Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Episode 29: AI and Pattern Writing
226 Plays2 years ago
Nicky and I took a bit of a dive into how AI works and if it could write knitting or crochet patterns. What we found was both helpful and hilarious.
Recommended
Transcript
Introduction to Podcast and Hosts
00:00:14
Speaker
Welcome to Craft, Design, Edit, Sleep, Repeat with hosts Lisa Conway and Nikki Jensen. Listen, as we take a deep dive into the business of fibercraft design.
00:00:38
Speaker
Hi, Nikki. Hi. How are you today? I'm great. Thanks. How are you? Good. And how is your mid April weather? It is unbelievable. We had about three days of spring and now it feels like summer. Oh, you're kidding. Yeah. We had like three days of spring and this week we've had six inches of snow. Oh my goodness.
00:01:05
Speaker
We've got temperatures in the 20s Celsius. I'm not sure what that is in Fahrenheit, sorry. Oh, that's okay. That's actually around 70 to 80, I think, ish? Yeah, it's been warm. That is summer. I mean, that's July 4th for us.
00:01:32
Speaker
Yeah, that's the only time we get like above 70. Yeah, I definitely wasn't expecting that for early April.
Exploring AI in Business
00:01:44
Speaker
Well, Nikki and I were trying to come up with an idea for today's episode and Nikki very nonchalantly typed in to one of those
00:02:02
Speaker
idea websites and it got us to thinking the big topic of conversation right now is AI and using AI in businesses. So we thought it was a good time to hit this subject now while it's hot and people have lots of questions. Right, Nikki?
00:02:23
Speaker
Yeah, I think that's a great idea. And so my question to you is, what is AI? Well, you know, that was kind of a confusing thing. I know that it stands for artificial intelligence. But when you start researching it, one of the things that really got me was they would intermix AI and ML.
00:02:44
Speaker
And it was like, so what is ML? So I literally had to Google, what is ML? Well, ML is machine learning, which is what is the basis or the underneath aspect of all of this AI that we've been talking about, because what they're doing is getting machines to learn how to respond to these questions.
00:03:10
Speaker
And whether we know it or not, we've actually been using it for years because if you own an iPhone or a Mac of an Apple product and you use Siri, you're using AI. If you have an Alexa, you're using AI. If you have one of those Google, what do they call them? Google dots? Oh, I don't know. I've heard of the Google Home.
00:03:38
Speaker
Oh, Google Home. That may be what I'm thinking of. Yeah, that's AI. When you say, hey Siri, you're activating AI. So the basis is machine learning. It's the machine learning how to respond. So that's, you know, kind of what AI is in its core form.
00:04:05
Speaker
If that makes sense, I'm not sure I fully understand it either myself, but so the next question that we really were wanting to dig into and Nikki had some really great answers.
Limitations of AI in Craft Design
00:04:20
Speaker
So I'm going to ask her, let our listeners know what AI is not good at.
00:04:26
Speaker
Well, the TLDR on this one is no AI cannot replace a pattern designer at writing knitting patterns for you. It's just not there yet. And I don't know whether it ever would be.
00:04:45
Speaker
It can write patterns that sound like knitting patterns but so far they don't really result in anything particularly useful. Operation Hilarious Knitting Disaster was a group project that started in 2017.
00:05:00
Speaker
by an AI expert and then there's a Ravelry who were trying to teach AI how to write knitting patterns and if you go on Ravelry you can see the hilarious results of these patterns that AI wrote that people then went and tried to knit themselves.
00:05:18
Speaker
So what they found and what I've noticed in my research is that AI really depends heavily on the input it's given in order to generate a useful result. And it takes a lot of time and practice to learn how to write effective inputs. It also seems to focus more on the language aspect and not pay adequate attention to the math that's involved in a knitting pattern.
00:05:38
Speaker
And the Ravelry project, which is known as Skynet, found that AI was able to generate more useful results when it was fed patterns with only very, very basic shorthand instructions to learn from. As Ian, like maybe a cow or a shawl versus a garment, or where did they have any success at all?
00:06:08
Speaker
Or did they have any success at all? Well, I didn't see any wearable garments. They mostly were small, kind of free-form lacy swatches that they were able to produce. It would have to be very specific, like knit for
00:06:33
Speaker
bind off three that kind of thing rather than a mix of math and Descriptive instructions that you would see in a typical knitting pattern. Okay. I know the same thing happened with crochet Mm-hmm in in looking I ran across some YouTube videos of people that had attempted to crochet these things from
00:07:00
Speaker
AI generated crochet patterns. One was supposed to be a hat. And it was this really long strip with one little like mountain hump of stitches on it. And that was it. There was no in the round. There was nothing. It was hilarious. There was one
00:07:21
Speaker
Yarn Geek did a live video with her watchers where they were talking about chatgbt and this is a recent video by the way where she was talking about having gone to chatgbt and generated this crochet pattern for a Christmas stocking
00:07:41
Speaker
And it had her crochet a circle that was by no means flat or even rounded. And then she said if she'd have followed the next set of instructions, it had her do the heel on one side of that circle and the toe on the other and then it said it was done. And neither one of them would have created the correct shape.
00:08:07
Speaker
Yeah, yeah. It's really, the results are hilarious.
00:08:13
Speaker
Yeah, we'll put some in the show notes of some examples of things that people have tried to make. They're really funny. I used any word to try to generate a knitting pattern just to see what would happen. I had originally asked for nine sizes, but it could not compute and asked me to change my prompt. So I ended up writing it in five sizes that I wanted. I gave it measurements.
00:08:39
Speaker
And the instructions actually looked really convincing. But when I tried to tech edit it, the math did not work at all. I've never laughed so much at a spreadsheet. The collar was 20 inches long.
00:08:58
Speaker
yeah the the measurement that i gave for the sleeve circumference ended up being used as the armhole depth so now i've got a 10 to 15 inch long armhole depth and the stitch counts for the sleeves and body just turned out tiny tiny so at best this might create a halfway decent sort of template that you could start filling in the math from yourself but
00:09:26
Speaker
At this point in time, it just doesn't seem to know enough about knitting to work out the correct stitch counts and grading. When I asked any word to do a knit pattern, I didn't even specify multiple sizes. I just said, you know, give me a knitting pattern for a raglan pullover. That was the entirety of my instructions to it. And it actually came back with five sizes.
00:09:56
Speaker
Wow. But as you said, the math was just where you insert the sleeves just was completely off. And I didn't even really go into tech editing it because you could see that it really didn't have a body. You should try it later. You probably have fun.
00:10:23
Speaker
And I read a blog post by Jenna Sargent where she tried to tech edit a pattern very similar. And yeah, she got very similar results too. Like basically a neck and no sleeves. That one was hilarious. Yeah.
00:10:44
Speaker
All right, so let's, we know that it's not very good at generating patterns. Guys try it. It is so funny. It really is hilarious. But did we find anything that it is good at that might be useful in a craft business?
AI Tools for Productivity and Writing
00:11:09
Speaker
Actually, we did. One thing that AI and machine learning can do for us is help to compile our information into one place. Just like with you going out and searching for ideas for our podcast, it can help us trigger that brain dump that we need. And it can bring
00:11:40
Speaker
your ideas into a more cohesive pattern. There was a really interesting article on that at prowritingaid.com. And I will link to that because it's worth reading because it really helps you figure out how to use AI in your business. It can also help with generating advertising copy.
00:12:09
Speaker
And when we get to talking about some of the different places you can go to use AI, I'm going to come back to that idea a little bit. But idea generation, compiling your information, it can also help with summarizing your information. So if you have written this long piece of advertising copy and you know that it's too much for the space that you're going to have on your,
00:12:39
Speaker
pamphlet or what have you, you can use AI to shorten it, to break it down into its base elements. So it's really good at that.
00:12:50
Speaker
You could even take a blog post that you've written and use AI to condense it down into a social media post, for example. Yes, yes, excellent. That's the summarizing aspect of it. So there's really good things out there that AI can help us with.
00:13:11
Speaker
So this leads me to some websites that I found. And I know that you found a few as well. So what types of websites are available? And where can people go?
00:13:30
Speaker
yeah so there are a whole bunch of these and most of them are i believe based on the chat gbt in the background and they've kind of geared them toward different uses for example pseudo-write is one that we found that's for writing fiction but it's really neat because it can it can help you to change the
00:13:54
Speaker
the mood of your writing, and I'll get into that a little bit more. Any write and copy AI are other examples that can help you with copy, writing copy.
00:14:11
Speaker
Okay, so and you talked about chat GBT being the basis. I'd like to point out here that chat GBT is actually an open source. So what that means for people who aren't in the tech world is open source is where they have a
00:14:36
Speaker
way for lots of different programmers to put their input, their coding into a bigger picture. So let's I'm going to stray here just a little bit. Linux is an open source operating system that was created to
00:15:05
Speaker
kind of counter all of the issues that they were having with Windows, which is a mainstream for purchase operating system. When we go and buy a computer, we buy one with either Windows or the Apple operating system. And Apple has always been proprietary, and Windows has always been connected to specific hardware.
00:15:33
Speaker
Well, Linux wanted to break the pattern of Windows being that general go-to operating system. So hundreds of thousands of programmers contributed to that operating system. And it's used very successfully by even businesses
00:16:00
Speaker
still today totally for free because nobody's charging for it. Well, chat GBT is the same. It was open source with lots of different programmers and different experts contributing, and these other places are pulling that into their systems and using it with specific parameters like pseudo-write.
00:16:27
Speaker
So that gets really deep into the technical of it. Sorry, guys. But I do want you to understand why chat GBT is the base for all of this. It's because hundreds of thousands of people have contributed to that code. And that's why it can be a foundation. And anybody can use it for free. So.
00:16:52
Speaker
Yeah, and it seems like that's become the foundation for any sort of
00:17:02
Speaker
Because the prompts need to be so specific, people have written user interfaces for specific types of uses on top of the base of chat GBT. That's a really good explanation. It's the user interface that other people have written and layered on top of what chat GBT does underneath to actually create the words.
00:17:29
Speaker
And I'm also going to point out here, we've been really focused on writing, but AI can be really good for art generation as well. And I'm going to kid on that a little bit more later too, because there's a really specific aspect of that that's interesting.
00:17:48
Speaker
So in our research, what types of apps, applications that we can actually put on our computers or our tablets, have you found that can be used for AI?
00:18:03
Speaker
Well, one that I've been using for a while in my tech editing business is Grammarly because it includes a spell check, grammar check. It can help you with the wording and voice and choosing your audience.
00:18:21
Speaker
in your writing. So quite often when I get a pattern from a client and it's a PDF, my PDF writer or reader doesn't allow me to spell check. So I copy that pattern into Grammarly and I can spell check that way. One thing I noticed though is you do have to be a bit careful with that. It's not perfect. I noticed the last time I used it that
00:18:49
Speaker
It missed things like an extra space or a closed parenthesis was missing and Grammarly didn't pick up on those things. So even a task like that doesn't replace a human editor, unfortunately. That's so true.
00:19:09
Speaker
It also can't check any of the math. And quite frankly, the math is one of the biggest potential errors that we find, isn't it? And there are others besides Grammarly. When I was actually taking my tech editing course, I found a site for, it was actually designed to look for plagiarism.
00:19:38
Speaker
which I found really interesting too because and very helpful because it helped me in my in my writing prevent from accidentally plagiarizing when I was doing research type reports. So you can do a search for plagiarism
00:20:00
Speaker
check and find it. It's been so long since I've used it. I'll have to really dig to find the link, but I'll try. Another app that I have noticed is advertising new AI components is Notion. Notion is a productivity app, kind of like Evernote.
00:20:31
Speaker
I did a bunch of research, oh, probably about a year and a half ago, looking for ways to keep track of all of the little pieces of my business, right? My client contacts and my business contacts and my ideas and my marketing information. And Notion kept coming up over and over again because you can literally use it like a brain dump
00:21:01
Speaker
And then have the app help you move things around and put them into an organized list of things you need to get done or
00:21:17
Speaker
Ideas that you need to create videos or podcasts on and things like that. I have not used notion, but I know that they're advertising a lot right now for using AI and AI being added into that their app.
00:21:36
Speaker
Interesting. I'd be curious to try that and see how it would how it would help a business like mine. I'm always looking for the next like little tool that's going to streamline things more. I think you have to be a slightly different type of thinker to use notion because it is very stream of consciousness type.
00:22:05
Speaker
And then you have to help get it reorganized into the way you need it. It's a very big app. It's a very powerful app. But I think there's a really strong learning curve. I ended up, instead of going with Notion, I went with one called Infinity. And it's kind of similar to Trello, but not. Trello is more streamlined.
00:22:37
Speaker
in many ways, but infinity has recently come out with AI and I haven't played with it yet. So that's another one. What else have you found?
00:22:52
Speaker
Well, when I was researching for the podcast today, I tried out pseudo-write, and that one was really fun. If you ever feel blocked when you're trying to write, for example, your pattern romance, trying to create a descriptive, evocative type text, I think you'll really like pseudo-write.
00:23:12
Speaker
You can ask it to rewrite your text using a variety of different criteria. You can make it more descriptive, more concise, more emotive, more visual, and you can even click on individual words to see options for replacing them, so you can use it as a built-in thesaurus as well. So that one was really fun to play with, and I could see that definitely being useful in pattern writing.
00:23:37
Speaker
Oh, I definitely am going to have to check that out because writing my romance is the hardest part. Oh, always. Yeah. I'm a technical writer. I'm not like a flowery writer. So I feel like this could be really helpful for that. I think that you and I are very much the same in that I was always really good at research papers, but don't ask me to write anything.
00:24:02
Speaker
that was fiction. Research papers are my thing. And I've got one of my three kids that is a really good fiction writer. And it was really hard for me as a homeschool mom helping to encourage that because I'm not good at it. So I always have tried to surround them with people that
00:24:30
Speaker
helped with that aspect. I deliberately went out and looked for places and things that we could do that would help generate those and support those talents. The other one I found was Canva. How many of us use Canva to help generate our Instagram posts and our Facebook posts?
00:25:01
Speaker
all kinds of other, I mean, ad copy and they're really, they're not limited to just the writing AI either. They're actually utilizing the art AI as well. So they're adding a whole bunch of new features right now and it's really worth taking regular looks because from one week to the next right now it's changing and improving.
00:25:31
Speaker
And that was one of the places I ran into the summarizing because you can take your blog post, put it into Canva, and it'll summarize it into, like you said, an Instagram post or an ad copy for a pamphlet.
00:25:51
Speaker
And it did such a wonderful job of distilling and bringing it down to just base points. And it'll do the opposite as well. You can give it a prompt and say, I need five points in this area. And it will give you ideas for those five points. Now, you'd still want to edit.
00:26:13
Speaker
Excuse me, you still want to read it very, very carefully and make sure it says exactly what you want it to say. But wow, what a great brain boost.
00:26:26
Speaker
Absolutely. Yeah. I tried something similar with copy AI. I said, you know, give me five, um, give me five questions about technical editing and it gives you five. And then, and then you can build, you could build a blog post about that. Like you could say, you know, top five things that people are asking about tech editing. There's your blog posts.
00:26:51
Speaker
So it just, again, you do kind of, you still are going to have to write the blog post yourself, but it gets those ideas flowing. Right. And I think that's one of the best places to use AI right now is to get the ideas formulated in such a way that you then can add your voice to it.
00:27:18
Speaker
And that even just that can save a lot of time. I played around on Canva as well. They have options to use the AI for creating Instagram posts. So you can say, I want to create an Instagram post and it'll give you
00:27:34
Speaker
a bunch of templates to choose from. It'll ask you for your colors. It'll ask you for your style. Like, do you want a modern style? Do you want a classic style, minimalist style? And so I can see how that would save time over going through all the templates that they have and choosing one and picking your colors one at a time.
00:28:00
Speaker
So that I think is something that I would definitely use in my business to save some time on creating posts.
00:28:10
Speaker
I think that the release posts for this episode is definitely going to be a place for me to play and spice things up a bit. Yes, I love that idea. Because I just do this generic thing.
00:28:32
Speaker
Well, it's so funny because I when I was playing with copy AI, I just put in the keywords knitting and artificial intelligence. And it comes out with AI is a hot topic this year. And there's a lot of confusion about what it is, how it works and how to use it. So we thought we'd help you out with some tips for using AI in your business. I was like, wow, I could just say that on the podcast and I'd be good to go. Yeah, no kidding. What a great we should have used that as our opener instead of mine.
00:29:02
Speaker
But that's where it sort of started to go off the rails because it says, we're talking about a future where machine learning and deep learning will soon be able to help you knit your way to success. Let's take a look at how artificial intelligence can help you with your projects by creating patterns for you automatically and helping you find the perfect yarn and recommending the perfect color. I'm like, well, that's not true. It sounds compelling, but.
00:29:29
Speaker
I think it believes it can do more than it can. It seems that way, doesn't it? We haven't quite gotten to the point where it's thinking for itself. I'd love to have that kind of confidence, though.
00:29:48
Speaker
No kidding. I definitely am going to have to play with some of these some more because we came up with this idea and it was like in the middle of all of this other stuff so I did my research but I didn't have time to really sit down and play and I really want to sit down and play.
00:30:07
Speaker
And that was what I found almost right away. There's a learning curve to it. And the more practice you get with creating prompts for the AI, the better results you're going to get. So you do have to take your time and play with it. Gotcha. Gotcha. That's a good suggestion for people.
00:30:31
Speaker
So I have a question for you about what might be legal issues with the use of AI?
AI-Generated Content and Copyright Issues
00:30:46
Speaker
That was something that really surprised me when I was doing my research is coming across an actual video on YouTube that talked about copyright and the use of AI. Now, this was related to AI-generated artwork. So I'm going to put this big disclaimer. A, neither one of us are lawyers. B,
00:31:13
Speaker
While I see that we need to do more research on the writing aspect of the copyright, this was a good video on how the U.S. Copyright Office is looking at this.
00:31:32
Speaker
And the U.S. Copyright Office has actually issued rules on generative AI that basically says that AI-generated artwork is not copyrightable. The artist still needs to make modifications to that artwork that is large enough modifications
00:32:03
Speaker
to show that the artist has contributed. So they're judging each piece, piece by piece, case by case, and saying, no, this artist hasn't contributed enough. Yes, this one has. So if you've generated artwork or if you've generated, say, a book at Sudowrite,
00:32:30
Speaker
And you've gone in and rewritten where it didn't work. That might still be copyrightable. But the base work that you created is not. So anything that was generated by the AI could still be used by anyone else. So in other words, they could put in the same prompt, get the exact same
00:33:00
Speaker
generative material and then they would have to change it for their purposes and with their words and with their art to make it copyrightable. So it's really important that we understand that if we generate a blog post, for example, and someone else ends up copying that blog post but we haven't changed it, there's nothing we can do about it.
00:33:32
Speaker
complicated. Yes. And not every country has even addressed this yet. So this is based on US copyright law. It's and I don't know what other countries are saying yet, because a lot of them haven't even fully addressed it yet. Yeah, because if you had AI generate a blog post for you,
00:33:59
Speaker
And you didn't change anything about it and you published it. It would, you know, it would look like you published it first, it's yours. And the next person who publishes something similar would look as though they had plagiarized, but if they just put the same prompts in and got the same result as you, then that's not plagiarism. And, and even if they out and out copied your blog post, because you didn't put any of your voice into that.
00:34:29
Speaker
It would not be considered plagiarism because it wasn't individual to you. And that's basically what the U.S. Copyright Office is basing their decisions on. That makes sense. It does make a lot of sense because it means that you can't claim the work that the computer did.
00:34:56
Speaker
Right. Right. And it was a really interesting video. I will link it in the show notes, which can be found on my website at ArcticEdits.com. Or if you want these show notes in your inbox, go to newsletter and sign up for the show show notes newsletter and you will get an email for every show with all of the links right there in front of you.
00:35:26
Speaker
which I don't talk about often enough. I didn't even know that. Yes, you can actually get them right in your inbox, which is really nice. But this was a really good video and really, I mean, he very specifically says, I'm not a lawyer. This is just what I've read and how I interpret it.
00:35:51
Speaker
And the very top comment was, I am an attorney and you have done a fabulous job of distilling down what the copyright law states, and how it's applied. So, it's really, even though he's talking about artwork, artwork versus writing, I think
00:36:12
Speaker
that it probably applies to both. And if I were to actually dig into U.S. copyright, I would probably find that it states the same. I just, I don't even know where to go to dig into it, to be honest. I need to contact Denise and Randy and see what Randy has to say on where I can find it. Yeah. Yes. Harkening back to our episode with, well, your episode with Randy, I wasn't there yet.
00:36:41
Speaker
Yeah, Randy is a really great resource. Denise is getting ready to have surgery though, so I held off even thinking about contacting them because everything is kind of wild and crazy because this is a pretty scary situation right now.
00:37:04
Speaker
without going into detail, this is a big deal. And I want them to be able to focus on her care and not think about me, if that makes sense. Other than knowing that I'm there for support, moral support. But do watch this video because I think it's really worth digging into more.
00:37:30
Speaker
So now that we've talked about what AI is, what we think it's good at, what we think it's bad at, what is our overall feeling about AI? What would we recommend?
00:37:42
Speaker
Well, as a tech editor, I would not recommend using AI to write pattern instructions at this time. It can fill in some of the basic elements, but it just doesn't have enough knowledge of how knitting works to be able to write something that actually works. But again, I think that's always evolving. I think it is
00:38:10
Speaker
constantly learning from the input that it gets so we may see you know in a year in five years who knows maybe it will start writing knitting patterns for us um but i definitely recommend playing around with some of these gpt interfaces to see which ones can be useful for your craft business because i learned about several that i'm i'm gonna use myself
00:38:33
Speaker
Yes, I agree. I definitely recommend playing with them and getting familiar with them. If you've been using Canva, go in and play with their new AI and figure out how it might benefit you and your marketing and your content.
00:38:57
Speaker
take the time and don't just take the time right now and say, okay, well, this works and that doesn't because like Nikki said, it's constantly evolving. So in a month from now, what it's good at may have improved and what it's bad at may have improved. We don't know how fast it will evolve.
00:39:23
Speaker
The more people that use it, the more it learns, the better it's supposed to become. And so we don't know. There was one other topic that will be in the interesting reading on the show notes that I really want to talk about. In 2019, there was a big deal for about a week of a study that MIT was doing
00:39:52
Speaker
where they were trying to build an AI where you could upload a photograph of a knitted item.
00:40:02
Speaker
And it would then write the pattern and send that pattern to a knitting machine, a commercial knitting machine, and it would knit the item and then you would receive it. So basically you would then get a custom made item. And for about a week,
00:40:23
Speaker
It was big news. And then it just kind of has disappeared. And I have a couple of thoughts about that because it's worth reading and listening to what they were trying to do. One, what happens to not copyright? That's not the word I'm looking for.
00:40:47
Speaker
Patents, for example, or, you know, it would allow a person to take a picture of a sweater at a high end boutique.
00:40:59
Speaker
upload that sweater and order it to their specifications and there's legal ramifications to that that I don't know that they were thinking about at the time and to how were they managing sizing and you know all of those aspects the math of it but they were trying and
00:41:26
Speaker
It's worth keeping an eye out to see if they ever succeed. And if so, what the legal ramifications become. Are we going to move even further away from the storefront and do all of our clothing shopping online?
00:41:50
Speaker
Yeah, I read about that one as well. And I thought it looked really interesting. Of course, it didn't seem to
00:42:03
Speaker
It didn't seem to be creating hand knitting patterns, which obviously is my personal interest because I'm not a machine knitter. But it was interesting to see how it could take the photo of the stitch pattern because we know that commercially knit garments can have cables and color work just like hand knitting.
00:42:32
Speaker
So that was really cool. But yeah, like you, I couldn't find any follow up from those few articles that came out at that time. I hunted and hunted and hunted and there was absolutely nothing past August 10th. I think it was of 2019.
00:42:49
Speaker
Yeah, so they just and so it was like, and now it's gone. So I don't know what's happened with this, but it was kind of interesting because they were fairly successful in creating some knitted items from this. And
00:43:10
Speaker
I know that the way commercial knitting patterns are written for these machines still takes a lot of human intervention. And they have to be both a knitter and a programmer basically to understand the process, which was rather interesting. So it really tweaked.
00:43:40
Speaker
How does this play out in the long run? Question for me.
00:43:49
Speaker
But that does follow what we learned from the Skynet project where they had to really get it down to very specific shorthand type inputs, much like what a machine would use rather than the more descriptive type patterns that hand knitters use.
00:44:13
Speaker
Exactly, exactly. And that was, it was specifically designed for commercial knitting machines, which are different from in-home knitting machines, which you can still find. They're a lot harder. Funny story, back in the 80s, I actually, well, up until 2009, I had a knitting machine. Ooh.
00:44:41
Speaker
And in the early 80s, I did custom one-of-a-kind designs for people with that knitting machine. That was kind of where I was at at that time. Because I was a hand knitter, but I also used the knitting machine. And knitting machines were huge. They had big conventions, just like you have fiber festivals that was nothing but knitting machine-related items.
00:45:09
Speaker
Oh, see, I was wondering if you meant huge, like popular or huge, like took up a lot of space in your home. Nope. They were, they were a big deal and they were very, very, um, popular. And so to have them completely disappear, I mean, there's no longer any manufacturer anywhere.
00:45:34
Speaker
They are, you cannot find them manufactured new, except for the circular ones, like the Addy Centro, I think it's called. Or there is a plastic one that's still made, and I can't remember who makes it, but they're really hard to find.
00:45:58
Speaker
And these were metal, and they came in different gauges, and you could get them with or without rivers. Mine was electronic, so it would do all your color lace patterns electronically. Wow.
00:46:19
Speaker
Cables still had to be hand manipulated, but everything else could be done electronically. I even had a special electronic
00:46:34
Speaker
I can't think of the term. Isn't that terrible? Anyway, it would sit on the main bed and it would do knit purl patterns all on the main bed instead of having to hand manipulate to do anything beyond rib. Rib could be done with the riverbed, but if you wanted a stitch pattern that used both knits and purls, like a seed stitch, for example, you had to use this electronic
00:47:01
Speaker
A piece that that ran itself back and forth on the bed and it manipulated the stitches because for every rib stitch it had to
00:47:13
Speaker
Um, manipulate that stitch three times. Hmm. That sounds like a lot of fun to play with though. I have one of those, I kind of miss it. Yeah. Yeah. I have, I have one of those centros, but it's very, very basic. It makes a tube. It does that. Well, that's about it. Yeah. It doesn't even really do the things that a good circular sock machine would do.
00:47:41
Speaker
Mm hmm. That's right. And they're not durable at all because they're completely plastic. Yeah, yeah, exactly. Anyway, so AI is is really fascinating, all the things that it does. I do have one question for you. In your research, you ran across TLDR and you even used it earlier. What does that mean? Oh, I'm so sorry. TLDR means too long, didn't read.
00:48:10
Speaker
It's kind of a current way of saying long story short. Sorry about that, Lisa. I'll bet you any one of my three kids would have known that. Yes, they would.
00:48:30
Speaker
I don't think of myself as being up on the times at all. I just turned 38, but... There's a big difference between 38 and 62. Oh, that's funny. I was like, what does she mean by this?
00:48:51
Speaker
I just, as I was referring to my speaking notes, I just noticed your little comment and I was like, oh shoot, I should have replied to that. All right. Well, this has been really an interesting topic to dig into. And I'm sure as time goes on, we'll be digging into it even more. The show notes, like I said, have a lot of links that we're not coming, commenting on in detail.
00:49:20
Speaker
with articles and things that we ran into. Like we mentioned, Jenna Sargent's tech editing of pattern generated with chat GPT and a chat GPT and generated crochet patterns and some of the other things that we ran across that really
00:49:45
Speaker
were just really interesting things, but maybe we didn't have the time to read in detail. So please either go to the show notes on my website or go and sign up for the newsletters so that you can get those in your inbox. And if you have a specific request to receive these show notes and miss them,
00:50:10
Speaker
do feel free to email me at nittesignedit.com at gmail.com and I will send you these in particular because there's a lot of links in here guys with stuff that's really worth reading. Nikki thank you so much. Thank you Lisa. I hope that you have a great weekend and
00:50:38
Speaker
I'll see you on Monday. Thank you. You too. Come back next time when Nikki and I talk with Paul Hezemeier, knitwear designer and theater costumer from Canada.
00:50:56
Speaker
Don't forget to like and subscribe wherever you listen and join the conversation in our Ravelry or Facebook groups. For technical editing, find Lisa at arcticedits.com and Nikki at handknitsandyuga.com.