Friday, May 27, 2016

Computers and productivity - a net zero?

Today, for Emily and I, has been a not atypical mix of frustrating interactions across government, school and the private sector.

There’s a common thread in all of them. IT systems that don’t work — because they are old, or partly replaced, or one of multiple overlapping systems, or underfunded, or flaky, or not well integrated into workflows … or all of the above.

There’s something about “information technology” that we don’t understand. Something that makes it different from cars and planes and indoor heating and clean water. 

Unlike those things there isn’t a general trend to improvement. Systems get old and decrepit, transitions are very difficult or impossible, efficiency leaps up and down. We don’t understand how to manage their health and evolution.

Perhaps the past 30 years of economic behavior would make more sense if we treated computers and IT as quite different from all other technologies. A technology where productivity both rises and falls, even before one considers the impact on complexity and how complexity enables fraud.

Monday, May 23, 2016

Fermi resolved

MINUTES OF THE GREETING COMMITTEE 45255425.34543.5

"… Resolution passed. What’s next?”

“We have a confirmed candidate. Spiral arm, water planet, 27 stage II civilizations within 1000 light cycles. They are 55 standard cycles post phase II physics.”

“Wish they’d had a bit more time - but 55 cycles is pretty average. You’re sure?”

“Yes. It’s been 0.1 cycles since their physics community reconciled the usual physics models. The implications have been widely accepted. There was a lot of resistance of course, but the usual tests were done.”

“So they understand now. The way things are for all of us. How are they doing?”

“It’s a path 17 course. They are social animals, live young, two genders with code merged offspring, short lifespans. Typical path 17 grieving, some social disruption, suicides up 200% but now returning to baseline. They are ready.”

“Well, that will be a relief to the neighbors. Hard to keep quiet with that many nearby. Those spiral arms are crowded. What do they call themselves?”

“Well, it’s a water planet, so they call it ‘Earth’.”

“That was funny once. Ok, send the fleet. It’s time to welcome Earth to the Galactic Federation. Usual path 17 grief counseling protocol. Anything else today? …”

Sunday, May 22, 2016

What killed inline skating? A new suspect.

I like inline skating. I started in the 90s a year or so before it peaked, and I’ve watched it decline since. I still skate, as does my #2 son.

It’s fallen a long way. The last time I skated I noticed young children staring in amazement, some asking their mothers what I was doing. Our lovely night skate expired five years ago.

Sports have fashions of course, and inline skating has issues. It needs dry conditions and smooth surfaces; bicycles are far more versatile. If you don’t wear wrist protection, and you skate often, you have a good chance of breaking your wrist (wrist protection drops the risk, it doesn’t eliminate it). Pavement is not kind to skin. Head injury risk is probably higher than with non-peleton road biking.

So it’s not hard to see why it might fade. Skateboarding has many of the same issues, and it’s not doing well either.

On the other hand, inline skating (“rollerblading”) is fun. To me it’s a cross between downhill and nordic skiing — two great sports. It’s no good in the rain, but it doesn’t need snow. It’s an easy way to get around; when I dropped my car off for tire repairs today I skated home from the garage. It’s painful when you crash every few years, but it’s easier on the knees and hips than running or even mountain biking.

I think there’s another factor.

I blame sealcoating. Sealcoating is the application of a mixture of petroleum products and fine stone to asphalt road surfaces. We didn’t always do sealcoating, it grew rapidly in the 80s and 90s. Now most streets undergo sealcoating every 8 years in Saint Paul MN.

Sealcoating has environmental issues, but it does extend the lifespan of city streets and trails. It’s not a problem for cars or bikes.

It’s a problem for inline skates though. Sealcoated surfaces are miserable to skate on for at  least 1-2 years. They get tolerable after 4-5 years, but then they get coated again.

Once sealcoating became universal it got much harder to find smooth and fun surfaces to skate on.

I think sealcoating is what truly killed inline skating.

See also.

Beyond Simplenote: I still want a graph layer (concept map) overlay for my memory augmentation notes collection

I still want a graph layer atop my notes.

Yes, I want my personal locally stored data unlocked memex. Since the passing fad of the web is now, you know, passing, maybe I’ll get one.

Maybe someone will play with this now that there’s not much point in doing another searchable memory augmentation app. Apple and Google each have their “good enough” solutions. Those solutions have scary data lock issues, but for their vendors that’s a feature, not a defect. (For the record, I’m still on Simplenote/nvAlt, despite the extremely very insanely annoying search bug in the Simplenote Mac client. [1])

The idea is as old as time. Each open data format note has a title, a body, tags, and a unique identifier. The app maintains a separate data store of noteID pairs (relationships, no directionality or additional relationship attributes necessary).  When viewing a note one sees titles of related notes. There’s a UI for viewing the graph that also treats tags as nodes [3], and a UI for editing relationships.

The key is that the individual notes remain separate files and the note-note store is plaintext/rich text as well. [2]

One day…

PS. I think this was kind of what Gopher did

- fn -

[1] My own extended memory collection has moved through DOS text files, FileMaker Pro text base, PalmOS Notes, DateBk MemoAvantGo files, Outlook Notes, Evernote, Google Notes (killed!), Toodledo Notes/Appigo Notebook,  and Simplenote/ResophNotes/NotationalVelocity/nvAlt. No wonder I’m a nut on data lock issues and distrust Cloud solutions for extended memory even as I use them. Also: Before Simplenote, Palm Notes, iOS Notes, Keep, EverNote and OneNote there was Tornado for DOS

[2] Remember when Mac Classic gave every file its own unique ID? Those were the days. How to get the unique ID for the notes is the trick for a plaintext implementation especially across platforms. With rich text one can bury the unique ID in the metadata. Unique ID could be an IP6 URI.

[3] Remember when graph data visualization was a thing? That was the early 90s I think, around the time of VRML and MCF/RDF.

See also: 

(This is started out as a tiny post but I kept finding more old material I wanted to think about …)

Thursday, May 19, 2016

New York Times VR kit shows up in the mail

This came in the mail today.

Cardboard  1

I assume every subscriber got one, but I’ve not heard much chatter. The return address had Attn: Paul Ferrall on it; that seemed a bit curious. (Update: looks like they sent them out last Nov to Sunday print subscribers. Didn’t get much buzz!)

It’s a Google cardboard viewer of course ($15 retail). The NYT has launched a Virtual Reality channel; there are iOS as well as Android apps.

It was my first experience with Google Tech. I very clearly remember the last time I had this feeling; it was using a View-Master back when they they provided very high quality stereo images. There was a young woman sitting on the edge of a tall cliff with a boat far below.

The streaming wasn’t working today, but downloading was fine. Gigabit broadband would be handy.

The whale/echo video is the most amazing. I elbowed my daughter spinning about. First VR injury in our family.

It’s easy to see where this will go next …

Cardboard  3 

Monday, May 16, 2016

The Other Side of History: an audiobook

I just completed Robert Garland’s @48 lecture audiobook - The Other Side of History: Daily Life in the Ancient World (2012). It was terrific. The Great Courses has some very fine lectures, but this one set a new standard. I loved learning about how regular people worked and played - from Neanderthal to Medieval England. That’s where I learned that the pyramids were a Keynesian WPA project.

The one catch is pricing. The Great Courses have whacky prices — they periodically have 80% off sales. This course now lists at $250,  but on Amazon it’s $50 to $72 and on iTunes (via Audible) it’s $46.00. For that matter you could sign up for Audible.com, they appear to carry 430 titles from the Great Courses.

Just buy it and enjoy.

Sunday, May 15, 2016

Special relatively predicts the photon: Minkowski:Poincaré:Lorentz:Boosts from 2 assumptions & math.

From Backreaction: Dear Dr B: If photons have a mass, would this mean special relativity is no longer valid? I learn that while Einstein may have created Special Relativity with light in mind he could have done it in the dark.

To summarize …

  1. Make time a coordinate rather than a parameter. That produces 4 dimensional space time.
  2. Call that Minkowski space.
  3. Require that laws of physics be the same for all observers moving at constant velocity in Minkowski space (inertial observers).
  4. Mathematically, that means symmetry transformations are required. The group of these transformations is called the Poincaré group.
  5. The Poincaré group (again, math) has two subgroups. One is translation — all space-time points have same physics. That one is not so interesting.
  6. The other Poincaré group is called the Lorentz-group. It has rotations and boosts. Rotations in space aren’t so interesting, they just say all directions are the same. Boosts are rotations between space and time. Minkowski:Poincaré:Lorentz:Boosts give us length and time contraction and the like.

and so

Deriving the Lorentz-group …  is a three-liner … it is merely based on the requirement that the metric of Minkowski-space has to remain invariant … the  boosts depend on a free constant with the dimension of a speed. You can further show that this constant is the speed of massless particles.

… if photons are massless, then the constant in the Lorentz-transformation is the speed of light. If photons are not massless, then the constant in the Lorentz-transformation is still there, but not identical to the speed of light…. 

Giving a mass to photons is unappealing not because it violates special relativity – it doesn’t – but because it violates gauge-invariance, the most cherished principle underlying the standard model. But that’s a different story and shall be told another time.

Perhaps (this is me writing) another way of thinking about this is that making time a coordinate, and requiring physics have inertial constancy, gives us, through math, the properties of a massless particle. So if we didn’t already have personal experience with photons, we’d know what to look for. When we found them we’d then say that our science made a great prediction…

I’m really looking forward to Dr. Hossenfelder’s explanation of Gauge-invariance.

PS. OS X insists in quietly auto-correcting photons to photos.

Wednesday, May 04, 2016

What are the neanderthal diseases?

Europeans used to have more Neanderthal DNA than we have now ...

Ancient DNA from last Ice Age unveils distant ancestors underwent significant evolutionary changes | NH Voice

… the percentage of Neanderthal DNA in Europeans 45,000 years ago has declined from between 3% and 6% to around 2% in Europeans today. The researchers think that natural selection has reduced Neanderthal ancestry over time.

… Neanderthal DNA is slightly toxic to modern humans…

Perhaps Neanderthal DNA was more useful during the ice age, but is less useful now. Or perhaps it was never that handy — though hybrids were not rare.

It would be interesting to know what disadvantages/diseases moderns get from Neanderthal genes. Once we might have looked at gene products, but now we know things are rarely so simple. Genes are like letters in the English alphabet, not characters in Chinese. The letter ‘a’ contributes to both “Bad” and “Glad”, but those words (gene products) appear in many sentences and paragraphs (phenotypes).

There have been some early studies …

I’d wondered about osteoarthritis, but so far that’s not shown up.

Physiology in action - water loss on low carb diets

I love carbs. Bread especially. Ok, pastries even more.

I liked the days when diets were supposed to be low in fat, and even low in protein. (For every appetite there is a historic “best diet” recommendation.) 

Alas, carbs are somewhat out of fashion these day [1]. So when I am home, and being good, I control carbs. I also do a relatively extreme amount of exercise. When I am away I sin with carbs and exercise less — and my weight jumps 3-5 lbs very quickly. [2]. When I reform I “diurese”, which is a medically polite way of saying I pee a lot and drop the weight.

What’s up with that? I think it’s this …

Low-Carbohydrate Diets - American Family Physician

 …low-carbohydrate diets also initially induce significant water diuresis. The majority of this diuresis is likely the result of glycogenolysis from increased protein consumption. Glycogen binds water at a rate of 2 to 4 g of water per gram of glycogen. As glycogen stores are consumed for energy, two to four times that weight in water is shed through urine. Thus, a portion of the early weight loss in these diets is water weight.

I think my body is “good” at building glycogen stores, and not bad at using them when I reform.

 - fn -

[1] The Mediterranean diet has been fashionable for at least a decade (carbs as grains), though if the AIs ever figure out proteinomic-microbiomic networks we will probably have custom diets that may be quite different.

[2] I’m being circumspect, because I think my weight makes these sudden jumps after I quit sinning, decrease carbs, and increase exercise. If this impression is correct I might hypothesize that my body responds to the double hit of more exercise and fewer carbs by stuffing what carbs and fat I get into water rich stores. These then have to be depleted.

Monday, May 02, 2016

Things I didn't learn in medical school: Vertigo

About two years ago I had a sudden onset of vertigo. Between my wife (a family doc), my family doc, and me (also a family doc, albeit lapsed) we couldn’t tell whether it was “viral” labyrinthitis or “benign” positional vertigo. The more I read and experienced the less I felt there was a clear difference; there’s an ENT folk belief that the “benign” condition follows the “viral”.

Yes, the “quotes” are there for the usual reason. The more I get to experience diseases from the inside (yay oldness) the more my textbooks seem delusional. The semicircular canals (labyrinths) have a circumference around a cm. I couldn’t find any diameter on the average cross-section diameter of the actual canal, but illustrations suggest it’s on the order of microns. Even the very best research scanners cannot resolve these canals, much less the microscopic hair cells within them. The otoliths within our utricle and saccule are about 3 to 30 microns and can only be seen with a high resolution microscope. 

We really have very little idea what’s going on in the canals of someone with vertigo. We’re making this stuff up. (Hain’s article is more honest than most.)

My acute vertigo lasted a day or so. I had to delay a trip and I put off driving for several days. It was just as unpleasant as one might imagine, but short-lived. Many people live with far worse vertigo for much longer. 

Except it wasn’t as short-lived as I expected. After the first few days I couldn’t induce the characteristic nystagmus eye movements but I still felt “off”. I could induce a feeling of imbalance by shaking my head. My walking required much more attention than I was used to, though I had no problems with running. I couldn’t close my eyes and balance on a single leg (bit of a severe test for many at my age).

Most days I no longer notice these symptoms. I think that took about a year of extremely slow improvement with occasional worsening; Epley exercises (there are many variants) might have helped. I don’t know if the (presumed) insult to the hair cells resolved or my brain compensated for them or both; I suspect it was at least as much compensation as it was healing. I don’t think my closed eye single leg balancing is as good as it once was. I’ve taken a step to the fall-prone imbalance of infirmity (yay, age).

Living inside my relatively minor (so far) experience with vertigo taught me two things I didn’t learn from my textbooks. One is that our classifications of transient vertigo (positional, inflammatory, viral, etc) are probably bogus (not the only example). Another is that the symptoms may take months (years?) to resolve - or adapt to. Even knowing this I couldn’t easily find references on duration of post-vertigo symptoms, I found very few references, and they seemed to only deal with acute symptoms.

Our system of producing and disseminating medical knowledge is quite broken.