Some years ago I've used the Lottie (Bodymovin?) library. It worked great and had a nice integration: you compose your animation in Adobe After Effects, export it to an svg plus some json, and the lottie JS script would handle the animation for you. Anything else with (vector, web) animations I've tried is missing the tools or the DX for me to adopt. Curious to hear if there are more things like this.
I'm not sure about the tools and DX around animated PNGs. Is that a thing?
These are some screenshots of NeWS 1.1 running on SunOS 4.1.4 on QEMU that I took time ago.
Unfortunately QEMU SPARC decided to break again. Also, keyboard doesn't work yet (forcing me to use the SunView terminal emulator), but probably it's due to the keyboard script.
In any case, SunOS 4.1.4 comes with OpenWindows out of the box.
If you visit the site without the subdirectory [0], you can see a message upload interface. The handwriting is supplied by the user (may or may not be AI-generated); the drawing movements are generated with some (hopefully non-AI) algorithm.
I am not sure about the US market, but Skyscanner [1] works great in Europe, I wonder how this is different. Besides from being structured as an email, it offers quite similar functionality of finding optimal flights (in some sense of optimal). It may just be me, but I am definitely misunderstanding the inbox metaphor. Why is it an inbox? Why is there emphasis on _one email_? I don't book flights through email -- I go on an airline's website to book.
For now it seems to be always responding with "Hi there, I couldn't find any flights for your request. Please try again.", I could not test it out. Unless this is a demo version, and is working as expected.
I get results everytime. Maybe not the cheapest. But it's something, and it's fast...
Very fast. (Compared to going to a website.)
With a very simple sentence full of acronyms that we all use buy now, it is a fun experimental tool. (Then clicking through the Google link, you can actually continue and book.)
But, I agree, Skyscanner is awesome.
In this article, the meaning of _pen_ as in an enclosure, a cage, a tray, a box, a container [0]. A holding pen in this case means a holding tray, a holding box, i.e., a container for holding things. A pen (cage, tray) for holding things.
I agree that it's a very difficult problem. I'd like to mention AlphaDev [0], an RL algorithm that builds other algorithms, there they combined the measure of correctness and a measure of algorithm speed (latency) to get the reward. But the algorithms they built were super small (e.g., sorting just three numbers), therefore they could measure correctness using all input combinations. It is still unclear how to scale this to larger problems.
Talking about physical devices, I see you mention tablets in public places. Would you make them eye-catching, so people come up to them, or hidden, to be spooked by someone's stare?
There is an interesting project in Vilnius & Lublin that just shows a livestream on what's on the other side (it's still up today) https://www.theverge.com/2021/5/30/22460964
For the tablets I was picturing things that were pretty clearly labeled - you'd know what you're getting into. Like an exhibit or an installation (but in many places, not one). I don't think I'd want people to be surprised - I'd want them to get onto the call knowing that was the only time they'd see that person.
I love that installation! My old office had a (much less elegant) setup like this that linked some of the social / kitcheny spaces across our offices and I liked it a lot.
I'm not sure about the tools and DX around animated PNGs. Is that a thing?