vladde

> drive. you'll need the car at the car wash.

using opus 4.6, with extended thinking

my personal page :)

> Although our investigation is still ongoing, our early review indicates that your personal data has been accessed, which may include: Identity information: > - first name, last name, date of birth, gender; > - Contact information: email address, home address, telephone number, if provided; > - Passport information: passport number, country of issue and expiration date (not including copies of the documents).

haha this made me chuckle

looking at you USB 3.0 (or USB 3.1 Gen 1 (or USB 3.2 Gen 1))

i'm placing my bets that in a few thousand years we'll have changed calendar system entirely haha

but, really interesting to see the insane methods used to achieve this

my bank includes transactions that are yet not final (preliminary transaction where currency conversion still isn't final). these rows change both in both the title and the amount, but i can do some guessing on how a row updates.

however: when i import my transaction a couple days later again, i need to manually keep track and remove preliminary transactions (now removed from my bank's export).

related: when i imported a CSV into YNAB, i would have to manually keep track of updated entries and remove those. with some code and state handling, i could figure out which rows no longer existed – but i couldn't remove them with the import function.

i ended up abandoning YNAB's CSV import and use their API to remove transactions... but it would have been much simpler if the CSV import could just have removed certain rows from the get-go.

(while i don't think this acts as a budget, it think others will run into similar issues as i have when it comes to importing CSV files)

> cluster the inconsistent labels by embedding them in a vector space

I often found myself adding "use built-in features if they exists", just because because of this type of scenario.

It unsettles me that some people feel okay always accepting AI code, even when it "works".

(the coolest examples come at about half a minute into the video)

However, the model "A high-backed chair" has some floating pieces in the middle of the seat, that are fastened from above. Can these robots handle building these?

Bug report has been open for years: https://github.com/microsoft/PowerToys/issues/5962

When it turns out they were using a different version and didn't have that option, we still had to figure out what translates to what – if the option even is there at all.

this is just `display: grid` though, and not for non-fixed width elements – the hacker news nav bar is an example where the | between links—on small screens—will either appear at the end of the first line, or at the start of the second line

Also, I absolutely love that it's so easy to start playing (no signup, no popups, etc)

Question: For the countries I know and select first time, I get them multiple times between countries. It's a bit odd to have to select England a couple of times, but the countries I don't know, I seem to get less often, or no repeats at all. Is this intended?

(edit) Question 2: Is there any scenario where you'd actually guess wrong more than once? After the first miss click, the country is highlighted. I don't see I'd ever get "You found country in 3 clicks".

There is no way for me to even play with some players. I wish I could "block" them, because the fun is gone when every tile is gets completed before I can even react.

My best game was my first, where I played against people my skill. Whenever the user "rapidash" is in the lobby, I might as well just quit, since I won't be able to get in more than one or two correct placements.

When I'm driving and want to switch ANC mode, it's convenient as well.

(It loaded about 0.3MB/s for me)

Maybe it's my brain that's weird, but when I have a train of thought, and then read the auto-completion of my sentence, I forget what I was thinking and suddenly I can't think of anything other than the (usually incorrect) suggestion.

󠅤󠅕󠅣󠅤 edit: apparently not edit 2: oh wait, the bytes are still there! copy-paste this entire message and it decodes to "test"

Personally I think this is the root cause of most sloppy AI code. If you just look at the code that was generated, and you don't think "I would've come up with that", then probably the code is wrong.

    1. limit the text's container to `80ch`
    2. set line height to `1.5`
    3. (optional) change the document's root text size to `112%`

(i've never liked my browser's "reading mode", so i don't use it)

I don't think I've ever seen good management. Anyone care to explain what that would look like?