So this came with the 2.90.0 => 2.91.0 release.
The opt-out now it works. Thanks!
Posts by Torsten Curdt
hmmm ... for me that gives
! warning: 'telemetry' is not a known configuration key
GitHub appears to have opted anyone using the CLI into sending telemetry they will use to inform product decisions. This is sneaky and should have been an opt-in decision, not opt-out.
Disable it with `gh config set telemetry disabled`.
cli.github.com/telemetry
We all have days like this
I know this is more of a serious topic
...but I just had to really lol on watching the video :)
Two papers came out last week that suggest classical asymmetric cryptography might indeed be broken by quantum computers in just a few years.
That means we need to ship post-quantum crypto now, with the tools we have: ML-KEM and ML-DSA. I didn't think PQ auth was so urgent until recently.
Totally agree. This really isn't about AI as such. It's more how companies view their customers. AI will just amplify this.
Support can be seen purely as a cost center - or as a feature.
It's amazing what good(!) support can do to the customer relationship - and by that extend to the LTV.
I am not sure that's me just being annoyed, but here is a prediction:
Having a good human support team might become an even stronger factor for the success of a company than it was before AI.
Remind me in 3 years.
Either way - thanks for the insightful responses.
I'll see if I can give it a try with SvelteKit.
Thanks for the release and the help! ❤️
...although - that negates any possible offline functionality. But that was never on the roadmap for zero anyway, was it?
UserA vs UserB wasn't really meant as computer sharing. UserB is more the adversary that is after data at rest. It just isn't something you have to think about in a classic web app - at least not on the client side. IMO there is a difference. But again the memory option covers that.
Well, it is a very different scenario when your bank transactions linger in some IndexDB. But if memory (or session storage) are options that certainly changes the picture. So thanks for pointing that option out.
A session storage based IndexDB or client side encryption is the only way I see as a possible way to trust such apps.
Is something like that planed or already possible?
scenario1:
UserA logs in, data gets sync'ed to IndexDB, Closes
UserB opens the dev tools and queries IndexDB to find data sync'ed from UserA
scenario2:
UserA logs in, data gets sync'ed to IndexDB
I open ~/Library/Application Support/Google/Chrome/Default/IndexedDB/ and dig through the LevelDB.
But you need strong column based security - not just row based. And that still does not cover some other scenarios as this means raw data instead of rendered data makes it to the client.
You did it! 🥳
European Parliament just decided that Chat Control 1.0 must stop.
This means on April 6, 2026, Gmail, LinkedIn, Microsoft and other Big Techs must stop scanning your private messages in the EU. #PrivacyWins 💪
First off - congrats.
But whenever I looked at zero, I am puzzled by the security story:
What gets synced to the browser? (not all in the db should be shared to the client)
And how is the local data secured? (because a userId isn't really a security boundary in IndexDB)
Everyone wishes "fair winds" - I wish you a cheap and short refit ;-)
Did you check the bulkheads yet?
Aren't specs like that micro managing? ;-)
In fact, it's good to treat the spec as a planning conversation. You iterate until the model moves in the right direction and it starts to make sense (to you). That does not mean all details need to be spelled out.
The first thing I saw on this pictures was a boat project - the rust from the screws. Next was, the question "what chart plotter software is this?" THEN I saw the drink.
I am not sure what tells about me.
...but cheers :)
I agree that detailed specs are not far from code - which is why vibe coding is always destined to hit a wall ...but a good spec can get you a boost getting to the first 80%.
Sure, we know the last 20% take 80% of the time - but I would not be this dismissive.
ASF members meeting over.
So many old friends that I haven't seen in way too long.
Congrats to the new members.
Congrats to the new board.
This sounds like the proper tool for what I build a ugly/hacky/shitty version of back in the days at joost.
Basically defining queries, streaming data through and watch the query results change.
We had the batch layer but also wanted some real-time insights.
Tempting to play with.
He could ... but I fear all we get is another Puppet Regime video ... which are all great ... but ... no war would be better.
www.youtube.com/watch?v=g-4O...
Fun - but I fail to see a good use case. Got one to share?
Wow - even if you call them again and they "fix it".
Two calls on your time to fix their mistake. Uff.
Eine handvoll Nerds bereitet die Epstein-Files gerade in Form eines Gmail-Accounts auf, durchsuchbar, nach Themen und Kontakten sortierbar etc. und es ist unfuckingfassbar: jmail.world
Apple not supporting a login to the AppStore on a virtual machine is such a pain. It would be so nice for testing.
Why Apple? Why do you hate us developers so much?
github.com/utmapp/UTM/i...
Is that a slap for using it or for not maintaining it?
Or maybe even for the general ingress story?