All posts by robertmcgrath

Book Review: “An Absolutely Remarkable Thing” by Hank Green

An Absolutely Remarkable Thing by Hank Green

Hank Green is a fairly prolific digital creator who I know nothing much.  (Sorry, I just don’t have time or interest for podcasts, YouTube, and so on.)  His new novel clearly draws on his experience (“write what you know”), with an interesting level of critical comment.

The story itself involves a weird series of events, apparently but enigmatically a contact from extraterrestrials.  The events are told by April May who “discovered” the remarkable thing, and publicized it via digital and mass media.

Green recounts the trajectory of April’s fame, as well as its effects on her, her loved ones, and everybody on Earth.  He offers both details of how “fame” works, and what it does to people.  And it isn’t pretty.

This is possibly the most important event in human history, contact with another civilization.  But most of the story is about us, people acting out and arguing and generally trying to figure out “what it means for me”.  This rings unfortunately true to life.

Much of the story makes little logical sense.  OK, if they are alien, we probably don’t understand how they think.  But still, much of the “communication” is in the form of mysterious puzzle solving, and the puzzles betray a deep connection to human psychology and culture.

If this process is a test for humanity, it is not clear what the point is, or how well we might have performed.   One is certain, though, we humans don’t seem to be able to learn much about the putative aliens.

But the crux of the story is April’s life, and specifically her philosophy of life and attitude toward the Remarkable Thing.  April makes a bunch of mistakes, damaging her own relationships and getting sucked into the extreme addiction to attention.  On the other hand, she also articulates a positive, collaborative view of humanity.

I kind of like April and her posse, so it is painful to watch her troubles (some of them self-induced).  Things are not really resolved by the end of the novel, and we are left hoping against hope that it will be alright.

The ending is ambiguous enough, though, and there is certainly room for a sequel.

Green is a good writer (OK—he seems to be a multifaceted “creator”), and even though I didn’t particularly identify with these characters or the situations (honestly, most of it isn’t so much “remarkable” as “unbelievable”), but I still was swept along.  From his experience he seems to have interesting insights into how collaboration works, which he portrays nicely without too much preaching.  This isn’t especially profound, but I liked it.

  1. Hank Green, An Absolutely Remarkable Thing, New York, Dutton, 2018.


Sunday Book Reviews

Book Review: “The Labyrinth Index” by Charles Stross

The Labyrinth Index by Charles Stross

The latest story from Stross picks up events after the catastrophes in The Delirium Brief.

The story follows Mhari Murphy’s troubled career under the New Management. In this case, she is assigned to lead a cunning operation to penetrate the US, which appears to be under its own, competing and extremely dangerous New Management.  While the UK’s new management is murderous, the US seems to be marching toward annihilation of the solar system and all life therein.

Things are bad, and get worse as the operation unwinds.  Apparently, management hopes to throw off the opposition with inexperienced leadership and a tossed together band of bureaucrats and technicians.

It a gripping tale, full of Strossian touches.  For example, he raises the very good—and deeply scary—question of just what all these extradimensional invaders are running away from. It must be something much, much worse than the horrors they are bringing to Earth.  That’s not a pleasant thought.

At the end, not everybody dies, though several people are infected (and therefore possibly not “people” any more). There are still a number of shoes yet to drop, so I really hope Stross is not done with the series.  There sure are some things we’d like to know!

  1. Charles Stross, The Labyrinth Index, New York, TOR Books, 2018.

Sunday Book Reviews

Antarctic Hotspot Under the Ice

As the ice melts all over the Earth, particular attention centers on Antarctica.  If and when the Southern ice cap melts, the oceans will rise by tens of meters, drowning much of where humans live.  When will this happen, if ever?  We need to know.

Antarctica is a continent, and that means it is big and complicated. Some places are melting fast.  Others aren’t changing much, and some may actually be accumulating more ice.

One of the unknowns is what is under the ice.  We can easily see the top of the snow and ice (at least some of the time.)  But what is going on in the kilometers deep ice, and what is at the bottom?

The is actually a critical question for several reasons.  Ice that is resting on dry, cold, rock will be solidly anchored and will generally melt only where exposed to air and ocean.  But ice resting on water will melt and slide downhill, generally ending up in the ocean, where it will melt.  And ice resting on warm rock will melt from below.

In Greenland, there are areas where volcanic heat under the ice is thinning and weakening the ice, and speeding the rate of melting.

This fall, researchers from Europe report on studies using airborne ice penetrating radar near the South Pole [2].  The study identifies an area that is a geothermal hot spot under the ice, where the overlying 2+ km deep ice cap is sagging.

The saggy spot is approximately 100 x 50 km, and the sag is consistent with melting of about 6 mm per year.  This seems to be much more than would be expected from the rocks, so the researchers suggest that there is geothermal flows of hot water from deeper rocks.

This finding is important for numerical models of the ice, which generally use a uniform, lower estimate of the heat of the rocks.  This hot spot, and other similar anomalies could have substantial effects on the changes to the ice cap.

The study is also interesting because it uses ice penetrating radar to create a three dimensional view of the ice, from which they discover the ‘sag’ and inferred the estimated rate melting.  The result gives a way to estimate the heating under the ice, even where no direct measures are available.  (And direct measures of the rock under the ice cap are very difficult to obtain.)

And, as noted, in this case the estimate is much higher—more than twice as high—as previous estimates.  It is noted that these results may explain the presence of numerous subglacial lakes in that area.

The details of the techniques are beyond my own paltry understanding of either ice or radar. Given the complex dynamics of the ice cap, there will need to be additional studies and cross validation to make sure these estimates are accurate.

These results will also guide ongoing plans to sample the ancient ice from this area.  It is thought that this area has some of the oldest ice on the planet, and examination of embedded air bubbles would reveal a history of climate far into the past.  However, the ice in this warm, sagging zone will have remelted and would not retain the ancient bubbles.  So the search for records of the climate should avoid these areas [1].

  1. Jonathan Amos, South Pole: Rock ‘hotspot’ causes ice sheet to sag, in BBC News – Science & Environment. 2018.
  2. T. A. Jordan, C. Martin, F. Ferraccioli, K. Matsuoka, H. Corr, R. Forsberg, A. Olesen, and M. Siegert, Anomalously high geothermal flux near the South Pole. Scientific Reports, 8 (1):16785, 2018/11/14 2018.


Tribal Bison Summit on National Bison Day

November 1 was National Bison Day in the US, a day to celebrate the American Bison.

This year the day was marked by the first Tribal Bison Summit, which brought together tribes and conservationists to plot future work.

The meeting was marked by a ceremony that evoked the spirit of the Buffalo, which sounds like it was a very fitting way to celebrate National Bison Day.

The spirits were with us,” said [Eastern Shoshone member Jason] Baldes. “The eagles were flying over the top of us as well as cranes. It was kind of an emotional and spiritual moment to kind of convene the summit in that way through ceremony.

The group is looking to expand the rewilding of Bison in the American West, which is not only reviving nature, but is also reviving and healing the cultural heritage of the peoples who lived with the Bison for so many years.

  1. Defenders of Wildlife, First Tribal Buffalo Conservation Summit to be Held in Denver, in Defenders of Wildlive – News. 2018.
  2. Melodie Edwards, At Tribal Bison Summit, Advocates Look To Future, in Wyoming Public Media. 2018.


“Blockchains for Science”

Or is it “CryptoTulips for Science”?

I’m a long time veteran of scientific computing, including the extremely tough problems of electronic publishing and knowledge dissemination, which requires good ways to deal with provenance and trust.

So I’m a bit surprised to learn that there is a magical solution to these insanely wicked problems:  blockchain!

Specifically, the First International Conference on Blockchain for Science, Research and Knowledge Creation happened this month in Berlin.

This appears to be a conference of hammer-makers, so everything looks like nails to them. : – )  The “hammer” is the blockchain, and digital science looks like a box of nails.

According to the conference prospectus, the blockchain is more of a Swiss Army Knife, which can “to reestablish trust into scientific data”.  Maybe it can “fix the reproducibility crisis”.

“As the ‘trust machine’ Blockchain in Science bears the potential to reestablish trust into scientific data. Some claim that it might even be good to fix the reproducibility crisis. New ways to rethink research subject privacy and whole data marketplaces are on the horizon. Blockchain might even play a large role in scientific publishing, where it questions the current role and business models of scientific publishers. New ways to incentivise peer-review or reproduction of results may arise. “

More plausibly, it might be useful for “data marketplaces” (assuming that scientists can afford to participate).

And it might also be a useful begging cup to help finance publishing and peer-review.

Much of the conference program is about blockchain technology (“get your cryptotuliips here!”), not so much about the alleged problems to be solved, let alone working solutions.

I’ll note that I don’t see any of the big names in Provenance or escience  (e.g., Sensei Carole Goble really should have been a key note speaker, IMO)  How can you talk about trust and reproducibility if you ignore the work that has already been done?

The thing is, blockchain qua blockchain offers little that can’t be done with conventional data systems plus public key cryptography.  (I’ve said this before, and I’ll probably have to say it again.)

In fact, blockchain technology is a terrible fit for a lot of the problems, just as it is for other applications.

Science is a tiny, underfunded enterprise that does not need a global public blockchain.  Science deals in weird, unique, often bulky data that ain’t never going to be on the actual blockchain.  Reproducibility requires incredibly complicated records of information flows and processes, which could be recorded on a blockchain, but probably shouldn’t.

I’ll note that one of the most crucial operations of science is revision and retraction of errors.  Blockchains cannot support that operation at all.  Flooding the world with fake data that cannot be removed is not going to be a real advance.

IBM’s contribution to this field reflects this very  fact:  Zach Seward reports that IBM has registered a patent for using blockchain for “open science” [1]. IBM’s idea of “open” science may not be the same as mine, but they certainly understand the problem.  Their system is basically a change log for data, reports, and whatever. This approach tracks and makes a public record of who did what, including corrections and retractions.

The thing is, of course, that we already have change logs (see, perhaps Github, which is the fourth or more generation of this technology), and we have been using them for several decades now. We were working on these “digital notebooks for science” fifteen years and more ago [2].  I assume that the new thing is implementing this on a blockchain which is technically clever but who knows if anyone actually needs it.

And, by the way, “science” is not a single enterprise, it is a bunch of small, inbred communities. I might have use for some kinds of data, but almost certainly have no way to even understand 99% of the data out there. So the blockchain will carry rafts and rafts of data that only a handful of people actually are able to use any part of.

It is interesting to consider that “scientific consensus” bears no resemblance to Nakamotoan “consensus”, because—it’s too complicated to go into here.  The point is, scientific results are not valid of important because the author thinks so, or because of the number of downloads.  The blockchain may assure accessibility and tamper resistance, but the evaluation of results will still work the old fashioned way.

For example, take a look at Wikipedia.  It is a giant change log.  There is a public record of who did what.  This has worked amazingly well for a long time–without blockchain.  Reimplementing it on blockchain would do nothing much, because the hard parts of Wikipedia are what the humans do.

I wonder if some of these notions about “incentvizing” publication, reviewing, and replication are a good idea or not.  I understand why there is a temptation to scramble for funding, but it is a slippery slope to put science on a market driven model.  This must inevitably distort what is done and published, rewarding trendy and politically favored topics, and starving less popular work. It may also be used to further cut public funding, on the excuse that “those scientists are raking in all that cryptocurrency”.

To me, this conference looks like a bunch of Tulip merchants trying to convince people to buy their magic CryptoTulips.

I may have to create a special CryptoTulip of the Year citation, for “type 3 CryptoTulips”. A “type 3” error is “asking the wrong question”, so a “type 3 CryptoTulip” is a confident solution based on misunderstanding of the actual problem.

  1. Zack Seward (2018) IBM Says Blockchain Can Power ‘Open Scientific Research’ in New Patent Filing. Coindesk,
  2. James Myers, Luigi Marini, Rob Kooper, Terry McLaren, Robert E. McGrath, Joe Futrelle, Peter Bajcsy, Andrew Collier, Yong Liu, and Shawn Hampton, A Digital Synthesis Framework for Virtual Observatories, in UK e-Science All Hands Meeting. 2008: Edinburgh.

Cryptocurrency Thursday

“Shared Economy” in China: The Confusing Definitions of “Sharing”

I was intrigued by a headline about “China’s burgeoning sharing economy”[1].

For grey hairs like me, the notion of communist China developing a “sharing economy” is head spinning.  At the height of the communist era, there was no non-shared economy, at least theoretically.  Of course, China now has a burgeoning economy with a lot of private and corporate ownership.  So there is room for the reintroduction of “sharing”, as reported by Iris Wang [1].

Wang’s article examines several cases, and they reveal the ongoing confusion in the use of the term “sharing economy”.  The original concept of the sharing economy was a peer-to-peer, user owned, system.

One of the enterprises mentioned by Wang is a clone of Uber, which isn’t so much “sharing” as “share cropping” and monetization of worker owned resources.  This may be economically important, but it ain’t good for workers or poor people.

Another example is Vcomic, which is a digital comic book market.  Owned by and plugged into Weibo, this is the Amazon version of “sharing”—one giant gatekeeper lets individuals sell stuff to consumers.  Maybe it works more like Patreon (if Patreon was owned by Amazon), but  I’m not sure why this is considered part of the “sharing economy”.  (It’s part of the “monetization economy”.)

Another case is a “peer-to-peer” housekeeping app. This is basically a digitized employment agency.  In fact, it appears to be a company that trains and manages a pool of workers, using digital technology.  Again, it’s not clear to me how this is “sharing” in any meaningful sense—its an employment market.

Finally, there is the case of a home cooking “meal sharing” app.  This one offers on-line ordering, food delivery or pick up.  A third option is to eat a meal in the cook’s home.  The latter is a classic “sharing economy” feature, the other two are identical to every food joint in town.  Furthermore, the operation is run just like a franchise restaurant group, except people work out of their own home (and, by the way, branding their persona kitchen).

The enterprises discussed are interesting and may turn out to be important to the workers and consumers of China.  But I’m not sure they should be called a “sharing economy” in the original, altrusitc sense.

They all use digital technology, and they all require the workers to provide the means of production, while raking off fees for running the “market”.  These are basically a cargo cult “sharing economy”, using the digital trappings without the actual worker ownership or control.

It is also glaringly ironic that these “peer-to-peer” enterprises do not seem to be worker- or user-owned.  Even in communist China, workers do not own their own enterprises.

This is particularly unfortunate because the peer-to-peer digital technology works just fine for cooperative enterprises.  So these could be organized as coops or collectives, or whatever.  But they aren’t.

This is the biggest confusion of all:  a “sharing economy” should be about shared ownership and control, not about digital interfaces and exploitation.

I’ll note one additional very Chinese wrinkle.  This article about the “sharing economy” in China reports that this information comes via the State Information Center.  The concept of a peer-to-peer, user owned, system is basically the opposite of the central planning of the Chinese government.

  1. Iris Wang (2018) A glimpse at China’s burgeoning sharing economy. Sharable,


PortSmash: Yet More Leaks

The drum beat of computer security problems just never ends.  Most of what we hear is just consumer grade dumbness—phishing attacks, misconfigurations, and so on.

But deep inside, there are wonderfully sphincter tightening holes, things that show just how far we are from creating perfect systems (e.g., this, this, this, this, this)

This month, there is yet another interesting security problem identified, tagged “PortSmash” [1].  This particular problem is a side-channel, which apparently can leak information from one program to another.  As they say, “Port Contention for Fun and Profit”.

This one is interesting because it has nothing to do with memory or IO, but rather is a subtle effect of the internal design of the Intel chip and clones.  The nickname “PortSmash” refers to the issue:  inside the processor, there are resources shared by different computations including “ports” to the internal busses.  Since they are shared, activity by one program can interfere with another program that needs the same resource. For example, a program may be slightly delayed at certain points because some other program is running.

These collisions are far out of sight of the program logic, but the research shows that they can be detected.  And they may be correlated with what the other program is doing, thereby subtly revealing potential secrets such as encryption keys!

This is pretty cool, if extremely obscure!

The big picture here is that computer architectures have become miniaturized, but that mainly means that the designs we built in macroscopic modules and systems software in the 1980s are now built in to integrated chips, accessible through simplified interfaces.

So all the potential bugs and side channels have been pushed inside the (almost) opaque insides of these massively complex chips. The chip has many functional units connected by internal busses and complex internal communications.

In the case of PortSmash, the internal architecture is multi-threaded, allowing more than one program to be running at the same time.  Having pioneered multithreading at the high level, I know that it is quite tricky to get right, and inherently involves multiple programs sharing resources. Back in the day, we didn’t even try to deal with timing anomalies or side channels, which are surely rampant in any multi threading systems.  (You want security?  Control access to the system.)

Anyway, nice work all.

I’ll note that the report comes from a research group that has a mission and project to investigate precisely this sort of side channel. So we can expect further examples to come to light.

  1. Alejandro Cabrera Aldaya, Billy Bob Brumley, Sohaib ul Hassan, Cesar Pereida García, and Nicola Tuveri, Port Contention for Fun and Profit. Cryptology ePrint Archive 2018/1060, 2018.