gorgonical

What do you mean? You use the 5lb tank via the Aarke instead of the CO2 cylinders? Or you refill somehow the CO2 cylinders? If it's the first one this sounds like the best of both worlds with the convenience of the countertop device and the cost-efficiency of the bulk CO2. It would be great if you elaborated some more!

Works like a charm for me, just like for you. Getting ~8 tok/s on my middle-of-the-road workstation laptop. Thanks for this!

Musician-turning-tech anarchist (?) Benn Jordan is making a very interesting series of videos about Flock cameras, their poor safety, and their gray-area interfacing with local governments:

https://www.youtube.com/watch?v=UMIwNiwQewQ

https://www.youtube.com/watch?v=uB0gr7Fh6lY

https://www.youtube.com/watch?v=vU1-uiUlHTo

https://www.youtube.com/watch?v=Pp9MwZkHiMQ

I recommend them.

Mind sharing the fix as a patch? I would like to run it this way, too.

Regarding the teaching workload: This is not generalizable; during my undergraduate studies a significant fraction (maybe the majority? too long ago to be sure) of my classes were taught by graduate students, especially the math and computer science classes. At the graduate level, your statement was true for me at my second university. In fact, I'm not sure if a graduate student would be allowed even to teach a graduate-level class, considering their credentials.

My experience around universities (as an academic) is that, generally, the number of adjuncts scales linearly with overall funding/skill at grantsmanship in the department. That is, the smaller universities I know saddled professors and their graduate students with substantially more non-research work, including teaching and administration.

Re: the point about Zig: Especially considering I used and played a lot with D's BetterC model when I was a student, I wonder as a language designer what Walter thinks about the development and rise in popularity of Zig. Of course, thinking "strategically" about a language's adoption comes off as Machiavellian in a crowd of tinkers/engineers, but I can't help but wonder.

At the risk of sounding pedantic, you're (I think, implicitly) claiming that reducing Russian oil revenues doesn't impact their war capabilities. Why would that be the case?

Concerning how widespread DBT is: In the last five years in grad school, I knew many people doing a psychology PhD and DBT was one of the major training focuses for their clinical psychology curriculum. I would say the characterization given by this article matches very closely with what they told me over the last ~5 years.

My impression of DBT compared to CBT, based on what my friends told me, is that DBT is much more confrontational. I remember one friend even specifically said that it took her a long time to "unlearn" the therapist's natural response to affirm and validate, but then redirect negative feelings with skills.

I like it, but the fact that you don't have to input the whole word correctly and only each individual letter really conflicts with my muscle-memory to correct mistakes. I also think the animations and effects are a little over the top and they distract me from typing.

Objective statements require some evidence. Are you referring to evidence elsewhere in the conversation or have you forgotten to include it?

Probably an obvious addendum here, but the classes I remember having the most engaging lectures were flipped style where you didn't need to take notes necessarily, because the class was about discussing and deepening the understanding of material you saw already. That was true for my physics classes as well as philosophy. I think it was doubled up in usefulness when we were assigned material that asked us to act on our deepened understanding soon, e.g. before the next class period, such as one of the many "opinion pieces" we wrote for things like dualism/monism, etc.

Technical subjects achieve this with labs, too. It doesn't scale but we see clearly that scaling isn't always very desirable, especially if it leads to this regression.

I don't know but if you want to find out a good place to look is Luke Ranieri's Polymathy/ScorpioMartianus channels. He engages very actively with the Ancient Greek and Latin languages and his content is also useful for discovering other links.

The tl;dr with Ancient Greek as I understand it (warning) is that dramatically less content was written and over a much larger time period. Homer's works are even described as their own (Homeric) and as such it makes "Ancient Greek" a more nebulous term than "Latin," even when you account for Old/Classical/Late branches. In turn, making it in my estimation harder to have a Fr. Foster equivalent.

Look into it and let me know.

It's funny to hear someone describe my daily driver personal laptop in the same way as a 380Z. I use Slackware with Xfce4 and except for the lower screen resolution (1360x768) I have never noticed that it's not "modern." I even have a new battery so it gets 6-9 hours of use again.

Like you said, the giveaway is the poor performance, but if you're a systems developer that usually isn't a problem anyway. Emacs, C, assembly, some Chisel and Forth are all that I write on it.

Unless I misread they don't state exactly how the attack escalates privileges to install the driver. Could there be two versions of the attack with varying levels of severity?

This is very cool and very useful as a commuting cyclist. The routes that OSM and Google give me don't always have a high "reasonability" score in that they may be the most direct routes, but the road conditions are very uncomfortable for cycling for any one of many reasons.

One thing that would be very useful is to color the segments on the map based on the waytype. The proportions are given in the summary but unless I already know the route I can't tell where exactly those segments of difficult cobblestone are.

The use of LD_PRELOAD as part of the attack surface makes me think that a statically-linked binary has some value. Not a maximalist approach like some experimental distros, but I think there's clearly some value in your standard userland utilities always performing "as you expect," which LD_PRELOAD subverts. Plenty of Linux installs around the world get on fine using BusyBox as the main (only?) userland utility package.

This is presently how my website works. I'm no webdev so I'm probably committing all sorts of heresies, though. I use m4 macros to build the header/footer structure with links and coloring and whatnot. Then the content pages are emacs Org files I publish to HTML and include with the m4 macros. You can do a surprisingly large amount with Org Export.

Location: Pittsburgh, PA, US

Remote: Willing/preferred currently

Willing to relocate: Yes

Resume/CV: Send me an email

Email: gorgonical at fastmail.com

Techologies: C/C++, ARM assembly, RISC-V assembly, D

I'm currently a PhD student working on my dissertation in trusted computing. I have published papers in high-performance computing, virtualization, and trusted computing, with a focus on operating systems and system software. I have written hypervisors for Intel x86, developed OS kernels for ARM and RISC-V at national labs, and written hardware drivers for our research kernels as well as Linux. Our research focuses on multi-kernel contexts and the advantages that pairing an alternative kernel with Linux can get us. Right now I'm working on my dissertation but am eager to pursue opportunities that would occur in the spring.

I think this is a key point to catch in this discussion. Like a sibling comment said, this is the way things are done in his EU country, but the big difference is that EU countries are higher-trust with way more financial regulation and accountability. China's economy is fundamentally not predicated on the rule of law and that's an important point here.

The description of that video credits one of the creators as Reini Urban, which I'm gonna infer means yes in this case

As a Westerner with a lot of cultural knowledge of Christianity but very little of the structure of Buddhism, yes please give the long answer.

For the record it is not correct that NixOS's choice to store language packages in its own repos is unique: Slackware does this and IIRC Ubuntu and Debian both do this. In fact, this is so common that I occasionally see warnings on package websites telling you not to install it from your distro's repos specifically because it will be out of date.

There are a few ways of comparing these numbers. I think you're suggesting that something like deaths/kWh is a better metric, which I agree with; if we just go with total deaths than something like pedal-driven generators are the safest way to generate power, which is obviously an unhelpful statement.

However, the greater point is that although nuclear power is dangerous by default because of the waste and risks of meltdowns it can be made very safe with engineering and still be a cheap generation method. By all accounts I'm familiar with fossil fuels cannot be made safe for either the environment or people while still being cost-effective.

A major issue in the nuclear vs fossil fuels argument is perceived vs actual risk. I don't have the numbers, but even though Fukushima was a huge disaster, the death toll is officially 1. But the cleanup has been very expensive and very visible. Meanwhile, coal/gas/oil plants deflect the equivalent costs of their cleanup onto workers and people in the communities in increased mortality and healthcare costs.

More succinctly, nuclear can be safe with effort, but fossil fuels seemingly can't be safe, no matter how much effort.

Sure. gorgonical@fastmail.com.

I'd be interested in a copy. How can I reach you?

I did an internship at Sandia this past summer, specifically on the HPC programs. Being a PhD student, the pay for that internship was excellent, and the cost of living in ABQ is low. I lived somewhere that I could bike to my office. My manager was excellent and a technical staff member and really understood how to manage other technical staff; no pointless meetings, hands-off, and tried at every turn to shield us from boring administrative non-sense.

The work was interesting, though my experience was very skewed as I was working on a solo project. I met weekly with my "mentor" to discuss where I was and if I needed support, but I was working nearly entirely solo. Like the parent comment here says, from what I learned the norm is that you work on many projects -- my mentor certainly was. Even in the department meetings it was clear that while we were unified under a general theme, each person in the department was working on their own, many projects.

My work was entirely unclassified, and my understanding is that most of the people in my department worked on projects with similar levels of classification. My office building looked like every other building and infosec and opsec requirements were pretty mild; wear your badge, don't photograph things, don't tell people any specifics about what you do.

I was offered to stay on as a full-time intern as part of the hiring pipeline and if it weren't for that it's in ABQ I would have strongly considered it -- the work was very interesting and also like the parent comment says, you are largely in control of what you do there. It's a lab first, not a defense weapons company, so research is the name of the game for the department I worked in.

Echoing a sibling comment, it seems strange it's so common that people look at probabilities like this and say that because the thing that happened was less likely that the model was wrong.

They say explicitly how this works: they run simulations and count winners. That means among their simulations is very likely to be an instance that is similar to the real outcome.

I have read the paper. They showed photosensitivity by both autoclaving the sample and adding DMCU to prevent photosynthesis. Both reduced power production. They do suggest a component of power production is related to electrolytic activity because they observe a dark setting power production; under a light setting the power production nearly doubles.

They also discuss the possibility that the algae are just promoting the oxidation of the aluminum, but do a series of assays I don't fully understand and conclude that some of the current produced is most likely from the algae themselves.

Link to paper: https://pubs.rsc.org/en/content/articlehtml/2022/ee/d2ee0023...

This seems to be a PDF of the paper:

https://pubs.rsc.org/en/content/articlehtml/2022/ee/d2ee0023...

It sounds like they are making the claim that they can revoke access to data immediately, but they say you don't need to store the content on-device. I looked up "key displacement" and can't find any immediate references, though I'm not an expert.

My understanding is that if I acquire a ciphertext encrypted with key v1, I will always be able to decrypt it. Any "asynchronous" protocol will have this problem, right? I can be revoked access if I need a piece of the key from a trusted party or if that v1 of the ciphertext can be guaranteed deleted, but caching old versions of keys and files means I will always retain access to the versions of files I ever had access to.

Could someone with a bit more understanding shed some light on this?