Was writing “Airbus” in a message and mistakenly typed a P instead of a B. Found the results amusing, so I fired up ChatGPT’s image generator:

Okay, I too tried out ChatGPT’s new image generator, mainly it’s graphics design capability. Got it to generate a logo for our bocce club. I’m impressed by the results: it’s pretty much what I imagining.

It lives!

I know I’m late to the party, and that this has been reviewed already by those that write about Apple software and anyone with an iPhones. Nevertheless, I’ve upgraded my Mac to the version of MacOS that has Apple AI, and along with it came Image Playgrounds. I thought I’d give Image Playground a go.

Here’s my first attempt. May I present to you: generic white technologist in a bow tie using a computer with a confused expression on their face:

It took me a little while to understand how to compose the image. I knew that this was going to be a very on-rails experience, but I was still expecting that prompts that would work for DALL-E would work here too. But when I tried the prompt “a technologist using a computer with a confused expression on their face,” I just got back an error. Eventually, I figured out that I needed to build up the image by adding additional clarifiers, which would floating around a pulsating orb while the image was being generated. This was a little more on-rails than I was expecting but I guess it’s a way for Apple to make this approachable for those not use to using ChatGPT, along with controlling the suitability of the output.

As for the image itself, all I could say is that it’s… okay. A little bit on the cartoonish side, but I’m guessing Apple prefers that over something with a high degree of realism. The subject looks nothing like me but that’s by design: I dared not use myself as the subject, given how bad the preview photo I assume it was going to use was. I am a little disappointed by how expressionless the generated people are. I’m not sure how well the underlying model actually understands expression like “confused” given that most of the suggested images had a neutral expression, with one exhibiting something that I can only describe as “surprise and delight.” This is the best one I could find.

Finally, it seems the developers of this app has never heard of the file system. I could share the generated image to a bunch of Apple apps like Mail and Notes, but I couldn’t find a way to save it as an image file. If there is a way, it’s not obvious: I checked the toolbar, menu bar and context menu. Copy to clipboard was an option so I ended up copying it to Acorn and saving it as a JPEG there. This is definitely something that would get any technologist confused.

So yeah, I don’t see this being a regular part of my workflow. It’s better than I expected, but this is definitely something that Apple has built for people who have not used DALL-E or any of the other AI image generators out there (and, I suspect, to appease their shareholders). For myself, I’ll stick with what I’m using now.

I’ll close this with a gallery of a few other images I’ve tried:

In lieu of anything else, I thought I’d put together a gallery of the logos I use for test organisations. I occasionally need to create fake organisations for testing at work, and to add a bit of amusement to the mix, I sometimes make fake logo for them. These logos are typically made using ChatGPT, although if I’m particularly bored, I sometimes touched them up by hand (nothing too drastic, usually things like cropping or adding titles). Most of the fake organisation are film and media production studios, as that’s typically the client base of our product.

I do this mainly for fun, but there is some utility to it. A user can be a member of zero or more organisations, and can change the one they’re acting in at any time. Having a unique avatar for each one helps in distinguishing which one I have active. I do get cute with the names, and for that, I make no apologies. 🙂

Two weeks ago, I tried that GenChess thing that Google Labs released. When prompted, I asked for a chess set to be made with pieces resembling Australian birds. What it produced was a little underwhelming, but I was curious to know what Gemini itself could produce when prompted for a single piece: an image of a pawn shaped like a cockatiel.

Here’s the prompt:

Please generate a photo realistic image of a 3D pawn chess piece which looks like a cockatiel. The piece must look like it is made with wood. The image must be generated with a white background and with a parallel projection.

And here’s the result:

Not bad. Certainly better than what I could produce. And it’d be fun playing with such a set should one actually existed¹.

It got me thinking as to what birds I’d choose if such a chess set were made in real life. My current idea is cockatiels as pawns, emus for kings (large and mobile, and important enough for such a role) and maybe cassowary as queens (also large and mobile, and actually quite dangerous in real life). I haven’t got positions for the other roles just yet.

One other thing I did was finally address how galleries were being rendered in RSS. If you use the Glightbox plugin, the titles and descriptions get stripped from the RSS. Or at least it does in Feedbin, where all the JavaScript gets remove and, thus, Glightbox doesn’t get a chance to initialise. I’m guessing the vast majority of RSS readers out there do likewise.

So I added an alternate shortcode template format which wraps the gallery image in a figure tag, and adds a figcaption containing the title or description if one exists. This means the gallery images get rendered as normal images in RSS. But I think this sacrifice is worth it if it means that titles and descriptions are preserved. I, for one, usually add description to gallery images, and it saddens me to see that those viewing the gallery in an RSS reader don’t get these.

Here’s the shortcode template in full, if anyone else is interested in adding this:

<!-- layouts/shortcodes/glightbox.xml -->

<figure>
<img src="{{ .Get "src" }}"
     {{ if .Get "alt" }}
        alt="{{ .Get "alt" }}" 
     {{ end }}
     {{ if .Get "title" }}
        title={{ .Get "title" }} 
     {{ end }} />
{{ if or (.Get "title") (.Get "description") }}
  <figcaption>
  {{ if .Get "title" }}
    <strong>{{ .Get "title" }} {{ if .Get "description" }} — {{end}}</strong>
  {{ end }}
  {{ .Get "description" | default "" }}
  </figcaption>
{{end}}
</figure>

I’ve also raised it as an MR to the plugin itself. Hopefully it get’s merged and then it’s just a matter of updating the plugin.

Removing personal identifying information from logs is a laudable goal, but it does make troubleshooting issues in prod rather difficult.

I finished my experiment with htmgo, building the worlds most inefficient world clock. It uses HTMX swapping to get the time from the server every second.

It’s an interesting framework. Not sure it’s fully ready yet (you can’t change the bind port, for example) but might be useful in the future.

I was poking around Dave Winer’s Software Snacks — a brilliant name for those — and I stumbled across Little Card Editor. Decided to give it a try.