Apples model to generate Gaussian splats from a single image. Takes about 30 seconds on an M1 Pro.
It falls apart once you move too much, but for a little side-wiggling or a second-eye view for VR, it's great. And looks a lot better than the old approach of depth map + vertex shaders that I use in https://github.com/combatwombat/tiefling. But ml-sharp has 2.6 GB weights, a bit too big to run in the browser.
An LLM step also works pretty well for diarization. You get a transcript with speaker-segmentation (with whisper and pyannote for example), SPEAKER_01 says at some point „Hi I’m Bob. And here’s Alice“, SPEAKER_02 says „Hi Bob“ and now the LLM can infer that SPEAKER_01 = Bob and SPEAKER_02 = Alice.
Very neat. I like the channel idea. There’s also https://random-video.com which shows one of 4.5 Billion YouTube videos randomly (or with a filter for view count, language, year).
Pretty fun to discover what’s out there, without being influenced by YouTubes algorithm. It’s based on the YT Archive project and some other sources, since YT sadly has no randomize function.
There seem to be a few fancy boat projects that turn out to just be renderings and NFT scams.
Like https://www.pangeosyacht.com/projects, an $8 Billion „Terayacht“ maybe launching in 2033. Until then they sell NFTs for a place in the virtual Unreal Engine version of this (also still to be built).
https://font2png.com to browse font-icons and export them as PNG, with background/foreground color. Usually you want SVG, but sometimes a PNG is better. I use it mostly to generate quick favicons. It was also fun to make it work completely client side with canvas.
https://github.com/combatwombat/rmdb imports the IMDb database (at least the limited .tsv files they provide) into MySQL so you can query it. List the highest rated horror movies of the 90s, genre distribution by year etc. I made that mostly to c̶h̶e̶a̶t̶ ̶o̶n̶ help with https://www.reddit.com/r/GuessTheMovie, with limited success. Still fun though to SQL-query over all movies ever made.
> Although terminal emulation has been largely supplanted by the Web for online access, Kermit software continues to play a role in other applications such as remote sensing and data collection, management and troubleshooting of networking and telecommunications equipment, back office work, cargo and inventory management, medical insurance claim submission, electronic funds transfer, and online filing of income tax returns. Kermit software is embedded in network routers and switches, in cell-phone towers, in medical diagnostic and monitoring equipment, even in cardiac pacemakers, not to mention the cash registers of quite a few big-name "big box" retailers. In 2002 Kermit flew on the International Space Station, and Kermit software is the communication method used by EM APEX ocean floats (left) supplying realtime data to hurricane researchers and trackers to this day (the hurricane project entered a new expanded phase in 2010 based on a new version of Embedded Kermit).
But I added MIDI keyboard support to the secret fart piano recently and thus Firefox throws up a scary warning. It's now disabled unless you access https://frt.rip#midi. But you can also use your computer keyboard to play/fart a little tune, see Web Inspector for instructions :)
Alright, the background can now have rounded corners with smoothing / squircles like in iOS, using Figmas algorithm. Thanks for the suggestion! → https://font2png.com
https://font2png.com to create PNGs from icon fonts like Font Awesome. SVGs are usually the way to go, but sometimes you need a PNG. A previous tool that did the same went away, so I built my own.
Also https://frt.rip, a whoopie cushion with spatial audio, because I had the domain lying around.
Both don't make money, but they also have no users besides myself so they can run on my cheap server for free basically.
> The scrolling issue in the Finder is due to the Finder's way of quirky overriding scroll events in this particular mode. I'm not sure how to fix it tbh
> Upload to the private server: I don't plan on monetizing the app and would gladly replace uploading to my server with something else. Which protocol would you prefer, FTP, SFTP, SSH?
Everyone has different setups, but SFTP and SSH with custom ports and private key auth should work. And ideally the option to copy a URL to the clipboard and/or show it. So I upload a file bla.png to folder /var/www/domain.com/images, it adds a timestamp or unique key to not overwrite files and returns https://domain.com/images/12345-bla.png. But that's just me, if others need different things they'll write you.
> I thought it works ok because I mostly use key shortcuts for everything, but for the mouse users it should be a nightmare. I plan on fixing it [...] It's always a balance between intuitiveness and efficiency for experienced users.
Yeah, power users like shortcuts and GUI people like to always see the options, but both are new users at some point. Even the future power user likes to see the best options first, before deciding to delve deeper.
> How can I pan a big screenshot with a mouse: there's also a Space+Drag trick
Oh nice. Also add a Pan tool to your sidebar :D
> I'm building the app mostly for myself and favor efficiency whenever possible.
I appreciate that. Building apps for oneself, with lots of "must-have" features missing, is freeing. Who needs pagination, I only ever have 1000 things :)
> But now that I have a decent user base, I guess I need to accept some compromises to improve experience for others.
Also, you're posting this app on Hacker News and you have a nicely crafted website for it, so people will treat it as a product.
But, just as Skitch doesn't have a sidebar and tons of options (which is why I like it, compared to https://app.prntscr.com/en/index.html for example), maybe disregard all feature requests like mine and just keep building it for yourself and see where it goes.
This is neat with some fun features. It's hard to make something that complements the pretty powerful built-in screenshot and measurement tools of macOS.
I take a lot of screenshots with annotations and use Cmd+Shift+4, Skitch and a custom script/app to upload images to my server. After using Shottr for the first time, some unordered feedback:
- The erase tool in addition to blur/mosaic is great. And non-destructive, like the other options.
- Multiple arrow styles, even curved! But, a more curved arrow style would be nice. I love the dynamic Skitch arrow [1], the main reason I still use it to be honest.
- Drag&Drop an image quickly to save it somewhere, nice. But it's not as discoverable as Skitch, with its prominent bottom tab. Took me a bit to find the option, after mousing over the toolbar buttons. And apparently you can just Ctrl+Drag on the image itself to drag the file, interesting.
- Scrolling capture is fun. There are other apps/extensions that capture full web pages, but this does arbitrary capture of anything that is scrollable. Although it doesn't quite work in a multi column Finder [2]
- Undo/Redo would be nice. I guess it's planned or buggy, because the Undo menu item is there but disabled?
- No upload to my own private server. Skitch does that, but doesn't support private key auth (thus my custom upload script/app). I guess you plan to monetize this by offering your own image hosting service? I would pay some bucks for a good upload function to my own server, with various options for auth, file names etc.
- In general, trying to put all the options in the toolbar might not be the best way. You have to hover over buttons to see their function. Buttons have a second level once you do something, so to get from the erase-tool-settings back to adding a line, I have to click the "back" button, then the line button instead of just an always visible line button. There will probably be more features in the future, with more options. Maybe it's better to add a sidebar and have more room to explicitly show the options and grow functionality.
- Always-on color selector and pressing Tab to copy the value is good. Although it doesn't save time compared to the built-in Digital Color Meter, which works without taking a screenshot.
- The size of a thing should change as you're dragging the handle, not only after you're finished [3]
- Dragging an image on the Dock icon or the "Welcome to shottr" window doesn't work. To edit a screenshot taken with Cmd+Shift+4 I would have to go to the shottr menu icon > ... more > Open File?
- How can I pan a big screenshot with a mouse? There are no scrollbars and Cmd/Option/Ctrl/Shift-mousewheel doesn't work, neither does dragging. Edit: Right mouse button...
tl,dr: Cool stuff, but the workflow and discoverability should be optimized. I'll keep using this, if only for the measuring tool and the curved arrows :) How about some Catmull-Rom arrows that follow a path?
