I've seen/noticed this simply from being on a low carb (aka KETO) diet.

Besides AI grossly over/under estimating values even when you give it a photo of the packaging with nutritional table and tell it weight you used.

The other thing that surprised me, at least until I read up on how LLMs are actually working. Was how it would confidently BS you for your daily total.

Even when the chat/messages are just "Ate ABC with XYZ values, what's my daily total?"

While I guess new chat for each day, or some MCP for storing and retrieval of record/meals would've helped with those daily totals.

The total would still be wrong - unless you explicitly specified each of the values you need to track (e.g. carbs, fat, protein, kcal) to be put into records.

At which point of course - you're not really using AI/LLM but basically an CRUD application.

How/what are you basing your "sweets/desert impacts my sleep more than a drink or two" on?

I was waking up an hour or more before the alarm (so waking up <=6AM with 7AM alarm). And I thought my sleep was good - after all Fitbit sleep score was 80-85.

Then after stopping alcohol I started sleeping longer. Specifically waking up later, at least for the first few weeks.

Seemingly alcohol was causing earlier waking due to spiking cortisol too early.

While waking-up time took a few weeks to recalibrate. Already 2-3 days after stopping alcohol - Fitbit was showing clear improvements in actual sleep quality metrics - HRV was increasing and RHR/BR were decreasing.

And now my "bad nights" have Fitbit sleep score of ~85, and it's regularly 90+.

Lab results are night and day - e.g. CRP was 20 (>5 signals inflammation), just few weeks after stopping alcohol it was ~10 while I was having cold/fever, and now it's <1.

The biggest/hardest problem for me was that after stopping alcohol, my sweets intake increased, especially in the evening. I was doing almost no carbs during the day, and then in the evening ... I guess brain was lacking some easy dopamine that it would previously get from alcohol... I would crave sweets, ice cream ...etc.

It took almost 3 months to be able to stick with strict keto diet. I'm finally doing <=20 gram of carbs because from past experience - any higher and I have a hard time limiting carbs to say just 50 (which would still be low carb/keto).

It's based on the sleep tracking data my fitness watch gives me, but also just based on how I rested I feel the next day, since I know there's limitations to the data quality a watch can give you.

I have noticed the early waking thing with alcohol, but for me that's never really an issue after a drink or two - only at higher consumption levels do I really have that problem (along with the bad dreams/nightmares that many have after drinking).

One other thing I've realized while experimenting with this is that I tend to be sleep better when I'm cooler - so anything that would raise my body temperature has a huge negative impact on sleep quality. This could be large amounts of alcohol, or it could be lots of sugar in the evenings, or a salty/starchy meal late in the day, etc. This seems to be a more important variable than anything else I've observed.

Since it was Dutch news and Dutch ship - I'm betting they used ~€5 Bluetooth tracker available in Action stores.

Not sure how different it works from Apple/Samsung trackers. But my Motorola Android phone can set it up, and phones out and about are reporting where it is.

In The Netherlands I wouldn't be able to login to any government or adjecent websites (e.g. portal of my local health center/GP, health insurance, retirement/pension insurance) without a smart phone running DigiD app for 2FA.

The non-EU Serbia has the equivalent app, but also you might be able to get individual/personal e-certificate (for logging into e-government or signing e-documents) added into smart card chip of your ID. But in practice it seems thats only used for business purposes, like CEO/Accountants/etc to sign/submit business records/taxes.

In Belgium the eID software runs on GNU/Linux, so I can log in to government websites using my ID card and a card reader. In my experience it even works better on GNU/Linux than on MS Windows.

It's one of the only things that Belgium does right.

The DigiD app isn't required. You can log-in with DigiD using SMS 2FA. This is what I currently do, because I don't want to install closed-source software on my device.

DigiD used to work fine without an app. I think it still does, because I have to explicitly select using the app to log in.

Dumb phone works as well with sms verification.

I used to work for a GSM messaging gateway/SMSC. And seeing first hand how most of those SMS messages (2FA, password reset, bank transaction/balance ...etc) are usually routed (sure over SSL but stored/forwarded as unencrypted GSM packets) through several different companies around the world - before reaching your mobile operator ...

And on top of that you add stuff like sim cloning, and all the other things that one gets by having a direct SS7 connection (there were blog posts/YouTube videos - IIRC Linus Tech Tips calls/SMS got routed to Australia).

Using SMS for 2FA or anything similar is my last resort.

Granted I stopped working there 15+ years ago - but I imagine that the basic economy reasoning where it's impractical for every mobile operator to have a direct peering contract with every other operator in the world - is still the same.

And messages originating from non mobile users/operators (like DigiD 2FA) always start at one of these messaging gateways/SMSCs (e.g. InfoBip.com), and often go through a few different ones before reaching your mobile operator.

I didn't come across those narrow seats when we looked into solution for fitting two kids and an adult (grandma) in the back row.

So we went for (especially in Europe) rather limited subset of cars where all 3 of the 2nd row seats are proper sized, with Isofix on each of them.

Usually same makes/models that offer the option of additional 2 seats in the 3rd row.

1990-01: "At last, an assistant that follows your directions"

Setup AdGuard-Home for both blocking ads and internal/split DNS, plus Caddy or another reverse proxy and buy (or recycle/reuse) a domain name so you can get SSL certificates through LetsEncrypt.

You don't need to have any real/public DNS records on that domain, just own the domain so LetsEncrypt can verify and give you SSL certificate(s).

You setup local DNS rewrites in AdGuard - and point all the services/subdomains to your home servers IP, Caddy (or similar) on that server points it to the correct port/container.

With TailScale or similar - you can also configure that all TailScale clients use your AdGuard as DNS - so this can work even outside your home.

Thats how I have e.g.: https://portainer.myhome.top https://jellyfin.myhome.top ...etc...

Can't answer if you should add them or not...

But if you do - you would get some notifications from Google about that website/domain.

I've only ever seen emails of the "There's an increase in 4xx/5xx errors on site/page(s)"

I also get “there were crawl errors”, which upon investigation are for pages that never existed (and I’ve owned the domain for 20 years, so its not a previous owner/operator thing)

Countries in Europe realized that if USA sanctions International Criminal Court judge - that judge suddenly loses access to their email/calendar/docs/etc because Microsoft/Google/etc have to comply.

For the rest - yes.

What do you mean by "stack" MRs?

Just like with plain git - in GitLab you can merge a branch that has multiple separate commits in it. And you can also merge (e.g. topical/feature) branches into one branch - and then merge that "combined" branch into main/master.

Though most teams/project prefer you don't stretch that route to the extreme - simply because it's PITA to maintain/sync several branches for a long period of time, resolving merge conflicts between branches that have been separate for a long time isn't fun, and people don't like to review huge diffs.

I guess what I'm saying is: for very large complex features, I don't want one big commit. I want to review a series of commits and then I want to have that series of commits persist in the history.

This is how Gerrit operates "natively" - the commit message and everything is part of the artifact under review exactly like the diff.

If the model is to squash an MR into a single commit before merging it, I'd then want to be able to have MRs that depend on each other.

You can "chain" them and there's some native support for this in Gitlab, but I can't say I've ever tried using it. If I really need a feature branch, I just create a separate branch and target my MR's to that until the whole thing is ready to land in main. Again, it seems less natural to me than how Gerrit does it.