Nick

Theoretical physicist by training (quantum open systems/quantum information), Currently paying the bills as an engineer working in free-space optical communication. I also enjoy the outdoors and learning about any area of science, philosophy, and many other academic areas as well.

Public Key

npub1254wcjqdese7zgtsnxcf3j2g2a2erlk5rm6l5rtl00spfhj9qcmsqceqaf

NIP-05 Address

internic@mathstodon-xyz.mostr.pub

Profile Code

nprofile1qqs922hvfqxucvlpy9cfnvyce9y9w4v3lm2paa06p4lhhcq5mezsvdcpz4mhxue69uhhyetvv9ujuerfw36x7tnsw43que2l2g

Publishing to

relay.ditto.pub

Author Public Key

npub1254wcjqdese7zgtsnxcf3j2g2a2erlk5rm6l5rtl00spfhj9qcmsqceqaf

Show more details

Last Notes

2026-08-30 00:14:28 UTC

I just installed #GrapheneOS on a new (to me) phone, and man was that shockingly easy. I remembered it being easier than i was expecting the first time I did it, but either they made it even better or I forgot how quick and simple it was.
#android #privacy

2026-07-19 23:31:09 UTC

- reply

@nprofile…0xjm I would welcome correction by those more knowledgeable, but for the benefit of people like me who were wondering, it looks like this just covers MPEG-4 part 2, i.e. the video codec that is roughly equivalent to H.263 and was used by things like DivX, not the more prevalent H.264 (which I believe is MPEG-4 part 10).

2026-06-11 18:48:52 UTC

- reply

@nprofile…yjg9 They're also easy to talk to, because they find everything scintillating. 😉

2026-05-31 04:10:23 UTC

- reply

@nprofile…0hde I'm not at all sure it suits your needs, but I wonder if https://magicgrants.org/ might be an option. It's used by PrivacyGuides for their subscriptions.

2026-02-04 23:12:52 UTC

- reply

@nprofile…t6k2 @nprofile…4hp5 It looks like you can also choose to make posts (by default) quotable by followers only, which might be a happy medium.

2025-12-28 05:53:25 UTC

- reply

@nprofile…atf6 @nprofile…00p2 One of the experiences this made me think of was learning Lebesgue integration. After I had finished the class, the topic came up somehow with someone else who basically expressed it as "integrals where you sum the areas with horizontal rectangles rather than vertical," and I thought "why didn't anyone express it in that simple way before this?" While you can perhaps quibble with the correctness of the description, as a "motivation" it works pretty well.
In that case this wasn't a big problem for me, because the idea of "generalizing length/volume" worked well enough as a motivation for measure theory, and from there integration just looked like an application. But it illustrates the general issue with omitting obvious motivating ideas.
@nprofile…ag4u

2025-12-27 22:06:33 UTC

- reply

@nprofile…00p2 A thing I've found challenging is a significant subset of math books that dive right into a bunch of definitions and theorems without giving the "motivation" often seen in physics books, describing what sorts of questions they're trying to answer and how we might intuitively guess that a certain approach might be productive. Ultimately I'm happy to have the rigorous exposition, but it's still useful to have a sort of overall plan for what you're building so that you know how to arrange the various building blocks in your mind.
@nprofile…ag4u

2025-12-26 18:24:16 UTC

- reply

I think I have similar frustrations to yours @nprofile…5gem. I remember many years ago when I heard some traditional Chinese music for the first time, and hearing that in comparison to the European-derived music I was used to made plain to me how much of what we think of as "the way music works" is historically or culturally contingent. It seems to me that music theory arises from three things, math/physics, intrinsic human psychoacoustic perceptual characteristics, and cultural conventions. But most of the discussion of music theory doesn't seem to distinguish which is which, and I find that rather confusing.
I suspect that if I spent more time reading music threads by @nprofile…8eua I might be able to resolve some of that confusion, though.

2025-12-26 16:19:29 UTC

An acquaintance is thinking about jailbreaking their iPhone. Not being an Apple user, I haven't followed the topic closely, but I thought this was generally believed to be a bad idea from a security perspective (at least for a primary device).
Unfortunately what I can easily find online are people with no obvious credentials saying it's fine and security companies (who often overblow risks, like the infamous "juicejacking") saying it's bad. I'm hoping the Fediverse might be able to give me a more nuanced or fact-based perspective.
#iphone #security #mobileSecurity #mobileSec

2025-12-10 00:03:12 UTC

- reply

@nprofile…00p2 I'd definitely be curious to hear which uses cases you find it particularly useful for. My general sense is that it should be good for tasks that are vary roughly analogous to an NP problem, i.e. not necessarily easy to produce the output but relatively easy to check if the output is good. So, as an example, it makes sense to me how it could be useful for coding, where you have a compiler or interpreter and, hopefully, a test suite to tell give you some indication of whether the output is good (in addition to looking over the output)..
The idea of semantic search is interesting, and I've heard that some models, like Perplexity, are particularly good about providing (real) references. I'd be more wary of using many other text generators in lieu of a search engine, unless I was trying to just remember details about something I already knew (for fear of getting incorrect information). I have heard some stories of people asking it academic questions where the response wasn't fully correct and the references weren't real, but it gave them good leads on terms and authors to look into..

2025-12-08 05:19:53 UTC

- reply

@nprofile…00p2 I'm curious to hear your thoughts, if you continue your "fiddling". Most discourse on LLMs that I see seems to slant so far to one side or the other (either LLMs have nearly human capabilities or they're completely useless) that I don't find it very credible. And I haven't personally had much occasions to use them (I haven't been that motivated to use them on a personal basis, and they're difficult to make use of on my professional work due to data restrictions). I probably fall more on the skeptical end of the spectrum currently, but I can easily imagine that LLMs are significantly useful for some purposes (e.g. as a coding aid).

2025-11-30 23:26:16 UTC

- reply

@nprofile…9s72 This just makes me think about that Sagan quotation, "If you wish to make an apple pie from scratch, you must first invent the universe."

2025-11-20 22:43:26 UTC

- reply

@nprofile…stct @nprofile…w8t5 In fairness, the Brits sent us Andrew Wakefield first. 😜

2025-11-09 23:49:23 UTC

- reply

@nprofile…stct The other option that the article I posted doesn't address (much if at all) is going with an appliance that just keeps water continuously warm (often sold by Asian companies). Since my kettle is just for home, where we rarely make more than 6 cups of tea in a day, that didn't seem like a good solution for my use case.

2025-11-09 23:17:30 UTC

- reply

@nprofile…stct But generally kettles seem to be another example of an infuriating trend in appliances, where you can't actually buy quality at any price: paying more gets you approximately the same quality of workmanship with a more attractive exterior and a specific brand name on it.

2025-11-09 23:15:00 UTC

- reply

@nprofile…stct It's a reasonable guess. I think there were several that looked similar. I specifically tried to identify one where the opening mechanism was metal (which it is on the Cuisinart) rather than plastic, since we'd had that fail on an even earlier kettle.

2025-11-09 22:58:49 UTC

- reply

@nprofile…stct I have the Cuisineart model from that review. I don't notice off flavors, but perhaps I'm not as sensitive. It does had a transparent window, so there is some plastic and sealant involved.
I used to have an electric kettle that I loved, which was reasonably cheap and had multiple temperature settings and an all-stainless-steal interior, but it broke after ~10 years of heavy use, and then I found out they don't make that model anymore.

2025-11-09 22:20:27 UTC

- reply

@nprofile…stct You may be interested in the Wirecutter review. Specifically, I found this tidbit enlightening:
"Unfortunately, the longevity of electric kettles can be unreliable across the board. We sent kettles to the Ohio State University metallurgy department, and researchers took them apart to study their components.
Based on their initial observations, as well as information from an OXO representative, we suspect that nearly all of the kettles we’ve tested use the same circuitry and temperature controls — most likely produced by a company called Strix. Pricier kettles don’t necessarily have more-durable electronics — you’re paying for looks, construction, and fancier features."
https://www.nytimes.com/wirecutter/reviews/the-best-home-kettle/

2025-11-04 06:01:39 UTC

- reply

@nprofile…pklh This just makes me think of when I was teaching physics, and I had pre-med students taking the course who were trying to prepare flashcards as a study aid. But, yes, prizing memorization seems even more absurd in that context.

2025-11-03 05:52:05 UTC

- reply

@nprofile…00p2 If you're vague about white holes, today is your lucky day, because I ran across this shortly after reading your post 😁
https://mathstodon.xyz/@johncarlosbaez/115445374242954199

2025-11-03 00:53:12 UTC

- reply

@nprofile…9s72 While I am a sometimes listener to Mindscape, I didn't listen to that specific AMA. I guess I would think that the early phase of the big bang is not all that analogous to a white hole in terms of its specific properties (for reasons @nprofile…00p2 mentioned, and questions like "in that case where is the event horizon?"). I would think that the white hole might instead be an instructive example in the following sense: The time reversibility of GR together with the solution for a black hole tells us that a white hole must also be a solution; likewise, if we grant that a Universe uniformly filled with matter (that is initially contracting) can collapse to a singularity under gravity, then time reversal symmetry says that the initial expansion of the big bang must be possible.

2025-10-31 01:39:37 UTC

- reply

@nprofile…00p2 I believe that archive.ph is one of a set of sites that are designed to mirror web content, sort of like the Internet Archive but with fewer rules, I think. Others include archive.is and archive.today. It's not clear to me if they're all the same site or maybe just run the same code. People often use them to circumvent paywalls.

2025-10-31 01:27:39 UTC

Someone I know said that while attempting to use archive.ph they were presented with what appeared to be a #Cloudflare captcha that asked them to copy text from one box on the webpage and paste it into another. They felt this seemed too much like some sort of phishing and did not do it, but now the question is this: is this actually some sort of captcha that is used by Cloudflare or a similar service?
In searching form information on this, all I keep coming upon are articles about the ClickFix attack, but as far as I can tell that's usually about tricking the user into pasting some text into a system shell, whereas this just seemed to want the text pasted into another window in the page. So I'm trying to figure out if this was some clever, more innocuous-looking variant of the attack or a legitimate (but weird) captcha mechanism of some sort.
#infosec #security #phishing

2025-10-04 11:57:57 UTC

- reply

@nprofile…5k3r Have you encountered Frankel's "The Geometry of Physics"? It's purpose is develop those sorts of geometrical ideas and show their applicability to physics.
https://www.cambridge.org/core/books/geometry-of-physics/94894F70DB22055BD7BC2B84C135ABAF

2025-08-27 01:24:07 UTC

A friend found a tab open in Safari on their iPhone that they had not (knowingly) opened and whose history indicated it had visited a succession of locations, including multiple web searches and several different websites (some related to the web searches, some not). What are the likely ways that this could have occurred? Could this indicate a larger problem?
My best guess is that a site they opened had ads from some ad network with malvertizing on it that opened the extra tab and caused it to visit the various locations, but I'm really not sure.
#infosec #security #malware #iOS #iPhone

2025-06-19 14:14:52 UTC

For those who haven't heard, astrodon.social is shutting down.
I don't blame anyone running a server for needing to stop, and it's good that they've given users a fair amount of notice. But I do think that closures like this highlight the need to find sustainable models to sustain and administer servers in the Fediverse that don't rely on one or two people holding it all together and supporting it out of pocket. They also highlight the need for a better "credible exit" mechanism that makes sure people can retain followers and posts, even if an instance shuts down without warning (a problem that, for example, Bluesky has solved).
#mastodon #fediverse
https://astrodon.social/@sebinthestars/114595973387772639

2025-05-26 17:54:00 UTC

I often see people using "AI" (often non-specific as to what tool exactly) to summarize a longer text source (or spoken sources with speech to text). My immediate reaction was that you could give a generative model (with sufficient memory) a text and a prompt and ask it to produce a summary, and it would likely produce something that looks like a summary but fundamentally isn't; rather it would be some text it made up that looks statistically similar to a summary and contains text statistically correlated to the input text. That would often look similar to an actual summary, but would differ from what we'd normally think of as a summary because it could contain things that were nowhere in the input text (just correlated to language in the original text) and might actually contradict the input text if it happened to be unusual compared to what was in the training set.
On the other hand, it did seem to me that it would be possible to train an LLM model specifically around the task of summarizing text. It would have the usual constraint that it would only ever be probably approximately correct (PAC), but it could be less or highly unlikely to produce fabricated output or output contradictory to the input. Based on a little reading, it looks like such specialized models exist. My question is this: When someone uses one of the major commercial chatbots (ChatGPT, Claude, etc.), and they ask it to summarize some text, is it employing such a more specialized model or is it just employing a more general text generation process as I described in the first paragraph?
#ai #llm

2025-05-20 18:00:15 UTC

- reply

@nprofile…fa4k Maybe @nprofile…uyg4 will want to get in on this discussion.

2025-05-18 12:03:21 UTC

- reply

@nprofile…fa4k Yes, that's true. I once had someone ask me to help refute Electric Universe/Plasma Cosmology nonsense (because they knew someone who believed it), but when I tried to look into the claims they were just incoherent. You could point to various statements that were baseless or implausible, but it wasn't really even clear enough to say what precisely was the claim to be refuted.

2025-05-18 11:44:18 UTC

- reply

@nprofile…fa4k I know that they're inherently bad at math (because they're just spitting out sequences of tokens that are probable based on their training set), but I thought some had implemented a kludge where when the LLM recognizes a math problem it writes code to solve it (something it *can* do, to a degree) and runs the code to get the answer to the problem.

2025-05-18 11:31:39 UTC

- reply

@nprofile…fa4k I know a common crackpot trope used to be "I have this revolutionary theory. I just need someone to do the math for me." I wonder if the LLMs have now gotten good enough at math to solve that problem for them.

2025-05-16 20:42:14 UTC

- reply

@nprofile…gt6n @nprofile…qz4x Look on the bright side, I guess, you can just attach the remaining edge together and you'll have a cool Klein bottle hat. 😉
(Disclaimer: May require a Universe with an extra spatial dimension.)

2025-05-04 17:11:37 UTC

- reply

@nprofile…fa4k @nprofile…u4rg Yeah that was the bit that also raised my eyebrows. Using power series and then truncating them is your big insight!?
Also notably lacking in the article were quotes from other researchers in the field not involved in the research.

2025-05-02 21:36:40 UTC

Reviews for many types of appliances have become less and less useful for me, because increasingly they focus on how good the appliance's app is and whether it integrates with the Apple, Google, or Amazon ecosystems. But to me these are generally anti-features that mean that the appliance will surveil me, be a security vulnerability on my home network, and potentially lose many or all of its features when some back-end server in a data center somewhere is eventually shut down or access suddenly becomes contingent on a new subscription service at an unreasonable price..
I really wish more reviewers recognized that having an app is a mixed blessing at best. It can be really useful in the short to medium term (if you're willing to give up some privacy), but it's generally bad news in the long term. In some ways the worst option is having an app backed by a service run by the manufacturer that doesn't (yet) have a subscription requirement, because that's an unsustainable model unless the product has manufactured obsolescence.

2025-04-29 22:51:16 UTC

- reply

@nprofile…m2ru What do you mean? It's trivial to isolate a system. For a system of size R, just add an additional amount of mass M = Rc^2/2G, and it will become quite difficult to exfiltrate information. 😜

2025-04-29 22:10:27 UTC

- reply

@nprofile…m2ru I agree that highly secure systems have been shielded against RF for very long time, I guess TEMPEST security was a concern as far back as WWII. So arguably "air-gapped" was always a poor descriptor, but it seems more absurd to me now that RF is not just a potential side channel that could leak a little information but rather one of the main ways computers interface.

2025-04-29 21:58:38 UTC

It strikes me as kind of funny that people still talk about systems being "air-gapped" to mean that they are isolated from machine to machine interaction, despite the fact that wireless networking has been widely used for decades.

2025-04-25 04:21:42 UTC

- reply

@nprofile…5k3r I haven't read it but now I'm quite curious!

2025-04-22 23:59:00 UTC

- reply

@nprofile…59k6 While I generally agree that social media isn't great for that, I've actually been able to find stuff that I knew I posted on Mastodon using the search function with the 'from:' keyword. I do desperately wish that you could also view all the posts from an account (your own or others) without the boosts in addition to the search option.

2025-04-20 15:37:22 UTC

- reply

@nprofile…fa4k Yes, I agree that if you're teaching a class or having a more in-depth discussion and have equipment at hand, you can go the empirical route (as @nprofile…pd3k also suggested). I guess I was thinking more about the context where you're speaking in brief (without props) and just trying to convey the most basic idea of what polarization is in a few sentences.

2025-04-19 16:48:27 UTC

I'm curious if anyone has found a good, succinct way to explain light polarization to someone who has no formal knowledge about electricity and magnetism.
I find this vexing, because polarization is conceptually pretty simple: it's just about the direction in which a transverse wave does the waving. So it's easy to explain if you're talking about, say, waves on a string. The problem is that what is doing the waving here, the electromagnetic field, isn't really familiar to most people (if they haven't taken some physics classes). I've tried before to give explanations that sort of talk around the "what exactly is doing the waving" angle, but then the explanation just sounds oddly vague or abstract.
I've also talked to other people who have had the same problem, so I suspect there just isn't a very good way, but maybe someone out there has come up with a particularly clever way to approach it.

2025-04-16 13:57:15 UTC

- reply

@nprofile…zwgs At the risk of asking a serious question, what's the deal with all these alternative ATProto apps? Are there any that have significant independent infrastructure?
https://techcrunch.com/2025/04/04/beyond-bluesky-these-are-the-apps-building-social-experiences-on-the-at-protocol/
I'm guessing that the answer is that they mostly share the same infrastructure and just provide a different AppView and possibly a different lexicon (not sure that's the right term) as far as the data they store and can interact with. But given that I have a very limited understanding of how Bluesky works under the hood, I'd be interested in a (much) more knowledgeable perspective.

2025-04-15 22:37:38 UTC

I'm curious, does anyone know of a good way to find archived copies of videos from the video streaming app/service Periscope? I'm trying to find helpful suggestions for an acquaintance.
I took a look at archive.org, and it looks like there are a fair number of videos under the subject "periscope", but not a complete set and not containing what my acquaintance was looking for.
I did find this collection of files from @nprofile…w4d5, but I have no idea what they are or how to make use of them:
https://archive.org/details/archiveteam_periscope
#archive #archiving #socialMedia #internetHistory

2025-03-26 04:39:39 UTC

- reply

@nprofile…5lc3 @nprofile…ksf8 Do these actually work on up-to-date iPhones? I know I've seen a lot of claims about tools to gain access to mobile devices that then turn out to only actually work on older OS versions (presumably old enough to have known exploits).

2025-03-09 22:35:44 UTC

I still use an alarm clock rather than a phone to wake up. The fact that I'm still using it is a testament to its durability and how well it does its job. But the one thing that will always baffle me is that it has a dedicated button for toggling daylight saving time on and off; a button that one would use, at most, twice per year. It does not synchronize it's time over the air or anything like that, so I cannot imagine why someone thought that made sense.