In past instances similar to this one, when Heroku wasn't acknowledging or giving any details it was a case of DDoS. It makes sense to stay quiet to avoid other potential attackers to pile in. Maybe they have found an AWS endpoint that was suffering DDoS attacks and they brought down a lot of people with it. Heroku is surely on AWS, I'm positive that also OpenAI is using something there and all the others affected. Let's wait and see, it will take a few days for sure.

From Datocms here :) We are also actively experimenting on porting our main app. We are doing progress, so far so good!

The Nextjs site lasted 4 years. Before that we had a Gatsby one: https://github.com/datocms/old-website/

A 4-year lifespan for a marketing website doesn't seem so bad

I'm helping our local Fablab to manage physical access with a series of ESP8266 and esp-rfid https://github.com/esprfid/esp-rfid/ (of which I became maintainer. If you want to use it as well I can help!)

I would pay real money for a system like this with actual security. The obvious starting point would be CTAP2 — the protocol is open, high quality fobs are inexpensive (not as cheap as Mifare, though) and widely available from multiple sources, and the protocol has been analyzed for real. One could probably even extract an actual production grade implementation of the NFC side from the Android sources. Apple Home Key support would be nifty, too. PIV would be another credible choice.

Extra bonus points for support for real commercial readers using OSDP’s transparent mode or whatever they call it these days. As I understand it, an early standard involved a horrible hack that was so horrible that HID managed to patent it, but the protocol was redone to avoid being a horrible hack, and the new version is also unencumbered. Although maybe the spec costs $30.

Working in a SaaS, I'm always surprised by these companies giving two months notice before shutting down services that are powering pretty significant websites.

This move is going to cost companies using Gatsby Cloud a lot, especially if they had built tooling and workflows around Gatsby Cloud.

Why are they rushing this shutdown? 2 months seems really too little for me.

Also, Netlify is targeting enterprise customers more now, doesn't this two months notice give a bad impression to those kind of customers?

I'm honestly not sure if I am too scared by these quick moves or if it's going to be a real problem for someone.

Agreed. Unimpressed by netlify’s migration guide, which missed things like migrating headers, whether or not cms preview builds are offered, and whether or not there is incremental build support. It feels like a rushed document smothered in corporate platitudes and dns instructions.

You mean openai decided

Is this Ok to upload all your messages and your contacts to these services? I'm very worried about that, really

I’d assume it has exactly the consequences you think it does, but I don’t know that and it becomes more of an unknown as more third parties are layered between you and the LLM.

I’d assume the right prompt by another user of the same underlying LLM on another platform could well expose your private information / content / passwords and would treat this with a lot of caution until you get satisfactory evidence otherwise.

Yes, that's what concerns me. Since it's not really clear what's going on, how can we trust a third party service by uploading personal data?

Very useful guide for proper parenting. Thank you!

I'm very interested in visual ways to explore ebooks, papers, articles.

A physical library helps discovering new stuff or occasionally bumping in something. With files you need to be much more deliberate and I feel like you can accumulate a lot without really knowing what you have.

Even if this is a simple visualization it already gives ideas on how much more information you could convey with something similar. Like number of pages, color of the cover, to remind you of the book, different font, etc. A set of cues that can help you navigate in the list of books, at least I suspect :)

I would love a way to flexibly view and move around lots of PDFs at different scales, to simulate piles on a desk or the floor. For some purposes, something is lost by constraining organization to a regular grid or table.

When you've got thousands of books, you lose track of what you've got, just like with files.

It's not necessary to ask how I know this :-)

Sure you lose track, but it's so much easier with physical things to bump into already seen items and recognize them. Or also casually being interested by a cover or by the size of the books.

I feel like there could be a better way to handle lots of digital books/articles/songs/films/photos/etc

Have your lock screen show a random book cover from your collection!

I'm intrigued by the idea of "bumping into something". Totally agree! How would you describe for a digital environment if it's not a pure recommendation algorithm based on similar books?

I was thinking to a 3D representation of a library or a zoomable interface. It could leverage our good spacial memory while also allowing accidental discovery of books

check this coworking space: https://toolboxcoworking.com/ there are several tech meetups of nice people and it's also hosting the local fablab: https://fablabtorino.org/

after covid there are less meetups, but I'm sure it's going to pick up again