gregw2

Per his GitHub repo ( https://github.com/pfalstad ) last week he ported his electrostatics sim fro. GWT/Java to Typescript+Vite ...

He also wrote zsh ( https://en.wikipedia.org/wiki/Z_shell ) as a sophomore in 1990, moved onto other things a few years later, and a couple decades later in 2019 it became the MacOS's default shell.

It was a joke/tease question.

I have yet to be able to create diagrams or visualizations I am happy with with LLMs. I can get a diagram, but tweaking it via prompts is extremely painful.

I do think it's a interesting moderately straightforward test case for an LLM that is beyond today's frontier.

What a great visualization!

Now can your favorite LLM make me a similar one for the Real #s?

The conclusion that the sun and moon look the same size in the sky is most intuitive when you think about how ecipses appear to us. The relative curve shapes when the moon covers the sun, and the ways total eclipses appear to match sizes do suggest they do "look" the same size to earth observers.

Regarding seeing gigantic moons, I know what you mean. There is a weird psychological effect where objects, particularly the moon, look bigger near the horizon than at the top of the sky that is apparently in our heads and not due to any physics of, say, refraction. I've tested this for myself by pinching a large moon between my fingers and watched it visually then suddenly seem much smaller.

Is there any reason Zuckerberg/Meta should get huge lawsuits and criticism for addictive design and say, Pinchai/Google/Youtube should not?

Not a troll question, just wondering.

As someone with kids, I still don't see a scenario with YouTube fading... What are you talking about? The video slop problem? If so, it's sort of like saying you don't think the Internet can survive if GenAI keeps going... which I've also heard but it didn't seem like that was where you were going...

If certain political problems to be solved are zero-sum, and others are win-win, I'm not sure how adding more intelligence will solve the problem of the zero-sum ones that are the source of conflict and resource constraints.

The premise (hypothesis?) that all political problems are resource problems is also a curious one.

I'm not an academic but I've worked through practical problems in this space for longer than I'd like to admit, often in ignorance of the literature. The most CS-y fundamental paper would probably be Lamport(1978) below, but the database papers are pretty fundamental on this topic in their own right.

Here are some pointers:

* Astrahan et al., "System R: Relational Approach to Database Management" (1976), oldest paper talking about logs in a database which is kinda what at its root this is recapitulating

* Jim Gray, "Notes on Data Base Operating Systems" (1978), covers logs with the goal of transaction management, undo, redo, the famous "two phase commmit" process for aligning state across a network boundary in two systems

* Leslie Lamport, "Time, Clocks, and the Ordering of Events in a Distributed System" (1978), not a database paper, more about state synchronization in general

* Bruce Lindsey, "Notes on Distributed Databases" (1979), standards for replicating data across multiple identical database nodes

* Jim Grey et al, "The Recovery Manager of the System R Data Manager" (1981), describes "write to the log first, then the database (WAL)" pattern

* Silberschatz, Stonebraker, Ullman: "Database Systems: Achievements and Opportunities"/"A Architecture for Heterogeneous Database Replication" (1980), discusses tricky bits about replaying logs in client database systems that work differently from the producer

* Leslie Lamport, "The Byzantine Generals Problem" (1982), on the math and needed consensus error handling when distributing state

For other later topics on state synchronization, read up on Paxos (Lamport, 1989/98), RAFT (2013), CRDTs (2011), etc.

The above approaches of "write-to-the-log-then-update-state" were applied to UNIX filesystem first in AIX 3.1 in 1990 and then adopted by other UNIX vendors and then in Linux ReiserFS/ext3/XFS(SGI) in 2001ish.

Lotus Notes took a different path in the 1990s to synchronizing state within documents which was not ACID-oriented like the above, but was more like the modern append-only-log with optimistic eventual consistency. Not sure about the best paper on this.

Then Martin Fowler popularized event sourcing in 2005 with Enterprise Application Architecture, and later described CQRS in 2011 and microservices. So "we" decoupled everything with Webhooks (2007) and Kafka (2011) and reinvented this problem. Oh, and did I mention Blockchain (2008), etc. Oh, did I mention Bitkeeper dvcs (1998) and git (2005) handling of distributed state?

Along the way databases (Snowflake, BigQuery, Iceberg, etc) started exploiting logs to show "older" state via features like "Time Travel" queries. Which is actually what Stonebreaker tried to do in the earliest versions of PostgreSQL but the first implementation in the 80s didn't perform well without good compaction support and on much more constrained resources.

Ask your local LLM for a good survey paper and it might be a bit easier to digest, but the above gives you some color and keywords.

I was a little surprised to find someone experienced in time series modeling not mentioning bayesian modeling nor the use of ensemble models... but I do think its more fundamental points were worth making.

Between the difficulty and value propositions for forecasting timeseries "disk usage" versus "financial markets" there are some rather relevant time series such as "company sales" or "demand for company product" that come up again, and again, and again but are neither "easy" nor "predicting-the-stock-market-hard".

What seems intuitively wrong as a layperson new to this about Coase's theory, is that the "surprising" collapse in prices to marginal cost "in period 1" assumes that consumers have no marginal utility, and thus no price sensitivity, of consuming the good sooner rather than later.

If that fails. Coase's argument fails. No?

There are a limited number of bitcoins and Satoshi started out as a monopolist of them... so Coase expects them to get sold at the marginal cost?

There are a limited number of iPhones....?

Do either of those examples shed light on where Coase went wrong that agree or disagree with the authors?

The earlier Spanish measurement system that changed to meters in Gaudi's lifetime was, like the English system, more 12-centric than 10-centric.

In a way, base-10-centric measurements are the victors of a three thousand year battle between group-by-10s beancounter arithmatical people in Egypt/Mesopotamia and the also ancient 12-centric make-things-easily-divisible architect/construction geometers.

True, 12 doesn't match the number of fingers or toes... but have you ever considered it does match the number of spaces around our fingers and toes?

When I saw the emphasis was on 12, I wondered why it was combined with 7.5 so much rather than say, a more 12-centric nearby number like 7.2.

7.2 = 12 * 12 / 10 / 2

But having integers for ones' for lengths (e.g 90m rather than 84.4) does seem a bit nicer in a way.

I think you make a good point about 15 vs 7.5 and symmetries around 1/2/3/4/5.

It undercuts the 12-centricity though; 15-centricity is just so much more man-centered, not fitting for someone about to be beatified as a saint!

It's an interesting time window you chose. Why would there be an anomaly during that window (if there is one)?

Perhaps it is due to the outward-facing, civic-oriented values coming out of WW2?

There was a lot of reflection in America on what went wrong in German pre-war thinking and culture coming out of that period.

The WW2 men in their 20s in 1940 were in their 40s in 1960s and their political power would have kept growing through peer older politicians into the 90s.

I don't see the underlying economic dynamics of the relevance of substitution costs driving the way we do things going away. And substitution costs are all about limitations.

Abundance and limitations are a bit of ying/yang phenomena in terms of driving things, you don't have one without the other.

Igor Stravinsky: "Constraint drives creativity"

(I also don't see Amdahl's Law --which is fundamentally about limitations -- going away any time soon.)

I do agree that there are compounding abundancies present.

Legalization reduces the risk of violence to sellers and buyers and external parties. This in turn reduces the risk premium and quick score potential of the "business".

The problem with the legalization strategy to reduce violence is that it has its limits.

It turns a high-risk game into a high-volume high-scale game.

It validates the creation of a corporate ecosystem who then are incented to create demand for the "vice" while simultaneously concealing what they know (or come to learn) about the vice's side effects.

I can confirm. Redshift support is mediocre even for a F100 firm with TAM support if the workload is large and complex and you have some needle in the haystack causing problems like you allude to.

Practically speaking keeping an eye on locks and transactions is a good idea, as is watching out for your statistics on key core columns going bad when they shouldn't. (analyze and vacuum sometimes don't actually do anything when you need them to...)

You raise an excellent point. I'd only counter that a number of the factors I raised are manipulated to retain mass surveillance even in the presence of mass encryption. Let's start with manipulating random number generators or controlling elliptic curve constants...

It would also be interesting to understand how much "imposter syndrome" seen at Ivy institutions stems from just the inevitable "little fish big pond" dynamic and how much stems from cheating.

(And how much that "imposter syndrome", from either source, then drives later hard-work/success and how much doesn't.)

That's a very interesting call out, the connection between gaming academics and (gaming) finance.

They both do have very concrete point systems with a parallel set of less-measured but very real externalities, don't they?

This brings me to a bit of a related story.

A family member of mine who attended Princeton and was an undergraduate Residential Advisor (RA) in the dorms responsible for care of freshmen recalled hearing a presentation in the early 2000s to parents of students from an academic dean or faculty member. The dean boasted to the parents how great their kids were, describing how each year in the last decade they kept adding more work to the students and the students kept rising to the challenge. My family member RA, very aware of the resulting stress the students were under was horrified. This family member thrived at Princeton and loved it, but is quite wary of trying to put their own children on a track to get there or go there.

This event correlates with the increasing fraction of students at Princeton going into finance which began in the early 1990s and which peaked in 2006 with 46% of students at Princeton going into Finance. I had not considered a correlation between student psychological stress and psychology of "gaming"/cheating and the psychological going into finance until your comment.

At that time, there was some sense that perhaps many Princetonians went into Finance because they had to pay of the huge loans from the price tag. After a couple decades on working on financial aid improvements, now that Princeton (tuition) is free for people with family incomes under $250k/year and has been for a while, and still large numbers (admittedly not quite as large) are still going into finance, I'm not sure some of the psychological factors around taboo topics like gaming/cheating and/or more prosaic related factors like reducing cheating while properly sizing the expected workload for the non-cheating population have been explored.

The light source tech was pivotal, but the supply chain mastery of 100,000 parts and patience to invest "200 billion dollars" in development over decades is deserves massive respect for the Europeans, no? This effort did not start in 2013.

This does not work if your communication enpoint is the same as your encryption endpoint.

Or you don't control your key material.

Or your tech supply chain.

Or leave your device unattended.

Or aren't susceptible to the same "five dollar wrench" attacks used by certain in-person Bitcoin wallet thievestgat are also available to state actors.

I could go on...

You raise a good point.

Skimming the original article, I didn't really understand why the author didn't discuss "WITH" CTEs (for SQL newbies, common table expressions, see https://modern-sql.com/feature/with ) as alternative composition mechanisms.

Or even SQL views. But your ergonomics comment makes sense to me.

The hidden problem many password managers gloss over is how unbelievably insecure the Windows copy-n-paste is in Windows for decades.

IIRC (corrections welcome) Windows's window manager broadcasts the contents of the "copy" operation to any application that requests to receive the ? WM_CLIPBOARDUPDATE event... so any windows malware or even legit application with legal fineprint basically gets a plaintext message with the contents even before the "paste" occurs. All running apps are trusted.

Here's an example blog entry from "grumpy-sec"(?) laying this out (2018): https://share.google/0til1YzbF4xFRY7ls

Not to mention newer Microsoft conveniences like logging your clipboard history to your disk so it doesn't go away when your computer reboots, and/or syncing it to the cloud.

I have a related question, is anyone developing standards on how agents can proxy the requestor identity to backend database or application layers? (short lived oauth tokens perhaps, not long lived credentials like the ShowHN seems to focus on?)

I heard about this feature first from Snowflake but there are similar options around in other ecosystems which may be of interest to someone here and one thing to keep in mind with even Snowflake's implementation...

Snowflake's implementation only works within a single Snowflake account, not cross-account, which implies if you want to clone across dev/qa/prod you must manage those environments within a single Snowflake account.

BigQuery has a very similar "table clone" feature. It works across GCP projects (accounts) but not across organizations.

Redshift and Azure Synapse do not really have this feature at all.

Databricks, Microsoft Fabric and the Iceberg Nessie-only catalog do support something similar, often called shallow cloning.

(Nobody really supports cross-region cloning... which makes sense if you think about it.)

Who says it started in 2011? I find that hard to believe.

Gates was having interviews with John Brockman's edge.org in 1996: https://www.edge.org/conversation/bill_gates-digerati-chapte...

Gates was at The Edge dinners in 2010: https://michaelshermer.com/articles/my-dinner-with-bill-gate...

Jeffrey Epstein funded The Edge in the late 90s early 2000s, then not around 2008/9 during his first trial and prison, then again after he got out:

https://www.buzzfeednews.com/article/peteraldhous/jeffrey-ep...

https://epsteinweb.org/john-brockman/

I throw out this observation more to be provactive than persuasive, but I haven't seen it elsewhere..

People before me have observed how Trump's moves all are ego driven, or self serving or serve Putin or Israel or gas companies, and I'm here to add to the mix a different conjecture.

Trump's moves all tend to increase inflation in a plausibly deniable way. Tarrifs, fed-fighting, wars, etc.

And that is a deeply unpopular but elite-viewed necessity for handling America's national debt.

Inflation allows the wealthy class to get away with extending government spending without admitting/pursuing austerity which was political suicide under Carter.

The wealthy shelter in their land and stock portfolios which keep growing unlike cash and also benefit from said spending, while ordinary people pay the extra regressive tax that is inflation. The elite can then turn around and blame the little guy for supporting Trump and their hands are clean.