from Hacker News

Hacking physics from the back of a napkin (2020)

by g0xA52A2A on 12/14/24, 5:31 PM with 14 comments

• by terminalbraid on 12/14/24, 7:48 PM

  MIT OCW has a free textbook "The Art of Insight in Science and Engineering: Mastering Complexity", which I usually point folks to on the subject of estimating physical systems. I think this is a real gem. Another work of the author is referenced at the bottom of the article. I'll also vouch for the referenced Guesstimation (though I hate the title).

  https://ocw.mit.edu/courses/res-6-011-the-art-of-insight-in-...

• by getnormality on 12/14/24, 6:23 PM

  I saw all of these hacks in my undergrad physics education and yes, these were the highlights of the whole experience. They should be spread as far and wide as possible.

• by atoav on 12/15/24, 2:16 PM

  This back of the napkin math is one of those things I whish more people just did.

  E.g. when people talk about harvesting vibrational energy on the streets or in wearable electronics you should be able to do a ballpark judgment of how much energy such a technology could roughly output in an ideal world.

  It always amazes me how few people in finance apparently have no clue about even the most basic laws of physics when they repeatedly fund things like these.

• by mncharity on 12/14/24, 9:37 PM

  For biology, bionumbers[1] is a fun resource.

  [1] https://bionumbers.hms.harvard.edu/Search.aspx?task=searchby...

• by sourcepluck on 12/15/24, 2:43 PM

  > Hacker spirit. Nowadays, the word “hacker” conjures up visions of Russian trolls, Julian Assange, and Angelina Jolie’s 90s pixie cut. But a nobler usage predates this. Hacker culture, in the original sense, grew out of places like MIT in the 60s, with its tradition of highbrow silliness and elaborate technical pranks.

  Implying that Julian Assange's endeavours were not noble, or in some sense comparable to "Russian trolls", or in some sense antithetical to the "hacker culture" of yore, is heinous and ahistorical. I mean that specifically in the sense that I think history will show it to be false.

  Not only that, it's also a gross misunderstanding of what Wikileaks was - a huge "hack", in the early MIT sense. A problem was observed, and a solution with genuine "hack value" was applied to it.

  The facts illustrating this are already in the public domain, although I am of course aware that they're floating around in a sea of insinuations and fake scandals and half-truths and propaganda and bare lies.

  Source: I followed Wikileaks and Julian Assange since around 2009. It naturally goes unmentioned in the majority of newspaper and documentary treatments of the subject, but Wikileaks is deeply rooted in the cypherpunk ethos of the early 90s, which itself is historically tied to the earlier MIT hacker culture.

  Stallman, who this introduction goes on to cite as a bona fide representative of the early culture and an "open source gnuru" (shudder), is a vocal supporter of Assange, and has stated clearly his belief that Wikileaks has that hack value I mentioned above.

  I recommend the book Cypherpunks to get a feel for the actual cultural and technical ethos surrounding Wikileaks, for anyone interested. For a rebuttal to the incredible amount of crappy reporting on the legal side, Nils Melzer's book on the Swedish case is good. There are plenty of good articles too (in the middle of a couple of orders of magnitude as many bad ones), but here's one that might be a good start for someone interested:

  https://theindicter.com/the-significance-of-wikileaks-as-inv...

• by zburatorul on 12/15/24, 7:00 AM

  Boltzmann constant is not fundamental. There's no need for it at all if we switch from measuring temperature in Kelvin to Joules.

• by mahdihabibi on 12/14/24, 6:24 PM

  Is this good book for beginners?

• by mncharity on 12/14/24, 9:05 PM

  A feel for reasonable values, and rough-quantitative understanding, is widely thought important for expertise. Despite the limited attention it usually gets in education. But even when instruction is relatively well resourced, as with college seminars devoted entirely to estimation, the content available seems less than inspiring. "Estimate turkey cooking time from first principles, without recourse to wrapper or google"... um, yay? Either we're profoundly confused about utility, or there seems an under-addressed opportunity to gather content which revels in that utility.

  If beyond call-and-response, memorize-and-regurgitate, plug-and-chug, problem-based-reasoning, and the-equation-is-the-phenomena, there is something more to aspire to, where quantitative understanding in content is pervasive and foundational, something which illuminates, simplifies, scaffolds, integrates and exercises understanding... then maybe we've a whole lot of content-creation work for the collective todo list?