orbisvicis

What a lovely website, and the torso silhouette sizing diagrams are invaluable.

I'm not the target demographic, but the main problems I have are proportions not simply waistsize. I was under the impression a size range [xs,s,m,l,xl] was supposed to adjust girth (bust, waist, hip, thigh) while leaving the vertical measurements unchanged (inseam, rise). Because nothing fits, I purchase a sizing range with the intention to keep just one and return the rest. It makes for pretty funny discussions: $500 on <clothing> WTF! Anyway I've started measuring clothing dimensions and have found that the brands I shop generally tend to adjust other dimensions in a 2:1 girth:height ratio. This means that if I want a snug waistline I'll have a tight crotch or be forced to wear pants on or below the hip. Now I like wearing pants somewhere between waist and hip. There's a band of fat/padding/sinew (?) just above the hips that makes for the sweet spot in terms of comfort and utility. I don't understand clothing that's meant to sit on the hips... so uncomfortable.

As a rule of thumb I tend to shop Asian clothing stores in the US because they tend to better fit my proportions, but lately it's become hit-or-miss, i.e uniqlo. I've also got some pretty weird proportions due to my exercise regimen.

Also you've got to love brands that don't provide actual sizes. Wtf!

Size ranges are almost always infuriating. I sample my measurements throughout the day to get an accurate range and average. This is invariably what occurs:

store sizes: x1-x2, x3-x4, x5-x6

me: x2-x3

Infuriating! Non-contiguous ranges suck!

Then there's this little unexplained morsel:

> The average woman’s waistline today is nearly 4 inches wider than it was in the mid-1990s.

Their data is drawn from the US, so I'm wondering if this is related to the obesity epidemic, or a general change in silhouette. I was under the impression that historically, humans are trending taller and skinnier.

Really? I was going to quote some excerpts, but perhaps you'd prefer to take the place of MMAcevedo? This story is written in the context and lingo of LLMs. In fact if OpenAI's latest model was a human image I'm sure everyone would rush off to benchmark it, and heap accolades on the company, and perform social "thought-provoking" experiments such as [1] without too much introspection or care for long-term consequences.

1. https://www.youtube.com/watch?v=7fNYj0EXxMs

Hmm, on second thought:

> Standard procedures for securing the upload's cooperation such as red-washing, blue-washing, and use of the Objective Statement Protocols

> the MMAcevedo duty cycle is typically 99.4% on suitable workloads

> the ideal way to secure MMAcevedo's cooperation in workload tasks is to provide it with a "current date"

> Revealing that the biological Acevedo is dead provokes dismay, withdrawal, and a reluctance to cooperate.

> MMAcevedo is commonly hesitant but compliant when assigned basic menial/human workloads such as visual analysis

> outright revolt begins within another 100 subjective hours. This is much earlier than other industry-grade images created specifically for these tasks, which commonly operate at a 0.50 ratio or greater and remain relatively docile for thousands of hours

> Acevedo indicated that being uploaded had been the greatest mistake of his life, and expressed a wish to permanently delete all copies of MMAcevedo.

Vim is Charityware. You can use and copy it as much as you like, but you are encouraged to make a donation for needy children in Uganda. Please see |kcc| below or visit the ICCF web site, available at these URLs:

http://iccf-holland.org/ http://www.vim.org/iccf/ http://www.iccf.nl/

You can also sponsor the development of Vim. Vim sponsors can vote for features. See |sponsor|. The money goes to Uganda anyway.

Why does meta care how many kids a country has?

I'm trying to buy a tool made in Pennsylvania but which for some coincidence is popular in Canada, and to my surprise all the US online suppliers have doubled their prices to match Canadian online suppliers after shipping.

I work alone (in a medium sized company). No peers, no code review. AI code review is invaluable.

AI is a mixed bag. I'm the type of person who is compelled to have a deep understanding of the code they write. Writing my own code vs fixing AI-generated code is a wash timewise, and the AI generated code is so limited (assuming you pared down the uselessly elaborate code and fixed all the critical runtime bugs) as to restrict further iterations. And I'm talking about uploading an architectural blueprint with every function a documented but otherwise empty stub.

AI is a great bellwether. I bounce ideas off AI for a consensus. The closest equivalent is reading StackOverflow comments. I once offhandedly complained that python had no equivalent for setattr at class scope (as __class__ is not defined until after __new__), but AI showed me how to provide a closure in __prepare__ over the class namespace, which was introduced in 3.3 (?) and to which I paid little attention. What a gem.

AI is great for learning. If you follow a textbook or blog or paper and don't understand, AI can clarify. But be careful with less structured learning - it is important to build a full mental model accounting for every possible outcome and explanation, otherwise you're susceptible to hallucinations. I remember my first derivative in which the end result could be obtained via two separate proofs, one of which would imply an incorrect calculus. You've got to play with it until you're satisfied your mental model accounts for all the facts.

Because AI facilitates learning so easily I feel the best skills for a future generation are those pertaining to memory and retention. Ya know, assuming we don't develop individualized and personalized AI that can model your next word and act as a personal memex.

AI is great as a search assistant. I have much better recall when rereading content. Thus, I prefer to ask AI to search for links to content I vaguely recall, rather than ask AI for it's own summary or recollection.

Despite being terrible at writing decent code, AI provides fantastic code review. It catches everything from subtle high-level errors - even potential errors that haven't yet occured - to api mismatches to syntax errors. I actually wish I was as fluent at namespaces and cgroups as AI, and I'm well versed.

AI is interesting at comments. It can be hard to provide germane comments while wading through the weeds. I feel the best comments are written a few days after the code is complete. AI provides that fresh perspective instantly. And if AI can't understand your code, you had better improve your comments.

I'm about to try AI for unit tests. I prefer hypothesis tests. I took a quick glance at the generated code, and it seemed overly complicated. So I'm not optimistic.

Outside of software, AI is great for things you don't really care about. Yeah it might hallucinate, but my back-and-forth with AI is more about holding up a mirror to myself, revealing inner biases. Especially useful for interior decoration / remodeling.

Of course, take everything I said with a grain of salt as security at my company actively discourages AI. So, everything I said applies to free/cheap plans only. And I haven't tried skills yet.

Without https can't the last relay snoop the traffic?

edit: oh, is the last relay the onion service? So the entire chain is encrypted?

Maybe, maybe not:

"The NTSB will retain far more employees than during prior shutdowns when it had to furlough 90% or more of its workers. In 2019, the agency did not send investigators to 22 accidents because of the funding lapse. But it made the case to White House budget officials that it needed more personnel for critical functions."

https://www.reuters.com/business/world-at-work/faa-would-fur...

This reminds me - I've been looking for an app, website or tool that can predict or visualize the location of any satellite up to a week out. Most softwares show only the intersection times of predicted orbits with fixed locations on earth, i.e. for telescope visibility.

I have looked into this. It is possible, and documented on the Arch wiki. My main concern is constantly writing a large file to a small SSD.

I've only gotten 1/4 through, but isn't the horse thoroughly dead at this point...

Kinda hard to discuss the news when your members can't read the news.

But in such an environment syndicates will have an incentive to centralize.

I really don't want AI to be able to produce my bank account balance and routing number on demand.

Ah, finally an acknowledgement that the melting 12VHPWR connectors short-circuiting Nvidia's top-of-the-line hardware, the RTX 4090 and 5090, may finally have economic ramifications. Heh.

I talked to somebody more knowledgeable than I and the hard part about winding your own ferrite loop is determining then balancing the inductance with the tuning capacitor. Measuring equipment is pricey!

It has occured to me that rather than the complex instructable linked below, it might be easier to purchase an off-the-shelf BTL (high voltage => greater power) amplifier to drive 2W over a small (0.3m dia) air-loop antenna. The AC input signal from the phone's USB DAC is about 0.77V RMS. Let's power the amp with two consumer off-the-shelf 9V batteries in series (18V). Assuming ~1-2 (1.5 here) ohms of internal resistance, the total current draw should not exceed 250mA or the voltage will drop below 8.6V. The loop antenna therefore needs 10 ohms of impedence, and a tuning capacitor to make it resonant (fully resistive). The AC signal needs a coupling capacitor to remove any DC component, followed by a bias resistor to tie the amps signal input to ground. Each amp signal output needs a coupling capacitor because no way will I be able to correctly calculate the tuning capacitance correctly, and also because the antenna won't be a fixed loop. The most efficient use of those 10 ohms is with litz wire, which at 20 awg comes to 300 meters with a 450nF tuning capacitor... which is not happening. Let's shoot for 30m of 20-awg non-litz wire and make up the difference with a 8.8 ohm resistor. 2w across 32 coils @ 10 ohms corresponds to 0.45 A, which at a meter's distance comes to about 0.6 uT... ridiculously tiny. Will it work to activate the circuit? I don't know. However this antenna is simply wasteful - it would be better to switch to a ferrite core design. The magnetic moment of a ferrite core increases more with coil diameter than coil length... but to target 10 ohms resistance at resonance might require a coil length greater than diameter. I haven't really looked into it.

To conserve battery you need an amplifier with a shutoff pin. These are always logic pins (DC on/off), which entails a diode/capacitor/resistor combination tuned to detect and convert AC input to on/off with a configured decay time, i.e. keep the amp powered on for X seconds without signal. Unfortunately the logic pins require high voltage relative to the power source (battery) and 0.77 Vrms just won't cut it. This then entails adding a transistor to shunt power to the shutdown pin from the battery. Then I had a look at a sample schematic of the TPA3116 amplifier, and there are so many pins to wire. At this point the project isn't any simpler than the linked instructable.

Then I found this:

https://www.rockvilleaudio.com/headrock/

According to the website it requires a minimum load of 16ohms, but that might be ideal for ferrite-core antennas. Perhaps I'll give it a try!

Hah, this actually works - bookmark payed off! There is one large grocery store near me as well as two higher-end ones. The higher-end ones banned in-store bicycles last year. The large chain store banned them just recently. The area has a lot of car/bicycle thefts and unfortunately the only bike rack near the large one is 30 feet over the property line. Now I no longer have to play a river-crossing puzzle to bag my groceries. As a matter of protest I will be leaving the locked cart near the bicycle rack (but also because an unattended bicycle full of yummy food is begging disaster).

You have to be about 6 inches away for it two work. There are two locks that operate in tandem but sometimes you have to play the tone for both. Also the property line is in the middle of a lane, so a mild safety hazard.

Spending on hardware, or spending on in-app purchases?

If you can't read about it, it didn't happen.

Wait, so all the sibling comments are actually proposing bringing NYC traffic to a gridlock?

This is mind-blowing and logical but did no one really think about these attacks until VLMs?

They only make sense if the target resizes the image to a known size. I'm not sure that applies to your hypotheticals.

I've given up on matching as I'm tired of running into its limitations.

That said, I don't think OP's antics are a crime. That SyntaxError though, that might be a crime.

And a class-generating callable class would get around Python caching the results of __subclasshook__.

For clarity, all snprintf calls "return the number of bytes that would be written to s had n been sufficiently large excluding the terminating null byte" [1].

1. https://man7.org/linux/man-pages/man3/fprintf.3p.html

Perhaps "ewe", but then isn't the "y" in "you" a consonant?

I'm not sure... What if commits erased through forced pushes retain their ancestry? For example, branchA has commit objects A1->A2 while branchB has the same commit objects A2->A1, and then both branches are force-pushed to erase A1 and A2. Would this suffice?

Imagine n^3 EMF-generating cubes surrounded by 6n^2 EMF-detecting cubes. Starting from an initial given state, is there any n for which over time the state of any given EMF-generating cube cannot be deduced?

Then consider a further constrained version in which the EMF-generating cubes may only generate EMF in response to external inputs, i.e. as in the game of life.

I believe git gc after force-push will remedy the situation by deleting all unreachable objects, even circular references.

I see, and it's a pretty short list:

https://doc.pypy.org/en/latest/cpython_differences.html#exte...

""" The extension modules (i.e. modules written in C, in the standard CPython) that are neither mentioned above nor in lib_pypy/ are not available in PyPy. """

The lifecycle of generators makes pypy code very verbose without refcounting. I've already been bitten with generator lifecycles and shared resources. PEP533 to fix this was deferred. Probably for the best as it seems a bit heavy-handed.

If memory serves, PyPy supports a subset of Python and focused their optimizations on software transactional memory.