Not a genius. This thing is called a monogram, the most basic logo design. Used mostly by couples, law firms and couples’ law firms.

You’re right, at that size the AI is not very concerning either.

Do your and your partner’s names both start with L?

Thanks, those give me ideas for making more operator logos like these (No AI but mostly CC0 (public domain) because my creative input is questionable, some are just tracing of scaled-down images with a few touch-ups; I’m not too concerned about sharing non-FOSS trademarks under a permissive licence at such low res)

In that way, yes, but the logos that feature small text (Groovy, Lua) didn’t turn out well at all.

Here’s the list of logos row by row so you don’t have to awkwardly ask what they are

Steam (not FOSS)*
“LL” (likely OP’s own design, not a FOSS project)*
Zigbee
Obsidian (not FOSS)*
Brave
Protonpass

Tailscale*
Home Assistant
Raspberry Pi (not open HW)*
Ubiquiti (not open HW)*
Android (not really FOSS)*
Signal

DigitalOcean (service)*
Ubuntu
Linux
Claude (not FOSS)*
Proxmox
Nextcloud
Jellyfin (rotated)

Trilium
Nginx
Tabby
Bash
Debian (rotated)
Docker

NodeJS
Python
HomeBox
XPipe
PiHole
Prometheus
Grafana

* not in gallery of printable sticker images

What is the “LL” monogram in the second one? I crudely recreated it to reverse-image-search but got no results.

Did you look closely at Jellyfin? The things in the dark are OK, they add flavor, which was an artistic choice − the problem is unrealistic silhouettes of people & animals in one mosaic piece. Also, it’s rotated on the laptop, and so is Debian.

The hand and face (bottom right) on Gimp is way more awful.

Looking closely at Perl, it seems there are photorealistic (hard to tell if AI) photos toned blue, abstract shapes that might be edited/vectorized photos, and in the bottom right there’s a bead necklace that’s so unshapely it’s an AI giveaway. Well spotted, it was indeed made using AI-generated images!

FOSS source is here.

The second “S” in “FOSS” is “software”. You did not publish software, just its output: bitmap assets needed to print the stickers. Thanks for CC-licensing your creative work but source would mean showing what’s under the hood. We don’t know how you sourced the images used in each triangular tile: generated to best correspond with AI? Matching pieces from Wikimedia Commons photos?

Edit: Look closely at Arch for example. It’s clearly just the logo placed in a hexagon, approximated by a mosaic of 24 triangles with AI images of differing quality. Is that snow or whipped cream? How can PCB traces be as blurry as watercolor and go nowhere? At least they’re topical: for Arch the prompt was probably “mountain OR architecture OR arch OR technology”.

Presumably, the process for each tile is this:

1. you make some direct artistic choices to create the base image (place the logo in a hexagon, choose a background color, add a border)
2. you make some indirect artistic choices: pick keywords/themes for the AI to use
3. you use a script to divide the hexagon into 24 triangles (presumably with “overscan”)
4. you use generative AI to stylize the triangles’ bitmaps according to thw keywords, perhaps regenerating bad output
5. you use a script to reassemble the image

To consider this open source, I’d expect you to at least post the scripts you used in steps 3 and 5. To consider this good open source, it should contain a guide detailing this process, best with examples. I’d expect the AI part will be “bring your own model” but you could tell which one you used and its settings.

The idea is creative and “human” enough for me not to condemn it. “FOSS” or not though, you should disclose use of AI, especially since you’re selling the printed stickers.

Edit: yup, it’s made of AI images

This is not an AI-generated pic, it’s a photo taken with a real camera. The logos, however, are hexagons divided into 24 triangles each, and these triangles contain often thematically related (e. g. lions for Brave) photos or photorealistic AI images (the info I found online does not state either way) cropped to best correspond to what the triangle would contain if it just had the original logo. Basically, that corner of the whale surrounded by white was taken from a face photo (or AI pic).

Yup. “Public” is not a useful qualifier as it could mean decompiled or leaked.

If there are other posters…

Renumbering characters during font minimization? I haven’t encountered that, it would break searching and copying.

Anyway, PDFs for example don’t even say whether a line of text is left, center or justified – they usually store the coordinates of the first character and then spacing to each subsequent one unless defined by the font.

And what if the document contains text boxes, or other Word objects? Well, the text is separate from the underlying rectangle (if there is one) and it’s up to the conversion tool to guess if it’s part of the main text layer.

Sorry, it’s really hard to edit PDFs. You might want to use Inkscape for editing the graphical parts. If you also need to edit paragraphs, I suggest recreating the document by pasting them into Word/LibreOffice, and importing any graphical shapes as SVGs (use Inkscape for the conversion, then you can try Word’s “Graphic > Convert to Shapes” feature).

Really, every software that outputs PDF should treat it as an export process, hopefully making it clearer that “saving as PDF” is visually lossless but structurally lossy and messy.

It could have been a top Lemmy app if open sourced…

Reddit as in the company? Of course, the first Reddit search result has been manually set to a “Stay away from Lemmy” post.

Is there a tool to automatically check partitions for excessive log files, caches or other junk? The root partition of a Linux box I have is 60 GiB and almost full, and XFCE will fail to start when there’s no space. I would use WinDirStat on Windows but the Linux alternatives can’t do the job properly because they scan by file tree and some subdirectories of / are on other partitions because of symlinks… I guess I could boot a live USB and mount my ext4 root partition but not the NTFS storage one but I’d rather avoid that.

Try integrating with OpenStreetMap Traces and Tapiriik for ease-of-use. Recommend running your own instance for the latter. Not necessarily for the minimum viable product but consider this into the future.

Good bike computers like Garmin’s allow GPX export so HW compatibility is there. It’s a few manual steps but you can make the process automatic for example by syncing your HW tracker to Tapiriik (15+ brands supported), which then can auto-download GPX files to your computer via Dropbox (or without Dropbox if you run it locally), and then you can auto-upload those to OSM with one of these scripts running on your machine.

The map is a community effort and the lack of social features, which caters to introverts, keeps focus on the end goal - an accurate map of the world. Other platforms are suitable for social activities and you can link to your OSM trace from there.

Yes, seeing the trace geometry only with no map is a letdown. That’s why I suggested the visualizer in another comment. It would certainly improve the shareability of traces.

OSM doesn’t produce any hardware. They are a wiki-based world mapping effort. In addition, they run a PNG tile provider (so you can embed their map on a website), an article wiki for how to edit the map etc. and the trace repository.

You can use OSM and record traces using various apps mentioned on their wiki.