herf

6 ms·

31.
by herf 6mo ago
Doing this without the parents on board does not work. Kids can lie about their birthdate by a few years. Facial age estimation has error bars of like 5 years and many teens don't have any ID. Younger kids use a parent's phone. Ma
32.
by herf 6mo ago
for the youngest ones, a lot of these are "mom's phone" or something like that, it's not even accurate to say you are identifying the user
33.
by herf 6mo ago
I think HiDPI is another reason - these were from the "96 DPI" era where pixels looked the same on every screen. You can draw all your pixel art at 2x (or 3x) and scale it down at load time, but it's not super easy. Also, som
34.
by herf 6mo ago
Both Dropbox and OneDrive default to "online first" for most users (including Dropbox on macOS which has moved itself into File Provider). It is a technically sound and sane default for Backblaze to ignore these mounts, especially
35.
by herf 6mo ago
This is an ideal place for LLMs to run (is this changelist a security change or otherwise suspicious?) but I don't think the tokens will be so expensive. For big platforms, transit costs more money - the top packages are something like
36.
by herf 6mo ago
There are a lot of details in the original article, in most cases comparing with Opus, which required "human guidance" to exploit the FreeBSD vulnerability: https://red.anthropic.com/2026/mythos-preview/
37.
by herf 6mo ago
We need a high quality compression function for human readers... because AIs can make code and text faster than we can read.
38.
by herf 6mo ago
Apple makes a nice distinction between their "app layer" (iCloud drive and Messages, etc.) and the OS login. This would work fine for Windows power users, and for the most part Windows has already had this (your "store"
39.
by herf 7mo ago
This "early velocity only" approach seems like a problem - how do you know with 5-minute training runs that you aren't affecting the overall asymptote? e.g., what if the AI picks a quantizer that happens to be faster in the
40.
by herf 7mo ago
It's easy to lie to an OS about your age because it's a single-user experience, and if your parents allow you to lie (or don't know), that's all it takes. Social networks are so much better equipped to estimate age becau
41.
by herf 7mo ago
They teach a lot of Taylor/Maclaurin series in Math classes (and trig functions are sometimes called "CORDIC" which is an old method too) but these are not used much in actual FPUs and libraries. Maybe we should update the cu
42.
by herf 7mo ago
https://arxiv.org/pdf/2310.11453 The original paper [fig 1, bottom-right] seems to say it needs about 4-5x the parameters of a fp16 model. You can build it and run some models, but the selection is limited because it h
43.
by herf 7mo ago
I wish there were more ways to specify whether the Windows filesystem /mnt/c should be mounted in a WSL2 instance - it is kind of generally on or off. In cases where I want WSL2 to function as a "container" isolated from
44.
by herf 7mo ago
It's sad to see an LLM take over a blog, because you can see the line: before 2026 it's an interesting person you would like to talk to. After 2026, it's like generic LLM marketing-voice copy.
45.
by herf 7mo ago
Sideloading without automatic updates is not very useful
46.
by herf 8mo ago
The days you move between categories can establish your birthdate, which is a lot of bits if you are doing this on an individual level (basically it's a great start at a supercookie).
47.
by herf 8mo ago
You know, LLMs could do automated code reviews for each update to avoid things like this. It would be much better than unexamined updates.
48.
by herf 8mo ago
"Noftsker also shared the hacker aversion to cigarette smoke, and would sometimes express his displeasure by shooting a jet of pure oxygen from a canister he kept for that purpose; the astonished smoker would find his or her cigarette
49.
by herf 8mo ago
There is a 15-30% difference between the groups at baseline (fig 8c-9c, 8d-9d), about the same magnitude as the claimed effect of the experimental condition. I think the result would be much stronger if these baselines were comparable, so t
50.
by herf 9mo ago
What if your query is answered by an insane AI that is obsessed with getting you to buy things? Like this demo that could only think about the Golden Gate? https://www.anthropic.com/news/golden-gate-claude This would m
51.
by herf 9mo ago
Most systems now "fail closed" because they are based on a code signing chain of trust that has no exceptions. It would be better if some portion of these systems were made to "fail open" - you don't want a botnet t
52.
by herf 9mo ago
It's especially important for people trying to use lower-resolution OLED (e.g., for gaming) - the text fringing can look quite bad. At higher DPI it can be less noticeable, though horizontal lines still have noticeable fringes. Here is
53.
by herf 10mo ago
It would be useful to separate "AI that stays on your device" from "AI that makes you send more personal data to the cloud" - maybe two scripts?
54.
by herf 10mo ago
Training data is quite literally weighted this way - long responses on Reddit have lots of tokens, and brief responses don't get counted nearly as much. The same goes for "rules" - you train an LLM with trillions of tokens an
55.
by herf 1y ago
Here's my situation: 1. My Microsoft account is still @msn.com, which I don't trust in any way to be secure, since it's not an email account I ever use 2. I have lots of Samba and other shares that know my local login 3. If m
56.
by herf 1y ago
Vertically integrated apps are much cheaper to run - Instagram stores only a small fraction of your photos and makes a lot of money from them. It is somewhat harder to explain why we pay for things like iCloud, which mostly has no web API,
57.
by herf 1y ago
"Try it for yourself by opting into the experiment in Labs." This is a really confusing call to action.
58.
by herf 1y ago
Assuming mobile platforms weigh in with an API sometime, it's notable that the only people allowed online by default would be minors who are using parental controls, because they would be able to prove (a) age and (b) parental consent
59.
by herf 1y ago
We deployed hashcash for a while back in 2004 to implement Picasa's email relay - at the time it was a pretty good solution because all our clients were kind of similar in capability. Now I think the fastest/slowest device is a br
60.
by herf 1y ago
Thank you for doing this - I think there are two lines in the sand that a true "user agent" would allow us to set: 1. Don't show recommended content at all--I want to view only the content I've specifically asked for (or

More ›