to know where in particular, like just a show, not necessarily--
Chris: All you have to say is, "Hey, we've got one of those things. Go find it."
Jim: Yeah. You don't have to go to Spotify and download their app to then get access to our content. It's just wherever you find it.
Chris: It would be a worse ad, it would be a less effective ad if they said, "Go to Spotify and find us," because then somebody that has Apple Podcast, they might be like, "Oh, bummer. I don't use Spotify." But that's not what they meant. They didn't mean "Go to Spotify." They meant, "Just use the one you already got," and we don't know which one you've already got, so we've just got to say it that way. Like they were almost strong-armed into not saying which platform to use.
Jim: Yeah, exactly. And wouldn't it be amazing if you could extrapolate that out into the future where people could say that about anything? If you're a YouTube influencer, you wouldn't have to say, "Oh, go to my YouTube." You could just say, "Follow me in your content consumption app of choice," and that could be following a blog, following a YouTube video, following music, following a podcast. I think there could be a lot of potential in this space if we could figure out how, and I think podcasts are leading in this area. But it would be really cool if we could figure out how to do that for blogs and video and audio and anything else anybody wants to make.
Chris: You know it reminds of those, like, you ever see an ad that's just for, like, pork? "Delicious, white pork." [Laughter] You're like, "Who paid for that?" How is that--?
Dave: It doesn't matter.
Chris: [Laughter]
Dave: In your pork catcher of choice.
[Laughter]
Jim: It's like the old milk commercials, like, "Got milk?"
Chris: Yeah.
Jim: Whose milk? Whose milk are we talking here?
Chris: Yeah. They definitely mean not nut milk. But you know.
Jim: [Laughter] Or almond milk, yeah.
Chris: Podcasts are leading, I agree, but are built on the foundation of RSS. Obviously, much less popular phenomenon, but still share the same characteristics such as you can just provide a feed and then people can play with it in all sorts of other contexts. They can read it through a reader, and they can build an app that sucks it down and uses it however they want to. It exemplifies the same type of cool freedoms.
I guess it was last week's show, Dave. I can't even remember. We kind of talked about how Git is similar in that it's just this protocol. It's just this thing that you can use, and a lot of people use GitHub because it offers some nice things. But you don't have to. You can use any number of other Git providers or nothing or just use Git between your computers and stuff. These kind of protocol-level things are subversive in that way that you're not trapped to any one particular thing. It shares those same characteristics that are cool.
Jim: To follow on another blog post, I had another one recently that was about files, which is kind of the same thing. Files are amazing because you can have any third-party sort of manipulate and work on your data. Right? If you have a text file, you can open it in Sublime and Visual Studio. Whatever text editor of choice you want to open it in, you can open it. Right?
But if you have a Notion doc, that's Notion only. And if Notion one day gets bought or who knows what -- and you never know, obviously, what can happen -- who knows what happens to your data? There's no Notion file.
Chris: Right. That's a row in their database somewhere that's not your file. They support exporting of sorts, you know. Yeah, that's tricky. I do like that, though.
Isn't it funny to compare Sketch versus Figma then? It's like Figma may have won because they don't have the .sketch file sitting there because that then became users had to be like, "Well, what are we going to do now, keep it in a Dropbox? Do we need Google Drive for our team, too, to keep those files in?" Figma is like, "Nah, don't even worry about that."
Not that I'm defending that, necessarily. But I do kind of see the advantage of it and see why it was probably ultimately a competitive advantage for them.
Dave: One feature I wanted or still do but for Luro, we collect a lot of data. Lighthouse runs. Blah-blah-blah. I think it's valuable to maybe offer a CSV export or something for all these tables of data we're building.
I feel like, to use Luro in your workflow, you have to come to Luro and look at this data and maybe copy-paste it over. But it'd be so easy to insert ourselves into somebody else's workflow or somebody's existing workflow if we just exported a CSV and then we didn't have to build an integration to whatever app, Coda or Excel, online 365 Business Edition. We just let you download a file, and then you have it. Now you do whatever you want with it.
It's going to go stale in a week, but good luck. Have it. I want you to have that. It seems like a nice thing to do. I don't know.
The thing about hyperlinks that gets me... You know your boy, Dave, loves The Verge, and I go to The Verge. When I read an article on The Verge, it could be about, like, Apple's new iPad, or the Volvo XC 90 electric is coming out with 24 florps, or whatever. And they don't link me to the product. That just drives me out... like, just drives me up a wall because you're telling me about the thing. You kind of sold me on the Apple Vision Pro. You could drop an affiliate link and I might go buy it.
Why don't you link out? You just really want me to stay in your little content silo.
Chris: Do they do that, really? Oh, no. That's obnoxious. Like the New York Times definitely does that, right?
Dave: Yeah, New York Times is a great example. They just don't link you to what they're talking about.
Chris: Yeah.
Dave: They just kind of--
Chris: When it's like that grade-A journalism, you almost kind of get it because the nature of SEO and how it influences the Web and stuff might influence the journalism in a way. I sort of get it.
I guess The Verge is journalism, but anyway. What do you think, Jim?
Jim: I was just going to say that is a tricky one because the SEO part of it. It's like you don't want to lend credibility necessarily sometimes to some of the things you're talking about. But by linking to them, you're sort of lending this credibility of, like, "Oh, the New York Times linked to me." People would pay good money to have the New York Times link to their website. But if you're doing something shady and the New York Times writes about you, it's kind of like, "Well, we don't want to link to you," because of whatever the technical underpinnings of a search engine and stuff are. All of a sudden, the New York Times is linking to you, and so it's like, "Oh, well, they must be credible, the New York Times is linking to them."
Dave: Yeah. Yeah, I can see... Yeah. Well... but is that a problem with the Internet or search engine monopolies? [Laughter]
Chris: Yeah. I mean the policy should just be rel no rel. Isn't there a rel attribute that solves this problem?
Dave: Yeah.
Chris: No refer. Not no refer, but it's kind of like, "Don't give any credit to this link. It's just a functional link only." I forget what the value is.
Dave: I'll find it. I have it somewhere on my site.
Chris: It should be a CMS decision kind of thing. But yeah, to me, The Verge sits squarely.
Dave: No follow.
Chris: No follow. Yeah.
Dave: Rel no follow.
Chris: The Verge is squarely in the, like, "Y'all need to be linking to stuff. Y'all need to be hot-linking a lot." [Laughter]
Jim: For sure. That's a pet peeve because I do it all the time, too, Dave. I'm reading the article, and I expect a link to be right there on those words I just read, like XC 90 with 4 florps, or whatever you called it.
I can't click right there, and so then I highlight the text, copy, paste it into Google, and try to find the link that refers to what they were talking about.
Dave: Yeah. It's just... Man.
Chris: There's that, and you know what's even worse, though, is when they do link it and then it's just to the tag of that thing on their own site.
Dave: With the homepage. Yeah, just to the homepage of Volvo. It's like, "We tried."
Chris: Yeah.
Dave: Yeah. Yeah.
Chris: They just internally link it or make it some hover thing that shows an ad.
Chris: Did you know? How top of mind is this? Did you know that there is some Canadian law that if you link out to a news site, you have to pay the news site to do that?
Dave: Oh, really?
Chris: For example, if you search some news, and Google search engine pages link to the news, Google agreed to pay 100,000 million Canadian to those news organizations just for the right to link to them.
Meta was in the same bucket (right) with Facebook. And so, instead of paying anything, Facebook just says, "Nah, we'll just block all news." So, if you're in Canada, there's just no news on Facebook because Facebook is like, "No, I'm not going to pay you to link to you. You should pay me."
It's like a reversal of expectation, you know. It just doesn't make any sense at all. And I'm not trying to go to war for Meta because screw Meta, but I agree with them on this. I can't be paying you to link to you. Uh-uh.
Jim: No, I was just going to say that's... I have not heard that. That's pretty wild. I don't even know how you--
Chris: It's relatively new. I'm going to drop--
Jim: How do you police that?
Dave: Well, that was an early Web debate a long time, like Hotwired or whatever. This was debated during the Hotwired era, just like, "Oh, if you're going to link to me, you need to pay."
But I wonder if this is specifically sort of about, like, if you're going to be stealing news and surfacing it as a product, like inside your application, you need to pay news outlets that are informing your product - or whatever. Sort of like how you'd pay the AP to get a feed of AP news (right) or something like that.
Jim: Yeah, it's funny because I feel like... I think you guys talked about this on a recent episode, too. It kind of ties into the whole "let AI browse the Web for me" thing where... I don't know. I really like following links down the Internet. That's... I'd much rather just follow links around the internet than just have a feed that's an algorithmic-driven feed that tells me what I think I would like.
I'd rather just sort of old-school '90s Web, you end up on something like Wikipedia and you're just opening all these tabs and all these links. "That looks interesting. That looks interesting." And you go down this rabbit hole of links. And that's what makes browsing the Internet fun, to me.
And so, this idea of, like, let a robot tell me what this page says takes all the fun out of the Internet. I don't know. That's just me. But it kind of comes back to the linking, again. I really like following links, and that's the fun part of using the Web.
Dave: Yeah. I should say we got a listener submitted a question or a comment, I guess, from Alec Gregory.
Chris: Gregory.
Dave: Was saying, like, "I've been following the reaction to Arc Search. I was a little taken aback to hear it be negative." And he's just saying he loves it. It's useful.
Then he has two, three examples here. One to just do a little quiz with his kids at the table, you know, like, "How far is San Francisco from Bristol?" And didn't have to pull out the map and do directions. [Laughter] It just told you, and that was a pretty useful, quick way to do it.
Watched Poker Face and wanted to know if it'd been renewed for a second season. You Google, and you get all this junk, like whatever.
Chris: Spoilers.
Dave: Spoilers.
Chris: Spam crap, yeah.
Dave: Actor interviews, crap like that. And then Arc was like, "Yes, it has." Done. You know? That's cool.
Then the last one was a footbridge in the community they live in. it's like now closed, and there's a saying, like, "This hasn't been fixed in five years." There's a banner hung up, so he just was like, "What is happening with this bridge?" Then it came back, like, it's not just angry next-door posts. It's being worked on and here's a history of the bridge and why it was broken and stuff like that.
Anyway, those are three anecdotes. I don't know. Chris, what do you think?
Chris: I just need to add that the third one, he goes on to say that then, with the footbridge in his neighborhood, which is an insanely hard thing to search on the Internet. What are the chances of the bridge in your neighborhood coming up?
It did. It found information on it. It gave him the information that he wanted. And then it cited a source of the website where it found that information. He clicked the link to the source and then went and learned more and figured out how they got that information. That's why he's saying, "Oh, it actually made me browse the Web maybe more than I would have." Eh. Maybe questionable there. But the source thing is very interesting to me.
So, now all of a sudden, it's AI-driven, but it's saying where it got the information clearly. I see little examples of this here and there where AI is happy to show you the source where it got information. And sometimes where it's just more than happy to totally wipe that information away.
To me, it's very night and day how different those things are. If you are saying where you got the information and linking to it, I'm all like, "Woo! You did it! Good job, nerds. You figured out how to make the Internet better. Thanks for the link. That's what Google's doing forever. I wish they would just do that more better."
But as soon as you wipe the link away, don't say where you got it, don't link to it, then I'm like, "Screw you."
Jim: Yeah, for sure. I feel like... I think I wrote another post about this of, like, why are elementary school kids held to a higher standard of citing your sources than AI from billion-dollar companies?
Chris: Yeah, you've got to do a bibliography.
Jim: Yeah. You get an automatic F if there's something in there that you didn't cite your source of it looks plagiarized and you didn't say anything. That's an F. And I'm kind of--
I think, Dave, you said this on a previous podcast. I'm still not to that level that I trust the AI summary with some things, because I've done this, too. I want to just have some good dad jokes, like back of the Laffy Taffy jokes.
Dave: Oh, yes.
Jim: Right?
Dave: Ah... That's it.
Jim: Finding that on the Internet is hard. But with AI or you just ask a bot, it's great, and there's no sort of factuality relevant there versus even a simple example of, like, what's a good example, what's a good recipe for some pancakes.
If it says two cups of flour or three cups of flour, that's a big difference, and I don't know, ultimately, who wrote that. Or if it says three tablespoons versus five tablespoons. I want to kind of check my sources on that and be sure that AI is not just munging that into who knows what.
Dave: Yeah. If Kenji Alt-Lopez tells me three, I'm in. But if it's [laughter] Google Bard, I'm out. You know?
Jim: Right. Exactly.
Dave: I need to know who came up with the three number, right? Yeah. There was a good post over on Kottke who linked to -- I said link. That's right -- to The Verge, our friends The Verge.
We should just try to get The Verge on here just to round out the ShopTalk story. Anyway, we'll get what's his name from The Verge.
Chris: Nilay?
Dave: Yeah.
Chris: You think he'll just pop right on? Yeah.
Dave: Dude, what are you doing? He's not doing anything. He can come on. [Laughter]
It was interesting, just this idea that Kottke, for the few decades of the Web, the tacit agreement was that Web crawlers could take data from sites in exchange for traffic back. But now our AI crawlers, including Google, taking too much and offering too little in return.
Chris: Yeah. it was a good way of putting it, wasn't it? It used to be cool that you crawled our sites and made search results and linked and there was a handshake agreement in that if we told you not to, you probably wouldn't - and stuff. And that's starting to evaporate on the Web a little bit.
Dave: Mm-hmm.
Chris: I think I got that from you, Jim. You were one of the first persons I read a blog that you were like, "Do you even trust an AI company to read your little robots.txt file and not crawl your site? Pfft. I think you found it laughable, and I agree.
Jim: Yeah. I think I put a little picture in there. it's like those signs when you go to a national park and it says, "Do not go beyond this point," and then there's just a whole group of people standing out on the ledge.
[Laughter]
Jim: Nobody reads these signs. Nobody pays attention.
Dave: On the edge of the Grand Canyon getting a photo.
Jim: Yeah.
Dave: Down photo, yeah. Yeah.
Chris: Yeah. Nah, we're just going to get the good press from saying we support it.
[Banjo music starts]
Chris: This episode of ShopTalk Show is brought to you buy Elicit, the AI research assistant. Elicit makes it much easier to find and understand scientific papers.
I first learned about Elicit when we had their designer Maggie Appleton on the podcast last September. Excellent episode. Maggie rules. Go listen to it.
Since then, it's been my first stop for any science question, and I can't recommend it enough. They reached out to us because they're a team of just 10 people helping more than 200,000 researchers every month. So, now they're looking for another exceptional engineer to join them. I think it's a great opportunity to join an early, fast-growing startup with a pro-social mission.
I know many of you love tools for thought, you know, end-user programming, computational notebooks. That's exactly what they're looking for.
If that's you, go to elicit.com/shoptalk to apply. That's elicit.com/shoptalk. Check it out.
[Banjo music stops]
Jim: Chris, for you, having run CSS-Tricks for all those years and kind of got out of the game a little bit sort of right before the whole AI slurping up all of your info and giving you no credit, and now you're just sort of on CodePen, which seems a little bit -- I don't know. Maybe this is an overstep -- a little bit immune to that, or at least more so than CSS-Tricks was.
Chris: I don't know if it is. Maybe.
Jim: Do you feel like--? I guess, what's your sort of posture on that and how worried are you, or are you not, with CodePen exclusively now versus if you were still running CSS-Tricks?
Chris: I would have put that fricken' robot.txt on there immediately. That's for sure.
Jim: But would that have let you sleep at night or would you still--?
Chris: No, I just don't care. I mean the entire history of CSS-Tricks and partially because it had an RSS feed, it was scraped incredibly. You know. You'd wake up... Back in the day when I had my, like, WordPress has this feature, like, XMLRPC - or something - in it. It's probably changed. I forget what it is. But it's the thing that makes pingbacks work.
You'd get three, four emails from some blogs you never heard of that are linking to you. You get a little pingback from them. Then you click and see what it is. It's like, oh, a new scraper site. Neat.
You're getting a pingback because there are links inside the article that it just scraped to CSS-Tricks articles that they didn't bother to change to their own URL. They just scraped it wholesale.
it's like, would I wake up and try to figure out who that company was and get a takedown request? No. I'd just hit archive on the email. Who cares?
There were fun, little tricks you could do like, in the RSS feed, you'd put a link that linked back to CSS-Tricks, like the original rel canonical, "This article originally published," and nobody would ever bother to scrape that out of the feed. So, in some twisted way, they were kind of giving you SEO, but that can be dangerous because sites that then get listed as spammy who link to sites, you get kind of like negative SEO for or, theoretically, because as I know to this day, nobody has ever really known how SEO works. You know?
There is no published formula for how this stuff works. And when somebody like me says, "You get negative SEO for it," I don't know. How would I know? I don't work there. I don't... I think if anybody knows that stuff, they're sworn so hard to secrecy they'd probably be in prison if they said that live.
Anyway, I went off on that, but it's not that different to me than the AI stuff, except for that the AI stuff is almost grosser because it's not link-driven.
Jim: Yeah.
Chris: It's almost more gross feeling to me that this gets stolen and slurped up and reused with no credit at all. And they didn't ask in the first place.
Jim: Yeah. No, that's kind of interesting. What can we expect over the next ten years? Well, probably a lot of the same thing that happened with the SEO game of trying to trick and all the weird things that come out of that.
The only thing that's different is that they just don't give you any credit. There's no source link. But otherwise, it might just be the same game again. Instead of search engine optimization, it's AIO.
Chris: AIO. [Laughter] It's funny. There's a playbook for being really highly successful in tech. You know?
It reminds of, like, how do you get to be a great rock band? Well, you have to do a bunch of heroin. You have to drink until you're almost dead. Then you'll write the greatest blues rock song ever after you wake up from, you know, a near-death experience in a hotel room.
You're like, "That sucks. That's not how it has to work. There are lots of examples otherwise. But sure, if that's the lesson you take from that, great."
There's the lesson of, like, Uber. "Ah, we'll just break all the laws. We'll roll into a town, and then we'll be like, 'Hey, you can't let us leave now. Look at how useful we are. We're amazing.'"
Or Airbnb like, "We don't have to follow any laws. Let's just break all the laws. But look how useful we are." You know?
Then Open AI is like, "We'll just slurp up the entire Internet and ask nobody," and then be like, "Hey, you can't kick us out now. Look how useful we are."
It's like how many times does that have to happen? That's just the playbook.
Jim: I mean even before the Internet, that was just the mob shakedown, right? "Oh, it'd be a shame if anything happened to this very useful thing in your community."
Chris: Right. You're going to need a little protection, I think.
Jim: [Laughter]
Dave: I hate when people are like, "You can't put the genie back in the bottle." I'm just like, "Yeah, you can write one law and then the whole stinkin' thing stops."
[Laughter]
Dave: You just say, "It's illegal," or it's immoral. You know? I don't know. I guess you can write, "Thou shall not steal," and everyone is like, "Except--"
[Laughter]
Dave: For cool images. [Laughter]
Chris: But they're like, "Ooh, but it'll help me write my cover letter?" I'm going to actually let that one slide.
Dave: Yeah. Yeah.
Chris: Anyway, yeah. How would I feel? I don't know. I feel like it's not well but powerless. You know?
Dave: Ooh, goll. That's a dark one.
[Laughter]
Dave: We just take that out of context. No, I feel that on Accessibility Project. We have open submissions, and I pitched to the group, like, maybe we have to take that down because I don't know who these people are. Then just showing up and barfing garbage and, like, "Oh, here's an idea. I put it in ChatGPT. Now give me money." You know? It's like, ack. I don't have enough time to edit your ChatGPT nonsense.
Chris: Right.
Dave: I don't even have time to edit human-generated nonsense. Anyway, it's hard. I feel like you need that human connection to make it work, human-to-human stuff.
Chris: I should also point out I feel like I've got to be careful when I talk about CSS-Tricks and the fact that I feel like both of you probably wrote for it at some point and that - whatever - Sarah Drasner was a staff writer at some point. Robin Rendle wrote the newsletter. Geoff was the editor. There were a lot more people than just me, so when I make these big pronunciations, it would probably have to be factored in how other people felt, too.
And if somebody said, "Oh, I want my article to be training the robots of the future," I probably would have had to build some kind of system for that. I mean it probably wouldn't be that hard. It's just altering that robots.txt. But you know. It's easy to make big pronouncements just by yourself. More people are involved.
Jim: Yeah. That's what I was going to say. Good for you, Chris, citing your sources.
Chris: [Laughter] Yeah.
Jim: Where that CSS-Tricks came from.
Dave: I like the opt-in versus the opt-out. This is kind of the WordPress hot drama of this week, right? Ka-chung!
Chris: Yeah, isn't that interesting, too? If you're a big enough company, then you get to go shake hands in secret rooms and say, "I'll give you our content for a million dollars." You know? But what a weird meeting. It seems like they have power. But do they really? Because it's like, can't--?
It's just on the Internet. It's already been slurped up. I guarantee it has. So, are you just trying to avoid a lawsuit at that point or what? What power does WordPress think they have in this secret meeting?
Dave: Yeah. I don't know. But then, to be fair, Matt was like, "I'm on sabbatical and I'll tell you my answer in three months."
Chris: Yeah. [Laughter]
Dave: In May, we'll get an answer on what's going on with WordPress and Open AI, just for journalistic integrity.
Chris: That's fine.
Dave: Yeah.
Chris: I do feel like if you're going to not comment for three months, then not comment for three months.
Dave: [Laughter] Thank you, WordPress, for sponsoring this show.
Chris: I'm clearly apprised of this situation and I will let you know once the heat has passed.
Dave: Yeah.
[Laughter]
Chris: It's weird.
Dave: Hey, Jim. You've been talking about design engineering.
Jim: Ooh, yeah.
Dave: That's kind of... The case for design engineers part 2 came up, a sequel to your 2023 banger, a case for design engineers part 1. [Laughter]
Chris: There was a ten-year gap on that one?
Dave: No, one, one year.
Chris: One year.
Dave: Just '23, right? Did I say--? Anyway. Well, well, well...
Chris: Oh, I see. I see.
Dave: Why again? Why is it back in your mind?
Jim: A good question. This is always a little bit on my mind mainly because I guess, at the end of the day, I consider myself a design engineer. You know when I was... Back in 2022, after... when Remix got acquired and then I ended up not going over in the acquisition, it was kind of like, "Well, now what do I do?"
Honestly, I hadn't really been in that position in my career for a long time. I'd always just kind of been like, "Hey, come over." Oh, okay. And so, it was like, "Oh, I have to actually submit a resume and stuff, I guess, again."
And so, I was trying to figure out how to navigate that whole process. And as I began sort of talking to people, I quickly realized that there were a lot of companies that sort of had this attitude of, like, "Wait. You're doing design. Why would you be writing code? We don't need someone who does that. If you're in Figma, you should just be in Figma," or vice versa.
"Oh, you're a developer but you want to do design stuff, too? That doesn't make sense. We have people who can do design stuff. We just want you to be a developer."
That seemed like was the attitude of a lot of companies, or at least the people that I talked to is they couldn't quite grasp why you would want someone who does both because, "Well, we have people for that. We have people for one. We have people for the other. And we want you to be one or the other."
And so, it kind of kicked off in my brain this desire to try and articulate the value of someone who does both and where they fit. And I think it's a fit as much as for the company as it is anything else. Some people are just structured to be that way, and I try to articulate, I guess, just the rationale for why it would be beneficial to have someone who can do both and why that would be beneficial for whatever it is that you're building.
And so, that's sort of where this was born out of. I honestly could write way more about it. But it does come from that thought that I feel like people have of, like, "Why would we hire someone who does both?"
Dave: Yeah. Your example in your part 2 post was just this little slider thing, before or after kind of slider, and what happens at the micro pixel level. A designer very much cares, and the developer probably doesn't care. You need somebody who super cares about both to make it great. Right?
Chris: It is interesting. Right. Jim has got this picture of a one-pixel line separating two areas, or one pixel-ish. A very thin line, right? Then a designer might just draw the line. And the developer might be like div ID=separator, and then bind the separator to that one-pixel line, being like the designer did their job. That's how they want it to look. The developer did their job. They bound the interaction to that. Which one of those two people's jobs is too thin or what are the events that trigger it, and how does it un get triggered, and what is the hover event like, and what kind of cursor does it turn into?
There are all these things that are pretty specific. Some of that is UX, I guess. Then some of it is a little bit like you need, actually, some technical know-how to kind of know how to answer those questions, maybe. Yeah?
Jim: Yeah. I think that's kind of what I was trying to get at is I don't think that if you are exclusively a designer or exclusively a developer or exclusively a UX person that you have the tools to really get into what you want to build. If you're a designer, how do you express what you want to happen on this interaction?
It's so interactive of, like, I move the mouse over. It's like, okay, is there an affordance that happens when I move it over? Does the cursor change? How does it happen? Is it on mouse in or is on mouse down? When does it happen? Immediately or on delay? Do you get some kind of an affordance with that?
Is it just the line itself or is there some kind of snapping involved when you're dragging or is there a target area for when you hover over it, the interaction? There are so many tiny little questions here that if you use... I feel like if you use what you would consider really great software, you just feel, "Oh, this feels right. Yeah. That's exactly what I would expect when I move my mouse here and want to drag this thing. It works the way I want."
How would you express all of that stuff if you were just doing static mocks in Figma, or even prototyping it in Figma?
Or if you were a developer, how would you go back to the designer and be like, "Okay, well, it should do this on hover. What color is that? Then when I mouse down, maybe it's this. What color is that and how thick should it be?"
Neither sort of individual has, I feel like, the tools available to completely express what they want to do. And ultimately, the only way you can do that is building it in the medium that you're shipping it in, which is the browser. That's why I feel like--
Chris: Prototype, you say.
Jim: Yeah, prototype. Those might be useful.
Dave: If only somebody cared about those enough. [Laughter]
It gets to the idea of feeling, right? How is this supposed to feel when you click this little thing and have this interaction?
I think about car design or anyone can build a car. We can all build a car. It's just some electricity, some motors, some wheels, a platform, and a steering wheel, hopefully.
But "How does it feel to drive that car?" is the engineering part, right? How does it take a corner? What does it sound like when it starts up? That's kind of engineering. If it goes ka-chunk, ka-chunk, ka-chunk, ka-chunk and sounds like a bag of bolts being poured on the floor, that's not going to make you feel good about your purchase - or whatever.
Jim: Yeah. How do you spec feel?
Dave: Yeah, and is that a higher-order need or basic need? You know what I mean? Am I talking about a luxury thing or am I talking about--?
Chris: Ooh, that's good.
Dave: --this is table stakes for identity and brand or whatever.
Chris: Yeah. Maybe somewhere in the middle because it does really matter. You could design an app that just is a feature for some other app but doesn't get these little feel things right and it will be way worse. People just intuitively know, like, "No, this is--"
Maybe without even being able to articulate it if you have these little one-pixel lines. I'm just circling one thing, but obviously, we're talking more generically about all these little, interactive things in apps. If none of them have good polish to them about how they're used, it might be dead before you even get going.
Dave: It's got to be really compelling otherwise, I think.
Chris: Right.
Dave: You can have nothing, but it's got to be very compelling content, right?
Chris: I think most companies now just get lucky, right? You get somebody who sweats these details in the door and they work on it, but they just happen to have landed in the right place. Or somebody with taste high enough knows that these details are important so that they make it happen. But it certainly doesn't have a name, like Jim is giving it. I've never seen a job opening for a design engineer.
Jim: Yeah, do those exist? They've got to exist. I feel like I've heard people call themselves that.
Dave: I think they exist more and more, but it takes, again, a company that understands what they're looking for. It's hard. That's where I wonder if it's kind of a higher-order thing, like people don't know they need it until they do.
Even getting... I just read an article -- I sent it over to Chris -- about high contrast mode. It's like you don't know you need or you can use high contrast mode to draw borders. You can do border color transparent. We were talking about it in the D-d-d-d-discord. Melanie brought this up.
You can just use border color transparent instead of border zero. You basically have a transparent border now. But in high contrast mode when it strips out all the background colors and all the everything, you don't just have floating text. You have a border around your buttons. That's a pretty cool feature. That's something I think a design engineer would know or do, too. Just kind of like, "Hey, here is--"
Chris: I didn't know that. It colorizes your otherwise colorless borders?
Dave: It'll just draw a black line around - or whatever - or white if you're in the black on white.
Chris: That's cool.
Dave: Or dark mode. But yeah, it's just kind of like a very, you know, you don't know that until you have experimented or felt those. It's just a different world.
Chris: That's cool. How are we doing here?
The other one I had written down was the RSS in HTML kind of thing. We already covered the RSS angle a little bit, but it was kind of an interesting thought from Jim. XML, really? I don't know if that was your... Was that the intent, like, what a weird, old format? Sure, JSON is a little nicer, but there's not a lot of movement in that direction necessarily. I mean there's a little bit, but that's not particularly human readable, and maybe the chill format of HTML could be a follow-up to a new format for RSS.
Jim: Yeah, that one, I feel like people... I had a few people respond to me with that with very reasonable comments of, like, "Yeah, this isn't going to happen. We already have XML. It's fine."
I don't disagree with that. I guess it's that purist in me that, "Gosh, dang it. Sometimes I wish you'd just shut up," won't let that sit.
I think, Dave, you did a post on this. I think it was the first time I saw it. You referenced -- I can't remember who it was -- of how to style your XML feed. So, when somebody clicks on your RSS feed, it actually shows up in the browser and looks like something rather than just gobbledygook XML.
Dave: Yeah. Yeah.
Jim: It's really great, and I was like, "Cool. I want to do that." I got in there, and I started trying to wrangle XML. Granted, it was just one night after the kids had gone to bed. I'm like, "I could do this real quick," and I just couldn't. I gave up, and I thought, "That's too hard."
Dave: Yeah. It was Darek Kay who wrote "Style Your RSS Feed."
Jim: Yeah.
Dave: And it was, you use this XSLT, which is like an XML style sheet template or whatever, and you can basically just reference it on your RSS feed, and it'll come out styled. The browser still somehow miraculously knows how to lay out the webpage with this weird style sheet. I don't know. It's weird.
Jim: Which is cool, and I wanted to do that, but I just ended up giving up. Yeah, then I just started thinking, "Wait. I feel like at this point HTML has won."
It's going to be around for a very long time. Probably longer than XML. Are we still going to be doing feed URLs in XML in 50 years? Can't we just get something that when you click on feed, it just shows up in the browser and is understandable to a normal human being? Then they can just, "Oh, yeah, cool. I kind of know what I'm looking at. I'll dump that in my feed reader."
Chris: Yeah, that's so tricky. I remember getting the email from that same guy and writing back and being like, "This is weird. I know I'm not a normal person. But when I see the styled pages, then I don't know that it's a feed."
It doesn't register as a feed to me. It looks like a webpage. Whereas when I see that gross XML, I'm like, "Ooh, that's exactly what I'm looking for." You know?
Dave: [Laughter]
Chris: Not that I look at each one. Usually, I trust it and just copy the link when people say, "This is my RSS link," or just let it be auto-discovered or whatever. But the styled feed, I feel like, has limited utility. Who is that for? You should either look at my webpage or give me--
I don't know. It doesn't sing to me.
Jim: I mean that's a fair point. That's kind of like when you get the knockoff Doritos and it doesn't leave the little gross stuff on your fingers.
Chris: [Laughter]
Jim: That's what I want. I want that gross. Then I know it's authentic.
Chris: [Laughter]
Dave: There's a user experience issue, though. When you see RSS, an RSS feed, an XML file, you think, "I broke the computer." [Laughter] "I have messed up a big one."
It's like when you send HTML with the wrong mime type. It's like, "Oh, whoops." I'm thinking back to my PHP days. Yeah, there's something broken feeling about it. But it doesn't make it easier. I agree.
It's like, "Okay. Well, cool. You got to this RSS feed. But now it looks like a webpage."
Chris: Didn't browsers used to do better? Didn't Firefox used to apply some general styling to it and be like you could put a little banner at the top that said, "This is a feed."
Dave: Safari had an RSS reader inside of it. It was awesome. That was awesome.
Chris: That's wild.
Dave: IE 5.5 for Mac had -- 5.5, weird, but it existed -- had a feature where you could say, "Hey, I'm going to bookmark this website. Tell me if it updates." And it would go ping it every day or every hour - or whatever - and it would come back and say if it was different. It was very cool. That's cool. Why don't we go back to that?
Instead, we're like, "Notification badges, that's the way to do it." [Laughter] I don't want that, but I guess that would work.
Jim: I feel like Safari, if I remember right, they're the worst now because if you click on an XML feed, you don't even see anything. It just says, "Do you want to download this file?" And you can't even see.
Chris: Oh, it offers you to download? No.
Dave: Oh, that's bad.
Chris: That's awful.
Jim: At least Chrome shows you the XML ugliness.
Chris: Yeah. The pushback, if you just use HTML instead, it's not strict enough. HTML is just the wild west, right? Not that I'm some XML lover, but it's very, like, "It's got to be like this."
Jim: That's where, when implementing my ultimately HTML feed that I have, I just ended up... Everyone was like, "Oh, microformats." I was like, "Oh, yeah. I guess I kind of remember those things."
I actually already had them on my blog, so then I just put... It's funny because, ultimately, the HTML feed that's on my page now is basically what a blog used to look like in 2008 of just, "Here's the ten latest. Here's the content of the entire post." Then you just would paginate through them. That's basically what this is, and it's just the ten latest ones.
Chris: Yeah. But you have that and you have... It's not like they're the same page, right?
Jim: Yeah.
Chris: You intentionally made a separate version of your site that is that way, which I think is fine. But I wonder, if it's just microformats, those were kind of designed to be like you sprinkle them in your real website and they had some utility there. I wonder if you could just make actual websites.
The reason I'm bringing it up is one of the things that always annoyed me about Amp -- and many other things (the Apple news format and the Facebook weird format). Remember that era? That's a great era to be over when all the discussions were like, "Y'all should make four versions of your website to satisfy all of our requirements."
People were mad about it for lots of different reasons, but that was always my number one was that I don't want to make four or five versions of my website to satisfy all your requirements. That's not tenable. We just went through this era of responsive design where it was like, "We're making different versions of our website for mobile and not mobile." That's stupid. We should make one and have it work for everything.
We just went through that, so to all of a sudden be like, "Nope. You've got to make a special Amp one, too." I always thought that was so annoying. But at the same time, I'm not bothered by having this XML version of my website, which is kind of the same, so I feel a little contradictory in that way.
Jim: Yeah, I get that. I mean that's why I didn't want to do this because it was like, "Ugh. This is just... I've already got a JSON feed and an XML feed. Why do I need another feed?"
But at the same time, yeah, I didn't want to do XML. I was like, "Can't we just do HTML for everything and figure out a way to do HTML?" I don't know. It was basically just an exploratory thing.
Dave: Yeah. I still maintain if they would have called Amp "RSS 3," I probably would have been on board.
Chris: Ha!
Dave: Whoops.
Chris: [Laughter]
Dave: Google, call me.
[Laughter]
Dave: If you want to do this whole thing where you suck up everybody's content, slurp it, and then train your AI on it, which I think is what they were doing--
Chris: [Laughter]
Dave: --just call me.
Chris: Whoa!
Dave: Call me.
Chris: Amp was an early AI training tool? Oh, I have not heard that one.
Dave: Oh, buddy. Get your tinfoil hat out, brother. I've got some conspiracy--
[Laughter]
Chris: Gem and Amp.
Dave: Oh...
Chris: Wow. Okay. [Laughter] Well, that was pretty good mouth blogging. Is there any reader questions or other blog posts you want to bring up? We didn't do the state one. We could do that one real quick.
Jim: Which one was the state one?
Chris: The state one was that... Dave, you blogged very different things, but they were both about UI is a function of something. Dave said it's a function of state, which I don't think you linked to that, but there was -- who is the React guy? He had one that was like UI is a function of state and data and talked about stuff. And so, I think this was a little blog theme going on.
But Dave's post was all about how much state there is. Sure, there is some stuff like is this button clicked or not. But there's global state, page component state, element state, second-party state, device state, browser state, user state, third-party state. There's just... And he lists. What did you have, 200 bullet points on this thing?
Dave: Yeah, easy, I think.
Jim: It was pretty overwhelming.
Dave: It kept going.
Chris: Yeah> [Laughter]
Jim: And depressing, to be honest. [Laughter]
Dave: Well, yeah. A couple of people felt depressed, and I apologize. But it's... Also, back to this whole design engineering thing, hire people who are good at this. It's big enough that it takes dedicated brains to do it well. You can't just--
You're not going to find somebody who is super good. [Laughter] Maybe. I'm generalizing. Somebody who is just a Python wizard just training AI who is just going to be able to do this UI stuff competently. I think it's so different than what you're... It's apples and oranges at some point.
Jim: I have a post, actually, that links to that one of yours, Dave. I haven't published it yet. It hasn't published... the next one. But it's this idea of it's notes from a talk that I read by Peter van Hardenberg, I think is his name. He's the Ink & Switch guy, local first software, all of that.
He's talking about complexity in software. And one of the things he talks about is how independent dimensions of our software multiply problems. When you're building a Web app, you have three. You have Chrome, Safari, and Firefox. Then you have different platforms: Mac OS, Windows, Android, iOS. Then you have different screen sizes, which if you oversimply if desktop, tablet, mobile. And then you have different network speeds: wi-fi, 4G, 3G, whatever.
All of these different dimensions sort of multiply against each other. If you're talking, okay, that's three browsers times four of the latest versions of each times four platforms times three screen sizes times four network speeds. That's thousands of different combinations of things.
Then if you take Dave's post, which just lists out way more pieces of variability, that's millions of combinations of things that can happen. And it's just kind of insane when you think about it.
He kind of points out, like, that's sort of why Electron apps are popular because it seems to--
Chris: Oh, that's funny. It takes that binary tree and cuts a couple of branches off of it.
Jim: Yeah, well, what was interesting to me about it is he kind of says people see Electron apps and they think, "Oh... no..." People don't like using them. But it's sort of like building different applications for different native platforms, that's really hard. But even harder is managing the people who are going to build those different platforms for different--
Chris: Hmm...
Jim: And so, that's why Electron apps is like you can have one team that works on something that works across all of these, and so it's supposed to bring down that variability. But still, even then, it's just an insane amount of things, as I think Dave's post illustrates really well.
Chris: What was it? Why do people hate them so much? I feel like that's--
Dave: John Gruber. It's cultural.
[Laughter]
Chris: I didn't even remember.
Dave: Also Yankees. [Laughter] But I think there's... Yeah. Electron, I think, for a small team that is trying to not have five different OS covered for this native application and then is just trying to iterate fast. Everyone wants this "Write once, run everywhere" runtime, like Java or whatever or React native. That it's big pitch, too.
But guess what is kind of already that is a Web browser is running on multiple platforms. It runs on Linux. It runs on Windows. It runs on Mac. It runs on iOS, kind of. Then it turns on Android. Maybe the Web browser is the right way. Maybe it's the better way.
We give up some performance florps, I'm sure, right? You give up some FPS or something or memory thresholds to Chrome. But it seems like a good deal.
I don't know. That starts tying into why do we need Electron. PWAs, everybody. Except, *not available in Europe.
Chris: Hmm...
Dave: Yeah.
Jim: It was kind of related to Dave's post. And, Dave, you linked me to an interesting article. The one that I wrote on mine was UI is a function of your organization. The idea being you can only build a UI that can deliver on what your organization is capable of delivering on.
Dave: Yeah. Yeah.
Jim: The pizza tracker is such a... Having worked at a company that wanted to build a pizza tracker for not pizza but something else, it was such a... I sat in so many meetings where the designers that we had on the project were sort of like, "Here's how simple it is. It's five steps. There's this, and then this, and then this."
Then once we got to the rubber hitting the road on the ground of people who were going to have to actually deliver on all of these things and getting that tracker to show up, it ended up being like, "Okay, here's step three, and that could take two to six weeks."
Dave: [Laughter]
Jim: And it's like, "What is the point of having a tracker at this point?" That is not useful to anyone.
And you linked me, Dave, to the article about some guy who went to a Dominos.
Dave: Yeah, staked out his own Dominos. He put an order in that was kind of weird.
Chris: Yeah?
Dave: He knew it was his pizza, and he staked it out and followed the whole pizza tracker chain.
Chris: Whoa!
Jim: And basically found out that it wasn't literal.
Chris: It was just a big lie?
Dave: Yeah. His conclusion was it was not real, and very good evidence that it was just made up. So, either it's made up or Dominos made something very cool, which is also possible. They have QR codes on every - whatever - pizza box, and it's fully tracked to Diana who is delivering your pizza. You know? Either they did something cool or they did something genius in that every other company, large pizza company, had to replicate this functionality for millions and millions of dollars. [Laughter]
Chris: Hmm...
Dave: And they basically--
Chris: I mean isn't it... Or isn't it in the middle? It doesn't have to be all a lie or all true. It could be like it does know when it leaves the store, but that's about it. All the rest of the stuff is made up - or whatever - or that it's based on a bunch of averages or whatever.
Dave: I think it's based on averages. This was sort of the idea. But I don't... Yeah. I can post a link so everyone can read it.
Jim: And the link that Dave sent me was interesting because it linked to some other article. I think it was from The Atlantic. See, we're just following links. Links are fun.
Dave: Links are fun.
Jim: It was some article in The Atlantic that talks about... What was the word? It was benevolent deception or something like that, this tactic that designers can use to make it seem like your pizza was ordered and then John is preparing it. He's putting your pepperonis on. Then it's going in the oven, and now it's out and it's getting a quality assurance check.
None of that stuff is actually real, but it's being shown as if it were real and that people are like, "Yeah, that's okay. We're fine with that." [Laughter] It's not the truth, but I don't know. I guess it calms people's fears of, like, "Where is my pizza? I want to know where it is. Is it going to be here soon?" Versus just like, "We received it. It left the store. It should be there soon."
Chris: Yeah, I wonder because you're like, "It's probably true." Or if it's timed, it's definitely true at some point because my pizza is definitely going to get here.
I wonder if that's because you're so used to being lied to, you're so used to putting a swimming suit in your cart and then, two days later, having it be like, "There's only one of these swimming suits left. Do you want to have it be yours?" There's that feeling that you're lying to me. I don't know. You're like, "Well, at least somebody really did put pepperoni on my pizza at some point because I ate it." It's better than being lied to.
Dave: Well, isn't it sort of like security theater, kind of? Like, "Oh, yeah. Everyone goes through this line. But if you pay us $60, you don't have to do the super secure line that makes everything take a long time." For $60, everything is okay. [Laughter]
Chris: A lot of theater. Back in the Wufoo days, when you signed up for a new account, there was a good 20, 30 seconds, I think, after you signed up that just this long animation played that said, like... It had little silly things. A dinosaur running across, and it told you that it was building you a database - and who knows. I forget all the stuff that it said it was doing, but it wasn't doing any of that stuff.
Jim: [Laughter]
Chris: It was just winging.
Jim: Yeah, that's an interesting phenomenon to me. Everybody has experienced this, even with progress loaders. They make it seem like it's actually pulling the server that's doing the work. But that's not happening. They're just guessing. Oh, we're going to say we think this takes 20 seconds, so that's what we'll make the progress indicator sort of fill up over time.
I don't know. It kind of works on our, like, primitive brains, I guess, of, like, "Okay. Yeah. I can wait." But even though it's not actually representative of reality at all.
Dave: Well, I think... You know. I like this function of org thing because it is. It's like this sleight of hand, right? It's like you can have the best UI, but if your company cannot deliver on that, it doesn't matter. I think that's what got Microsoft Windows messed up. They tried to be like, "Here's a progress meter for your file going." Then it took forever. People hated it, and it felt slow all of a sudden, and people thought it was bad.
Then you go try your friend's Mac and it's got a little beachball that spins when it's thinking about something, not a loader that's stuck. Everyone is like, "Oh, the beachball is better, man. Look at that." It's just like what can your org deliver on is so important.
Jim: And I think it's a disconnect that I've seen a lot between designers within an organization who end up sort of in the kind of bubble where they're just like, "Yeah, I'm just going to imagine the best thing possible," which is great. It's good to be thinking those thoughts. But then there's ultimately this disillusionment of, like, "We can't do that. We can't deliver on that. It's going to take this person this long to put the rubber stamp on it and move it through."
You're not just designing a UI, right? You're designing whole new processes within the organization and how people communicate and work together.
Chris: Right.
Dave: Yeah. I have a story like that from my history, but that's maybe for another time.
[Laughter]
Dave: It'll take another 35 minutes. But I realize we're kind of at time here. But Jim, thank you so much for coming on the show. Really appreciate it. We'll have to cut it short. We'll have to get you on again. But for people who aren't following you and giving you money, how can they do that?
Jim: Well, you can go to my website. Maybe soon it will actually be clear when you go to my homepage that there are different subdomains that you can go to.
Chris: [Laughter]
Dave: Wee!
Jim: For my blog or my notes. Yeah, I need to do that. Thanks for the tip, Chris.
Chris: Yeah. Yeah. You're welcome. [Laughter]
Jim: Yeah, jim-nielsen.com, that's where you can go.
Dave: Nice. Jim's got also -- I don't know. I'm going to throw this out. Jim has got a lot of cool ideas on his blog that you should probably steal. Anyway, like tracking links, who he links to, and all this stuff. Articles on Hacker News and stuff like that. There you go. Just go. Anyway, go check out Jim's blog. It's blog.jim-nielsen.com.
Then, yeah, thank you, dear listener, for downloading this in your podcatcher of choice. Be sure to star, heart, favorite it up. That's how people find out about the show.
Follow us on Mastodon. And then join us where the real party is over in the D-d-d-d-discord, patreon.com/shoptalkshow. Chris, do you got anything else you'd like to say?
Chris: Hmm... ShopTalkShow.com, the link, the URL. Feel free to link to us.