current journal
FAQ
search
contact
rss/xml
atom/xml
spam notice
terms of use
warrant canary




archive


A character in a dream last night said that his desktop publishing software automatically replaced ss with ß as long as the line of (German language) text it appeared in was within 22 degrees of horizontal.

German has rules about this but they're nothing like that. A friend and I used to argue over whether Google was evil. I think we were arguing less about Google than about whether a yes‑or‑no question does justice to the issue. I keep coming back to Solzhenitsyn's take on such matters: "If only there were evil people somewhere insidiously committing evil deeds, and it were necessary only to separate them from the rest of us and destroy them. But the line dividing good and evil cuts through the heart of every human being. And who is willing to destroy a piece of his own heart?" (from The Gulag Archipelago)

That said, Google looks more uncool to me with time. I don't know which is changing more, Google or my point of view.

Google Maps recently started asking me about places I've been to. As fate had it, the first instance of this was creepy: my Android phone spontaneously asked me whether parking at a doctor's office was free. The same parking lot serves a bunch of businesses but Google knew I was at a medical office.

I usually have satellite navigation turned off on my phone but I had it on inadvertantly a couple weeks ago. And there are other ways a phone can sense its location, e.g. from which Wi‑Fi networks it detects.

Chrome was my (desktop) browser of choice for years but I'm trying out others now, partly because I'm more leery of Google and partly because of performance issues. I've been seeing Chrome do inordinate amounts of disk I/O. There are threads on various web fora about this kind of problem, some with suggestions on how to ameliorate it—but there's something to be said for a browser you don't have to figure out how to tame. first ascent
A buddy drilling a hole for a bolt on lead on a first ascent today.
the same garment featured in the 1 November postingOne of the more understated logos on anything I own.
Shorter Sarah Sanders (from today's press briefing, after being asked about allegations of Tr‑‑p's sexual misconduct):
He says it didn't happen. And besides, it happened a long time ago.
Following up on some old postings:

Ivan Chermayeff, designer of (among many other things) the big red 9 I mentioned on 25 Jan 2010, died on Saturday at 85. His obituary in the NY Times includes this paragraph:
Working in three dimensions, Mr. Chermayeff designed the sidewalk sculpture — an immense number 9 in red steel — that marks the entrance to 9 West 57th Street in Manhattan. The building, by Gordon Bunshaft of Skidmore Owings & Merrill, is noted for its convex facade that glides down to street level.
Well. 9 W 57 is concave, not convex. (You could argue that any concave surface is convex when regarded from the other side. But a façade is generally apprehended from without, no?) Obligatory screenshot here in case the Times fixes* their page.


On 19 May 2011, I wrote about metastability, defined by Wikipedia as "the ability of a digital electronics system to persist for an unbounded time in an unstable equilibrium or metastable state." Metastability rears its intractable head when (for example) a system has to decide which of two events happened first. See also Buridan's ass.

Bitcoin's blockchain system is subject to metastability. Each Bitcoin miner that strikes paydirt adds a new block to the blockchain and announces it to the network. Should two miners extend the chain at about the same time, the distributed database may fail to reach consensus as to who was first. As described in the original Bitcoin paper, the can is kicked down the road:
If two nodes broadcast different versions of the next block simultaneously, some nodes may receive one or the other first. In that case, they work on the first one they received, but save the other branch in case it becomes longer. The tie will be broken when the next proof-of-work is found and one branch becomes longer; the nodes that were working on the other branch will then switch to the longer one.
That works unless two miners again succeed at about the same time. In practice these disputes are resolved before too long but there is no guarantee of how quickly resolution will come.

* It's fixed as of 11 Dec 2017 (tacitly, i.e. without a correction notice)