HN user

FabHK

15,613 karma

Unrepentant ultracrepidarian

http://www.fabian-lischka.de

hackernews@fabian-lischka.de

[ my public key: https://keybase.io/fab; my proof: https://keybase.io/fab/sigs/XVCHiJ8si5GOpHUKjHh8QVkG9Czgvft0ZTZGHmRLnkI ]

Ceterum censeo Elsevier(um) esse delendum.

Posts134
Comments5,471
View on HN
blog.dshr.org 5d ago

Portents of Doom

FabHK
2pts0
danluu.com 12d ago

Agentic test processes, LLM benchmarks, notes on agentic coding from Galapagos

FabHK
1pts0
annas-archive.gl 4mo ago

If you're an LLM, please read this

FabHK
3pts1
blog.dshr.org 4mo ago

Tesla's Not-a-Robotaxi Service

FabHK
5pts2
blog.dshr.org 5mo ago

Data Centers in Spaaaace

FabHK
3pts0
electrek.co 6mo ago

The White House just said your life is worth nothing. Yes, literally.

FabHK
16pts2
electrek.co 6mo ago

Tesla to stop selling FSD package, moves to subscription-only: why a big move

FabHK
3pts1
www.nytimes.com 8mo ago

A cryptography research body held an election and they can't decrypt the results

FabHK
84pts40
www.ft.com 11mo ago

Brace for a crash before the golden age of AI

FabHK
4pts0
www.economist.com 11mo ago

Palantir might be the most over-valued firm of all time

FabHK
26pts15
libertystreeteconomics.newyorkfed.org 1y ago

Why Are Credit Card Rates So High?

FabHK
4pts0
www.ft.com 1y ago

The FT View: Encryption 'back doors' are a bad idea

FabHK
7pts1
www.afr.com 2y ago

Physicist can prove that economics has it all wrong

FabHK
2pts3
annas-archive.org 2y ago

You can help Anna's Archive by seeding torrents

FabHK
193pts114
www.wsj.com 2y ago

Flood of Fake Science Forces Multiple Journal Closures

FabHK
2pts0
blog.dshr.org 2y ago

Elon Musk: Threat Or Menace? Part 5

FabHK
20pts6
papers.ssrn.com 2y ago

Murky Consent: An Approach to the Fictions of Consent in Privacy Law

FabHK
3pts0
www.schneier.com 2y ago

Dan Solove on Privacy Regulation

FabHK
2pts0
jacobin.com 2y ago

The Philosophical Case for a Four-Day Workweek

FabHK
2pts0
www.experimental-history.com 2y ago

Startling differences between humans and jukeboxes

FabHK
1pts0
www.schneier.com 2y ago

Hardware Vulnerability in Apple's M-Series Chips

FabHK
2pts0
www.technologyreview.com 2y ago

Let's not make the same mistakes with AI that we made with social media

FabHK
3pts0
blog.dshr.org 2y ago

Elon Musk: Threat or Menace Part 3

FabHK
13pts0
daringfireball.net 2y ago

Apple’s Two-Pronged Annual iPhone Strategy

FabHK
2pts0
m.slashdot.org 2y ago

'Data Have Spoken LK-99 Is Not a Superconductor,' Says US Research Center

FabHK
13pts4
www.thenation.com 3y ago

Lessons from the Catastrophic Failure of the Metaverse

FabHK
4pts1
www.bloomberg.com 3y ago

Germany Sets the New Standard for Cheap, National Mass Transit

FabHK
2pts0
www.bloomberg.com 3y ago

Netflix loses 1M users in Spain over password policing

FabHK
745pts924
www.ft.com 3y ago

A beginner’s guide to accounting fraud (and how to get away with it)

FabHK
26pts1
futurism.com 3y ago

Leaked Messages Show How CNET's Parent Company Sees AI-Generated Content

FabHK
2pts0

BTW, there was a variant of this discussion with the iPod and mp3 players back then. Some people wanted to manage music "by hand" in the file system, and had various schemes for doing so (folders for genre, then artists, then album, etc.), and wanted that reflected 1-to-1 in the mp3 player.

iTunes (and the iPod) had a different idea: The app takes care of file management, and you have a frontend supported by a database containing all sorts of metadata to browse your music (both static metadata (artist, genre) and dynamic (play count, skip count, last heard)). iTunes has smart playlists which I love (all songs with 3 stars or more that I haven't heard in 6 weeks).

I'm firmly in the latter camp, and similarly prefer Jellyfin over a pure SMB share for movies.

My intuitive take:

Eigenvectors answer the question, if you have a linear mapping from a space to itself: which lines through (through the origin) remain unchanged after the mapping? And the associated eigenvalues tell you how much a specific point on that line moves along the line. For example: If you rotate things in 3D, the rotation axis remains unchanged. That's the eigenvector. If you reflect things through a mirror plane, any line in that mirror plane will remain unchanged (with eigenvalue 1). Other lines will be, well, mirrored, but there is one line that is reflected onto itself: the one perpendicular to the mirror plane. However, points on it will go from one side to the other, so have eigenvalue -1.

The SVD says something else: Any linear mapping from a space (to itself or a different one - higher or lower dimensional) can be expressed as 1) first a rotation in the old space, then 2) some coordinate scaling or dilation (expansion/contraction) along the coordinate axes, described by the diagonal matrix with singular values, 3) then another rotation in the new space.

Just to add: Credit is also due to Prof Christian Reinsch from Technical University of Munich (in addition to Golub and Kahan) for the invention of the practical algorithm for computing the SVD still in use today.

https://blogs.mathworks.com/cleve/2022/10/23/christian-reins...

https://people.inf.ethz.ch/gander/talks/Vortrag2022.pdf

https://www.mathworks.com/company/technical-articles/profess...

https://en.wikipedia.org/wiki/Christian_Reinsch

BTW, if you wonder about the dedication ("For Gene Golub on his 15th birthday"):

Gene Golub was a numerical analyst, and father of the practical singular value decomposition (his license plate read "Prof SVD"), together with William Kahan (the father of IEEE 754 floating point numbers).

And his birthday was February 29. (In other words, the article was on the occasion of Gene's 60th birthday).

https://en.wikipedia.org/wiki/Gene_H._Golub

RIP.

Indeed. I like to say, we would like three things to be true:

1) Every day has 86400 seconds.

2) Every day is from noon (sun exactly above) to noon (sun exactly above).

3) We use SI seconds.

You can't have all three. Pick any two:

1, 2, not 3: What you describe. Day has 86400 seconds, we keep in sync with the sun, but we tweak the seconds a bit. There are different versions, like epoch time, or UT1.

1, not 2, 3: Every day has 86400 SI seconds, and we slowly go out of sync with the sun. That's TAI.

not 1, 2, 3: We use SI seconds and keep in sync with the sun +/- 1 second, but need to add/drop seconds occasionally. That's UTC.

https://en.wikipedia.org/wiki/Universal_Time

https://en.wikipedia.org/wiki/International_Atomic_Time

https://en.wikipedia.org/wiki/Coordinated_Universal_Time

Not really. It's still the standard deviation, and it still gives you bounds on probability, for example the Chebyshev inequality:

P(|X-\mu| > k \sigma) < 1/k^2.

So, while for a normal RV, 5% of observations lie outside +/- 1.96 std.devs, for arbitrary RV (with finite variance) at most 25% of observations lie outside +/- 2 std.devs.

Even with equities, things are not entirely trivial.

A few examples: Some stocks trade in one currency, but pay dividends in a different currency. Some stocks go ex-div before the dividend amount has been determined (e.g. in Japan). Stocks trade on certain days, and trades settle on certain days, and they might be subject to different holiday calendars. Corporate actions are not entirely trivial, either (with rights issues, you need to issue two temporary securities for accounting purposes, etc.).

Well, that's not money created out of nowhere. The balance sheet extension creates money and an equivalent liability. That's the point of the statement, not money supply and its regulation.

If someone sells you 12345.55 EUR vs USD at a rate of 1.12345, how many EUR do you think you end up with?

1. If someone sells me 12345.55 EUR, I hope to end up with 12345.55 EUR.

2. That's the point though. They will sell you a certain amount of EUR at a certain dollar price. This, in turn, implies a rate (which might well have more than 5 digits behind the decimal). This is ideally close to the quoted rate, sure. But what counts is the actual EUR amount and the actual USD amount, not what rate was quoted or with how many digits.

you're interested in the risk metrics, like durations, convexity, vega, and so on, no one cares what your rounding convention is. doubles are just fine, thank you.

Doubles are not just fine, but required there. If you compute risk, as you allude to, by finite differences, then you need the precision of floats rather than rounding to integer cents prematurely.

it borders on bad advice (e.g. "Balance is never stored. It’s derived from the movements of money.")

Don't think that's bad advice, in particular as the article clarifies:

Derived state can be cached. Balances and projections can be cached or snapshotted for performance.

Of course you can cache or snapshot. But don't mistake that with storing the golden source.

There's a fairly obscure T-shirt designed by FT Alphaville that just cryptically says "Allen, Ben, Janet, Jay" ;-)

Sorry, that strikes me as the norm. Graduation is predicated on passing certain courses, and obviously you can't take it an arbitrary number of times. So fail an exam twice, and you're out. (Some universities offer a final oral exam after failing a written exam twice.)

But I can also teach a game designer or artist who has never heard of source control how to correctly and safely use P4 in about 15 minutes. Meanwhile you can tell Git is badly designed because there are 10,000+ tutorials explaining how easy it is! Spoiler: things that are actually easy do not need 10,000 tutorials telling you that it is easy.

100% and LOL. Sometimes I wonder where we'd be if Mercurial (hg) had won that contest.