kmfrk

Seems these days that the most marketable threat modelling for your product is against the company's own CEO.

I also suspect that the quality of transcription depends on the popularity of the video. So maybe it's decent for videos with 000,000s of views. Same way the video codec is changed from AVC to VP9 - and occasionally AV1 nowadays.

I recently redeemed three months of free Apple TV+, and it drives me crazy to get a preroll of other TV+ movies and TV shows I couldn't care less about.

I like to think it's only going to be there while I'm freeloading on the three-month offer, but all I know is that I don't get this with Netflix.

So far, I've had a hard time setting up an iCloud passkey on desktop Windows. Windows Hello keeps intercepting most attempts at a passkey, and 1Password will usually step in otherwise. The UX around it continues to be baffling to me.

At least there's a dedicated migration protocol for passkeys, but I'm not super stoked about how basic Apple's password management is, and how few fields there are for each profile.

But I also feel like it's time to retire my YubiKeys ...

I can't tell if this is about specific subreddits or Reddit itself. Subreddits often have to implement some threshold for karma or activity within the subreddit because of brigading - or these days, botted promotion.

I thought this was largely not possible anymore after the APIpocalypse, but maybe Reddit implemented some tools of their own that are either more opaque in their rules and requirements, or more prone to false positives than their third-party counterparts?

Anil Dash can probably call someone he knows if his account is shut down for upsetting the moderation algorithm. For people whose careers and livelihoods depend on their online presence, it's easier said than done for people with that kind of privilege. People losing access to their Instagram accounts with no option of appeal or support is the norm.

Social media is a big part of people's lives, and getting your account canned or shadowbanned might not be the most useful form of activism. People try different things instead of just the one. When nothing works to stop and hold Big Tech accountable, what really counts as actual activism? Sure, some people act way more oppressed than they are compared to others, but the "lingo" at least implies an awareness of the censorship. Some of these attempts at special language or emoji is too absurd to the point of being offensive like the grapes and overly coded language to the point where it trivializes the severity of the issue.

Some people make money by using social media. How do you talk about real issues without being at risk of effectively ending your primary source of income? Talking about "politics" on Twitter can get your content reclassified as "Political", which greatly diminishes the advertising on your channel: https://safety.twitch.tv/s/article/Update-on-CCL-for-Politic.... Political and advertorial censorship are similar, but separate things, both of which have major chilling effects. This isn't to excuse influencers etc from using their platforms responsibly, but let's at least understand their daily reality.

This might be a hot take, but I also think part of the problem is that the adults also need to realize that we fundamentally just need to be less online. Instead of coming up with "Twitter alternatives", it'd be better for most to just hand their kids an analogue diary to type all their cringy thoughts into so that only they will know the embarrassing stuff they would say when they didn't know any better.

After people getting pilloried for social media posts from twenty years ago, I really hope (sensible) people will have the wherewithal to think twice about what they hand over to their chatbots.

Trying to download all my 5GB of files from the web client gives me an "Attempted to zip too many files."

I guess the only way to get or export your data is through the desktop client?

Gonna need these people to at least prefix their prompts with "You are Edward Tufte and have an allergy to chartjunk".

Third-party calls won't eat into Apple's compute, so it's not entirely against the interests of Apple as they put caps on people's AI use.

I wonder what the margin on this stuff is for them:

    Some Apple Intelligence features in the Home app, such as
    video summaries, are available for a limited number of
    cameras depending on your iCloud+ plan — up to 1 camera
    in the 2 TB plan, up to 2 cameras in the 6 TB plan, and
    up to 5 cameras in the 12 TB plan. Supported camera
    counts and other iCloud+ plan features may change over time.

https://support.apple.com/en-us/127901

Sony TVs are handled by TCL now, so time will tell how Sony-branded models act (up) in the future.

https://www.tomsguide.com/tvs/sony-just-handed-its-tv-busine...

I like Kagi, but is it me or have their results been less than great for the past 1-3mo? Maybe the places they're pulling their results from have had some changes?

The AGI rhetoric in Big Tech reminds of Paul Samuelson who wrote of Wall Street indexes predicting "nine out of the last five recessions".

https://quoteinvestigator.com/2023/11/01/predict-nine/

Parents got one years ago, and to date, it is the loudest household item I've ever heard. Never trusted commercials without sound after that.

That looks amazing! I've been thinking to move to Kobo as well with iBooks feeling like abandonware, and it's also been interesting to see community projects like Nico Verbruggen's typesetting fixes for ebooks[1][2].

Honestly, seeing your fixes and his, makes me more excited about potentially getting one with all the tuning possible.

Time to wait and see if anything new gets announced at IFA.

[1]: https://nicoverbruggen.be/blog/kobo-typesetting-fixes

[2]: https://www.youtube.com/@NicoVerbruggen/videos

What format would people generally prefer for shorturls if they were designing a website for readability, maintenance, and SEO?

foo.website.com/{initiative} or website.com/foo/{initiative}? And are there any optimal names for foo that also make the scheme more intelligible to the average person?

Shame, because it's genuinely a great map - ahem, optimizations aside.

I remember giving the app a second chance - after assuming it to be bloatware next to all the other default apps that came with Windows - but after seeing it redeem itself in benchmarks[1], I gave it a shot and quite like it. The iOS app was too rubbish to bother with, though.

I'm curious what could options there are for decent predictions. Especially with Dark Sky never having been much of an option for Europe. I see some of the people behind it have a new subscription-based iOS app to try, though: https://acmeweather.com/app.

Do people have any desktop weather apps that strike a nice balance between decent code and actually useful predictions? Future weather's likely to be anything but boring, unfortunately.

[1]: https://forecastwatch.com/awards/2026/

As a European and Fastmail user, this is great news.

One thing to flag from Europe and GDPR is all the malicious compliance of companies and governments using the law as an excuse to not meet some basic obligations to users like storing and retrieving useful information, or getting pretty basic customer support. It's very reminiscent of companies deliberately making cookie warnings annoying instead of protecting user privacy. Who knows, maybe this will be used to obstruct FOIA too.

Because of this, be sure to pay attention to companies getting too clever in interpreting this law, even after it's passed. It ain't over 'til it's over, so be sure to keep fighting it until we actually get the desired outcome, and track the actual compliance with the law.

The law itself sounds great, just remember that people often aren't.

I understand why forums don't expose user IDs for scraping, but man it would be nice to just be able to tag users online with accounts created after the introduction of ChatGPT.

People have already tried this with C2PA, and it's not a perfect system. The main problem with these systems is that you risk certifying fake media, which is worse than uncertified fake and real media.

https://news.ycombinator.com/item?id=40313769

I do wish more people used basic watermarks on their media regardless - as in sloppy ffmpeg text. People are more likely to repurpose other people's older media than generate new media, and old time-y watermarks make that just slightly harder. Maybe a free app for uploading citizen journalism media that handles the watermarking with geo- and datemarking is the way to go.

This would also work amazingly with https://beliapp.com which is a tool for - only - sharing restaurant reviews with your contacts to serve as a personal graph of trust.

Or you could also implement your own on-site equivalent of it, if people are sharing their reviews anyway.

It's also excellent for vlogger content, if you want to get the word out, although influencers aren't historically known for tenacious attribution and credit in their content. :)

I forget the name, but there's also a company that does physical mockup of your house's floor plan that you can walk through, kind of like Von Trier's Dogville, but at least with some walls.

There's definitely a perception gap that needs to be closed for these things, whether through VR or something slightly less digital. That's probably also a big part of the appeal of prefabs.

Definitely. I think the appeal to some isn't GenAI per se but image-to-image generation like "make this item look like a talented photographer shot it".

But I can't imagine that your conversion wouldn't benefit by showing actual imperfect photos of your products, especially when you run a platform with 1k new listings per month. I'm personally turned off by something that looks like 3D, especially when all the different colour variations look the exact same.

It's not great as a Mullvad user: YouTube constantly complains and also shadowbans some of my livestreaming functionality, and my X account has now been flagged as a bot, which requires me to download the app and verify myself in ways that they won't explain. And X is practically useless without a user, much like Instagram. TikTok also shadowbans you with reduced functionality.

Twitch and Discord meanwhile feel remarkably unaffected compared to their counterparts.

VPNs in the aftermath of FireSheep[1] before we got proper HTTPS, but it's not likely to be the ban or censorship circumvention tool we think they are.

[1]: https://en.wikipedia.org/wiki/Firesheep

Swedish politics aside(!), I'd love to have other VPN services as exit points for Tailscale. More choice is good, and hopefully that becomes an option one day.

People also generally use Violentmonkey (https://violentmonkey.github.io) instead nowadays.