Am I reading the correct article? The one I just read does not include the work "immigrant" nor the word "asylum" at all?

At the risk of being a moron, why does an app that seems to carroonify a photo actually need to be able to delete/manage photos or files?

Good question, that's because it's a gallery app at it's core which lets you do all CRUD stuff on images/folders.

Ebay? MySpace, Usenet? Though the distinction was about features and adoption not about ethics. Back then it was assumed that tech owners (to the extent that something like Usenet had owners) acted ethically.

Google's motto of "dont be evil" was mostly tongue in cheek until of course it wasn't and then they stopped using it.

This is tech though and the idea that one company might have an ethical marketing advantage is much older. For example when I worked for GM we often talked about being like GE only with a heart.

> This is tech though and the idea that one company might have an ethical marketing advantage is much older. For example when I worked for GM we often talked about being like GE only with a heart.

That’s an interesting point.

Great visualizations but you can't buy a shoe without knowing that a 10 in one brand is not a 10 in a second brand or that for example you need to size down when ordering Dr. Martens then there is no way to expect clothing to be more accurate than a shoe.

Sure, not every shoe brand is equal, but if I know I'm a 9, I can generally start there and find a shoe plus/minus a half size. I have yet to go into a store and wind up in a shoe that is 3 sizes larger than what I thought my size was. Or 3 sizes smaller. Or a size 8 in one shoe from a brand and a 10 different shoe. I can order Nike/Jordan brand shoes without trying them on and they fit. Have done it for years.

I went to re-buy the "same" jeans ~8 months after my initial purchase and the size I was wearing didn't fit in the new jeans. Tried another pair with a different wash and was back to the original size. I have tried on jeans from the same brand with similar cuts and came away two sizes apart. I can swing several sizes as a starting point between some stores. I get it, not every jean is going to be identical, but it isn't a ridiculous ask to be able to have a size I can start at and be within a size of what I need.

Anecdotally, I discovered recently that I’m a full three sizes down in Vivobarefoot shoes versus normal shoes — but for a really interesting reason. It turns out that modern runner’s shoes actually are often shaped like a foot, rather than like a spatula, and so now that I don’t have to size up for my toebox width thanks to them creating shoes that are foot shaped in disregard to fashionable propriety, I now fit much better into a 3-sizes smaller shoe than I did into their older shoes at my prior size.

Part of why retailers are afraid to change sizing is that lots of women install their clothing into their ego and brag about it socially. I don’t approve, but I recognize the extrinsic cultural circumstances* that pressure them to do so. It’s a a lot harder to brag about being a size 49-42-48–8-30 than it is to brag about being a 20UK. (The /22 in 20/22 UK, the common size these days, is silent, because size lying is normalized.)

It would make more sense instead for them to choose an anchor measurement and a body type modifier; but that gets into the problems of having to annotate nine different body type letters onto a numeric size, not to mention having to design clothing that looks good on nine different body types, and having to hire models of nine different body types. The modeling industry is unprepared to staff that need, too!

* The phrase “pinup-hourglass male-gaze body-shape imposed-ideal” serves as an excellent starting point for research on that nightmare. For those unfamiliar, ask your friends who are women about that exact phrase, and remember to listen rather than critique their potentially-lengthy reply. I’m focusing on the sizing discussion and leave that topic as an exercise for the reader.

I've collected size charts for every shoe model across 200 brands and found that it's not uncommon for any given size to vary +-1.5 sizes (a range of 3 sizes total). And that's size charts from manufacturers.

Many online stores use outdated or outright wrong sizes charts. One of Europe's largest online retailers don't even bother providing official manufacturer size charts at all. They have one generic size chart for almost all of the shoes, so it's pretty much down to the consumer to return and retry. Kind of crazy state of things.

Here are my rules:

1 it is the responsibility of the employer to may an appropriate living wage.

2 there are very few cases where a tip more that 15% is warranted

3 never tip in advance before you have received service

4 never tip when it is not clear what you are being appreciative of

You might try recovery using all the prior email accounts you remember. Personally though as someone who hires programmers (rather than say ops) I put 0 stock in this kind of credential and I have never checked that a claim to be certified was valid let alone current.

Why do you categorize this as an "operating system"?

Fair point. When I said “operating system,” I meant that car buying today is super fragmented. Most people end up juggling a bunch of tabs and tools at once, plus a bunch of calls.

What I’m trying to build with Vehique is to put all that in one place. Users just describe what they want, it searches live inventory, finds matches, tracks new listings daily, and lets you compare cars side by side. When they’re ready to buy, we help with dealer communication via AI-assisted calls.

I’m trying to make it so you don’t need to visit other tabs or make calls anymore, by bringing the whole journey into one interface.

But yeah, “operating system” might be too much. Maybe “the only tab you need” is more honest. What would you call it?

Why do you feel an older cut off degrades the model? Perhaps it improves it as the model has limited training on AI garbage.

That's a possible hypothesis, but qualitatively the model seems to be so much less capable than it was even 4 hours ago.

And I'm talking about GPT 5.2 High (non-codex), which is the best trade-off model for capability vs. speed.

It's like talking to a senior programmer who all of a sudden experienced a traumatic injury. Noticeable.

Interesting. For what it is worth, I use Claude Opus at the moment and I am generally impressed.

I assume you mean outside the boundaries of context window size or token thresholds right?

Both in a way. I'd say it gets worse outside the context window it seems to get worse due to compression and loss of detail. Even in a single context window it's hard to keep ideas and branches clean and remember important insights.

Azure's services offer multiple levels of input sanitization/filtering and filtering input is one of the current cornerstones of mitigating injection attacks. If Azure dows it, I imagine all the main players do.