HN user
Is the telescope design available anywhere for hobbyists to build? I can't seem to find anything in the article or in a separate search. I'd be interested in perhaps putting one of these together to do radio astronomy with my kids.
“This isn’t just a new value of the Hubble constant,” the collaboration notes, “it’s a community-built framework that brings decades of independent distance measurements together, transparently and accessibly.”
Don't love that I can't read sentences like this without wondering if an LLM was involved.
Probably the biggest purpose for me is to have something to look back on. The last time I consistently journaled before this was during the time when I dated, got engaged to, and married my wife. Granted, that was more than a line journal, but it has been really nice for us to go back and read it together all this time later. With the line journal it is more of a prompt to help me remember that time period than a detailed description. But also I guess there is an autobiographical aspect to it that appeals to me--maybe someday I'll use it to help write an actual autobiography. And if not I also like to think of it as something to leave for my kids or grandkids to read when I'm gone.
I have been doing a one-line-a-day journal for the past two years. If I ever feel like writing more I have a separate section for that, but I only ever tell myself to write the one line. It's the first time I have been able to consistently write in a journal for a very long time before I started. My entries are not nearly as "composed" as the examples here, though, just tiny highlights of what happened during the day.
Right. All it takes is for one to work out, if you have several suitable options. If some of the options are only vaguely suitable, or it comes to light through the process that some of them are not suitable at all, then it takes more than just one working out. That's what I was thinking while reading this.
If you're referring to Cubit, they license the ACIS kernel under the hood.
I notice that your application asks about willingness to relocate. Are you able to do fully remote for the right candidate?
So much of writing is managing your own emotions. The virtue of “pseudowriting” is that it helps you preserve hope for as long as possible—hope that what you will eventually put in place of those square brackets will be good. Hope keeps you coming back: it is more pleasant and low-stakes to pseudowrite than to fix actual language into the draft; and it is less daunting to return to a document where it feels like all that’s left is for you to fill in some blanks. Do that enough times and you will, in fact, end up with something you can read top to bottom.
This describes how I write a new chunk of research code, often. I'll type along until I get to something like "oh, I'll need to calculate the foo of the widget here," and I'll just put a non-existent function call calculateFoo(widget) there until later, when I'll come back and fill it in. I feel like it keeps it manageable; I'm choosing the level of abstraction that I'm drafting code at, and I come in and fill in the details later. I hadn't connected this idea to the journal articles that I am working on; I typically feel somewhat guilty when I add a FIXME in my LaTex document, but with this framing I see now that that is probably the better way to do it than aiming for a finished paragraph from the get-go. The square brackets and placeholders also seem much nicer that the FIXME I was using. Glad to have seen this at a timely moment for me!
I didn't have time to get through the whole article today, but I did spend some time with my kids playing the Wikipedia first link game, which we enjoyed. We kept trying to find one that didn't end in Philosophy, and my youngest son said we should try Brick. Sure enough, it ended in a loop consisting of Existence and Reality.
My brother just finished building an automatic pipe organ for his Mechanical Engineering senior project. Or maybe you'd call it a street organ since it has no keyboard. It has midi input on some arduino-like device, 3d printed pipes, a hand crank pump to fill the bellows, and a bunch of shift registers, relays, and solenoids (on breadboards at this point) to open the valves for the pipes. It turned out really nice, and was a hit at the project showcase. I was involved in helping him figure out the electronics and the code, though I think he threw out the code I wrote in favor of using ChatGPT. All in all, it's not a technically difficult thing to make, though I have the benefit of his hindsight at this point. Maybe I'll get him to write it up at some point.
I'm the kind of guy who decently likes maps, and I pay attention to where I'm going and also to the map before, during, and after using a GPS (Google maps). I do benefit from Google maps in learning my way around a place. It depends on how you use it. So if people use LLMs to code without trying to learn from it and just copy and paste, then yeah, they're not going to learn the skills themselves. But if they are paying attention to the answers they are getting from the LLMs, adjusting things themselves, etc. then they should be able to learn from that as well as they can from online code snippets, modulus the (however occasional) bad examples from the LLM.
You're right, I was thinking of Combobulate, linked in a sibling comment.
I'm not extremely familiar with the details of incremental parsing, but I have used Cursorless, a VSCode extension based on tree-sitter for voice controlled structured editing, and it is pretty powerful. You can use the structured editing when you want and also normal editing in between. Occasionally the parser will get things wrong and only change/take/select part of a function or what have you, but in general it's very useful, and I tend to miss it now that I am no longer voice coding much. I seem to remember that there was a similar extension for emacs (sans voice control). treemacs, or something? Anyone used that?
Here is a nice article on a study of baseball bats using microphones. https://www.acs.psu.edu/drussell/bats/papers/AcousticsToday_...
All models are wrong, but some are useful.
If you’re discussing large groups of people, you have to somehow compress the data. On the other hand, yeah, you probably shouldn’t prefer things like this to explain your neighbor/friend/in-laws over personal interactions with them.
So these addition formulas that apply to series and parallel resistors seem more generally to be formulas for finding the new ratio when you have several ratios and want to add them together treating the quantity in the numerator or the denominator as equal for all ratios. Series circuits have the same current for each ratio, and differing voltages, while parallel circuits have the same voltage for each ratio and a different current for each. Similarly for these aspect ratios, where you want the new ratio after adding width or height while setting the other (height or width, respectively) equal.
Isn't IGA's shtick just replacing classical shape functions with the splines used to specify the geometry?
That's how it started, yes. The splines used to specify the geometry are trimmed surfaces, and IGA has expanded from there to the use of splines generally as the shape functions, as well as trimming of volumes, etc. This use of smooth splines as shape functions improves the accuracy per degree of freedom.
If I recall correctly convergence rates are exactly the same
Okay, looks like I remembered wrong here. What we do definitely see is that in IGA you get the convergence rates of higher degrees without drastically increasing your degree of freedom, meaning that there is better accuracy per degree of freedom for any degree above 1. See for example Figures 16 and 18 in this paper: https://www.researchgate.net/profile/Laurens-Coox/publicatio...
geometry and the fields of quantities of interest do not have the same spatial distributions.
Using the same shape functions doesn't automatically mean that they will have the same spatial distributions. In fact, with hierarchical refinement in splines you can refine the geometry and any single field of interest separately.
What is left in terms of potential?
The biggest potential other than higher accuracy per degree of freedom is perhaps trimming. In FEM, trimming your shape functions makes the solution unusable. In IGA, you can immerse your model in a "brick" of smooth spline shape functions, trim off the region outside, and run the simulation while still getting optimal convergence properties. This effectively means little to no meshing required. For a company that is readying this for use in industry, take a look at https://coreform.com/ (disclosure, I used to be a software developer there).
Interesting perspective. I just attended an academic conference on isogeometric analysis (IGA), which is briefly mentioned in this article. Tom Hughes, who is mentioned several times, is now the de facto leader of the IGA research community. IGA has a lot of potential to solve many of the pain points of FEM. It has better convergence rates in general, allows for better timesteps in explicit solvers, has better methods to ensure stability in, e.g., incompressible solids, and perhaps most exciting, enables an immersed approach, where the problem of meshing is all but gone as the geometry is just immersed in a background grid that is easy to mesh. There is still a lot to be done to drive adoption in industry, but this is likely the future of FEM.
For some reason the copy my library has doesn't have that page the same as it appears on hathitrust. Other nearby pages match the snippets I see on google books, but that page does not, and none of the nearby pages do, either. So I'm afraid my search here is a dead end.
I just put in a request at my university library for that item. I'll let you know what it turns up.
The biggest thing that keeps me from using Julia rather than Python for math prototypes is that it uses one-based indexing. I go back and forth between these prototypes and my C++ codebase, and the mental gymnastics to switch from 0-based to 1-based makes Julia a non-starter for me. I prefer Julia over Python other than that one issue, and the lower availability of tutorials, etc. for Julia.
I’m working on a PhD adjacent to computational differential geometry, and while I’ve made a lot of progress on the computation, I still don’t have much intuition for k-vectors and k-forms. I love coming across articles like this that help me build intuition. The article for which this is the second part was really helpful, but I’m going to have to come back to this second part a few times to fully grasp it. I also loved this quote from one of the articles listed as a source (but with a broken link, I found it at http://yaroslavvb.com/papers/notes/piponi-on.pdf): “Think of a vector as a pin, and a one-form as an onion. You evaluate a one-form on a vector by counting how many onion layers it goes through.”
Edit: This one also looks good: https://math.uchicago.edu/~may/REU2018/REUPapers/Bixler.pdf
This doesn’t seem to address the original question very well. The question is someone who is unable to find time to fill their social needs between working a remote job and caring for children asking for advice, and the answer is “it was an accident that in person work helped with this at all; you’re just going to have to find time.”
I made a calcium fertilizer for my potted peppers last month, soaking dried eggshells in vinegar. They did a similar “dance,” though the bubbles were formed through the chemical reaction rather than a supersaturated liquid. I spent more time than I would have expected enjoying watching it.
This is really cool. I took an optimization class a few years back, but haven't made the time to do anything fun with it since. This inspires me to do it.
I do kind of wish that the last note corresponded to a game over, though, and I wonder if a smaller screen or faster ball would widen the playing field a little. Maybe I'll fork the code and try some of those out myself.
Often times in areas I'm experienced in I will feel strongly that a certain decision is the way to go based on intuition or gut feel. I do think these intuitions are usually correct, but would be hard pressed to describe why. I even sometimes end up BSing something if I'm asked because I'm convinced there is a good reason, I just don't know exactly how to explain it right in the moment. I think the discussion here about the context does a lot to explain that phenomenon for me. The context is smarter than me, and the context collaborates with me without putting all the information in my head.
Of course, I could still benefit from practicing explaining the intuition...
"...arrested an amateur photographer for selling manually photoshopped pictures..."
Am I the only one who finds the phrase "manually photoshopped" in an article about the late 1800s to early 1900s amusingly anochronistic? How about "manually doctored" or even "altered"?
I'm currently working on a PhD dissertation part time, and this concept has been very helpful since I'm not working full days and therefore have more context switching. I end up drawing a lot of graphs and figures to think things through, so a big notebook of graph paper ends up working the best for me. Every couple of days or so I write out a list of the next few tasks as I currently see them, then think on paper to figure out how to implement the code for each task.
Funny thing is, writing it down helps it stick in my brain, so I need the write up less than I would if I didn't write it. That's got to be some kind of contrapositive of Murphy's law or something.
Sure, and even ideas that have been written about can be more precise and complete, perhaps by writing more about them, for example, so no one has fully formed ideas by this logic.
Okay, I can see that now, thanks.