tethys

Slightly off-topic: Refreshing to see these two authors link to their Bluesky and Mastodon profiles. No Twitter/X in sight!

It may be free, but it cannot be used without credits.

  Error: {"type":"error","error":{"type":"invalid_request_error","message":"Your credit balance is too low to access the Anthropic API. Please go to Plans & Billing to upgrade or purchase credits."},"request_id":"req_011CaGaBf6uTHfbmdZ39nx1Z"}

When I meet my friends and tell them about my last vacation instead of my work, how I slept, and how often I brush my teeth, is that also “manipulation”?

It works with prepaid credits, so there should be no risk. Minimum credit amount is $10, though.

Hard to believe when the page literally says

Create a viral Reel in just 2 clicks

If there's any others you'd like included, let me know!

Just this morning I came across HunyuanOCR which sounded very promising. https://huggingface.co/tencent/HunyuanOCR

This incident has been resolved. Posted 4 minutes ago. Nov 18, 2025 - 19:28 UTC

Better link for chroniclers, since the incident is now buried pretty far down on the status page: https://www.cloudflarestatus.com/incidents/8gmgl950y3h7

Official announcement (that gets straight to the point): https://www.manton.org/2025/11/11/microblog-studio.html

In addition, I can wholeheartedly recommend this video tutorial that guided me through setting up Coolify for the first time: https://www.youtube.com/watch?v=taJlPG82Ucw

Been running this setup for about a year now, and it's the first time I am actually self-hosting and feeling fairly confident about it.

As for why they are closing:

Fixed costs and a lack of financial support are forcing us to take this step. In addition, the general cultural situation in Berlin is very precarious. It was a very difficult decision for us.

Via Deepl, original here: https://www.instagram.com/p/DLuAW5DIANV/

I don't understand this trend of translating UI elements when 100% of the content is English. Who does that help?

I am not so sure. It currently highlights a story from Munich, and in addition to a few factual errors, the information is simply outdated; there have been numerous new relevant developments. (I also don't understand the selection of sources. Aljazeera? rt.com? South China Morning Post? As if there weren't enough sources of original reporting right from Germany.)

I would agree that a single daily news update is useful (and healthy), but this must also be reflected in the choice of topics and the type of reporting.

To keep up with updates, follow us on Instagram.

If only there was a protocol that would allow me to subscribe to new entries published on a website …

Fair point! My eyes are (mostly) fine, but even I would have a hard time telling these colors apart when used in the same chart.

I usually do very simple charts, using maybe 2 or 3 colors, and with this palette I feel the results are typically very pleasing, whichever colors I end up selecting.

Whenever I need colors for a chart, I type "12 bit" in my address bar and pray my browser hasn't forgotten the URL: https://iamkate.com/data/12-bit-rainbow/

The post (and this clarification [0]) state that you have to either long click or select "Preview Link" from the context menu.

[0] https://connect.mozilla.org/t5/discussions/try-link-previews...

if treatment significantly reduces the probability of a pilot's committing suicide

Psychotherapy significantly reduces the risk of suicide. https://pmc.ncbi.nlm.nih.gov/articles/PMC6389707/

No. You have to turn your head, and stuff. Also, even if this would work, they allow only one verification per person per 90 days.

I've generated (and downloaded) a couple of images. All filenames start with `DALL·E`, so I guess that's a safe way to tell how the images were generated.

That's not correct and also not what the article says. They are only talking about Excel and PowerPoint.

"We're working on adding the Enable Copilot checkbox to Excel, OneNote, and PowerPoint on Windows devices and to Excel and PowerPoint on Mac devices."

I am using Word on my Mac (version 16.93) and do have a checkbox that disables Copilot.

For comparison, this is what other models produce: https://github.com/simonw/pelican-bicycle/blob/main/README.m...

That's unfortunate! I would immediately pitch this tool to my SEO colleagues if it had a (fairly priced) API :)

Do you offer an API? My issue with most keyword tools is that their APIs suck, with tight limits that force you to only request data in bulk, and so on.

Can't tell if the guy, or the background, or both are AI generated.

Instead, the company told TechCrunch simply that Spotify’s features can vary over time, between markets and across devices.

I hate it. I recently tried to find out how to sort my liked songs by genre. Turns it, this feature has been released over 3 years ago [0], but hasn't made it to Europe yet.

[0] https://newsroom.spotify.com/2021-02-25/how-to-sort-your-fav...

As long as you use Safari for browsing, Notes for note taking, iCloud for mail …

Just a side note: It bothers me a bit that most audio player interfaces make it unnecessarily difficult to navigate within a title. This here is a good example: The waveform extends across the entire width of the window and could be wonderfully used to jump precisely to the desired spot (Soundcloud does this comparatively well). Instead, I only have an extremely narrow bar here, where a few pixels of movement mean a jump of several seconds.

their reviews are complete black boxes

They are certainly not "complete black boxes". This page [0], for example, explains in detail their test procedure for vacuum cleaners.

Just an excerpt on durability testing (via deepl):

Impact test (500 cycles): The hoovers drive over thresholds 10 000 times and bump into posts 1 000 times.

Drop test: The suction nozzle must survive 1 200 drops from a height of 80 centimetres.

Cable pull-out test: 6 000 pulls.

Test of hose attachments: 40 000 swings with the connection nozzle clamped.

Squeezing the pipes, hoses and auxiliary air adjusters: The testers load them with about 70 kilograms for ten seconds.

[0] https://www.test.de/Staubsauger-im-Test-1838262-1838266/

This Statamic blog post has (at least some) insights: https://statamic.com/blog/statamic-at-spiegel-scale-2

So how does SPIEGEL do it?

Curious about how SPIEGEL’s stack works? Here’s their high level approach, keeping in mind this is running on v2. Upgrading to v3 streamlines even more of their stack.

A huge flat file content store organized with each entry inside its own subdirectory (you can’t have more than 10k files inside a single directory, an operating system limitation)

The content store is using a cloud storage solution.

Statamic’s control panel pushes entries into the content store, and then into an ElasticSearch instance via message queue.

ElasticSearch runs as a content API with blazing fast response times and is consumed by the control panel and front-end

The front-end is built in Go.

User accounts are connected with an Office 365 Active Directory integration with OAuth.

I think he was supposed to fix search.