30 November 2013

Flying Blind

Another, not so, happy coincidence. It's been a while since I visited Nick Carr's blog, and we get two very unique and related posts.

First, the FAA admits pilots don't fly much, anymore.

Second, programmers don't write code much, anymore.

The first shouldn't be too surprising, while the second impinges more directly on this endeavor. I've recounted recollections (without a cite I can claim, alas) of studies done in the, I'll guess, early 90s that GUI-fied applications (Word, you loathsome bastard) deflect the user's attention from content to eye-candy. And, thus, earn my enmity. The comment stream is nearly as interesting as the post and the linked article of Nick's.

What I conclude from my experience, the post, and the comments:

1) Even coders who acknowledge that there might be a problem with automating coding don't get that the code structure is a function of the code generator. While the discussion was largely directed at IDEs, the more insidious problem is with application frameworks: RoR, Struts, and the like. IDE/framework is a crutch. The application of "higher level thinking" isn't in the use of such crutches, but in their construction. Tool maker versus tool user, once again.

2) None see the extension to The Great Recession, which is that "higher level thinking" is just codeword for "let's stop doing real brain work, and just all get rich doing finance thingees". Possibly a larger indictment of the quant squad, but it's the coders' fault, too (much of Wall Street runs on C++). One must not forget the Li issue, as told by Salmon. Recent times, last 3 decades, have been largely about devolving from productive employment (actual rocket scientists) to leech employment (automated financial despoilation), all in the name of "higher level thinking". Other than process size improvements, largely an engineering issue, not science or discovery, not much can be said about "higher level thinking" having moved civilization forward. Are Apple's round cornered rectangle patents progress? Are such "inventions" what we should be showering with the maximal rewards (enforced by the legal system, of course)? One comment mentions chip layout, but skips over the relevant part: one can still find photos of chips being taped out (1970s), with real mylar and real hands; these days one uses builders which offer up ever larger blocks of circuits which an "engineer" (for those few still doing such) Lego blocks (verb form) together. How those circuits actually work, and whether there's any alternative, is gone. Whether this Lego blocking truly implements "higher level thinking" is not obvious, to me. There's a reason most everything on a platform looks the same. Apple, in particular, asserted as a marketing meme that a "seamless experience" was justification for uniformity in applications. As biologists well know: monoculture leads to doom. This is not to assert we should all revert to various assemblers (and not nearly as many as there were in 1980, say), but it's equally true that most OO coding (mentioned by more than one comment) isn't OO at all, but olde style FORTRAN function/data with cuter syntax. Even the OO revolution was a sham.

3) Much of the defense of the status quo involved citing the likes of Instagram! Facebook, twitter, Google (even) and such are toys, not tools or anything actually useful. The great minds of the younger generation worry about keeping "in touch" with friends. Somehow, I don't think Einstein had that as a priority. Nor did Crick and Watson.

4) Humans have run up against an immovable object: there's little left to discover in the real world. We've found all the land, resources, and physical laws. We're on the edge of miniaturization of nano-object construction. There is no warp drive or time travel. There is no new continent to exploit (modulo, possibly, Africa), at least in the previously unknown sense. We haven't had a plague in centuries, so we're well past the heel of the population hockey stick, again. Again, the rocket scientists of today are in finance devising ever more opaque methods for extracting moolah for themselves from the productive economy.

5) Oddly, none see the explosion of healthcare.gov as indicative of bloat and inefficiency engendered by such automatons, which by their nature are code-centric rather than data-centric. (There's this disputed NYT report that the system, per se, is 500 million lines. Not likely. There could well be that much including all the foreign systems it connects to, though.) Back in the old days, it was the WinTel monopoly: Intel needed MicroSoft to make ever more greedy software to justify buying a new machine with a fancier X86 chip, while MicroSoft needed ever more bounteous cpus to run the clusterfuck it called Windows. Whether ARM or Intel or whoever will find a needy compatriot is an open question. That we can see the physical limit of node shrinking is not in question (and no, nanotech isn't the answer; once you get to the single atom as we nearly are, the next step is inside the atom which straps your ass to quantum mechanics and that's nowhere you want to be). Given my predilection, the continued ascendence of coding, as opposed to DRI in RDBMS, is the highest mark of IT's regression. While the Kiddies have all their new languages, they've reverted to the paradigm of COBOL/VSAM development that Granddaddy used. Where COBOL was invented, intentionally, so that (non-programmer) managers could read and understand what the application did, today's IDE/framework puts the coder in the same frame of ignorance. Just jam a bunch of them code blocks together. This is intellectual regression, not progression. And they haven't a clue that they're flying blind.

6) None mentioned Knuth.

29 November 2013

Angry Birds

When these endeavors began, the first post asserted "Why the Stimulus Will Fail"
In 2009, who are the unemployed? Not, by and large, workers in factories that will make goods for American consumers. The deindustrialization of the economy, in progress since the 1970's, makes any stimulus program a low probability gamble. Will the stimulus program re-employ the leeches in the financial services industry that sent us over the edge in the first place? It is important to realize that this sector of the economy had grown to a very large proportion; possibly unprecedented. Even if the Federal government chose to reward them with new employment, what is it that the Federal government could buy from the sector? Variable annuities? Would those who are re-employed as a result of the stimulus program spend their newly increased incomes in the financial services sector, thus re-employing all those folks? Would that be rational?

In later postings, which don't come to mind (and given the allusive nature of post titling used ...), this notion was made more specific. Any economic recovery from depression/recession faces a fork in the road early on: either re-inflate the businesses which had been employing those made redundant by the collapse (even, and especially in the current case, those who caused the collapse), or move the economy into new (and by intent anyway, less silly) modes of production. I've returned to them in both versions of this endeavor on occasion. Re-inflation is easier to do, since it rewards the wealthy miscreants, and that's exactly what the QE exercise has done. Total employed has barely changed over the course. Capital gains, for those with capital naturally, have exploded. Such a country!!

In sum: the raptor-quants intend to snip the gonads off the only part of Dodd-Frank that deals with the root cause of The Great Recession. They want all mortgages to be classified as exempt from the stricter rules (which aren't all that strict, particularly in historical context). They want to go back to 2004 when they could slurp jeroboams of moolah from the stream going from the saving class to the borrowing class. A zero-sum profit machine.

As Floyd Norris reports today, the Banksters, in harmony with builders and even some titular Left Wingnuts, have been hard at work to "Make it So", putting the oomph back into Banksterism.
The rules on qualified mortgages are meant to assure that consumers can afford them, and the requirements are rather low. Lenders must go to the trouble of verifying a borrower's income, and the total monthly debt obligation must be no more than 43 percent of pretax income. There are no requirements for down payments, or limits on how much is lent relative to the value of the property.

He later quotes the figure I've generally used: 35% as the operational rule of thumb. The new rules define three classes of mortgages, with the best having no restrictions. Naturally, the Banksters want all mortgages to magically qualify. "Welcome to Lake Homebegon, where all the Banksters are strong, all the men are feckless, and all the mortgages are above average."

So, the part of Dodd-Frank that's under assault is that which intends to rein in profligacy by home mortgage issuers (be they banks or mortgage companies). Both don't want to go back to the Good Old Days (which they always do, mostly) when savings' banks lent most mortgages and held them to term. Not much moolah to be slurped up that way. Better to make loosey-goosey ARMs and such, and sell them off for a quick few bucks. The builders don't want the new rules, since such rules reduce the number of high value qualifying houses; higher price thus higher profit. Just ask Apple how that works.

The raptor-quants are starting to feel hungry and need a hearty meal. I suppose it's just a coincidence that we hear about their hunger pangs at Thanksgivanukkah.

27 November 2013

Another One Bytes the Dust

OCZ has filed bankruptcy. No STEC, no OCZ, Fusion-io in the tank, ditto Violin Memory. What's a fast relationalist supposed to do? I know, NoSql flat files. Yeah, that's the ticket.

By the end of 2014, SSD will have coalesced into Samsung, Intel, and may be SanDisk. Such a disappointing revolution.

Snake Bit

Bitcoin is in the news again. Twice in the NYT in a few days (once, and twice).

What none of the pundit class has discussed, in the cyberink that comes my way, is what a fixed amount of currency (21 millions "units", as I understand it) would have on a global economy? If the "exchange rate" (e.g. $35/ounce as gold once was) is fixed, then dee-flation is the rule. It has to happen, since as commerce expands both across nations (growth by accretion) and within nations (growth by procreation) prices have to aggregate to the amount of currency available.

If one looks at 19th century USofA, when specie money was the law, that's exactly what happened. There was localized inflation in mining areas (shopkeepers charged steep prices for necessities, and more for amenities), but national deflation since the economy was expanding along with its territory. With a nearly fixed amount of specie, relative to economic activity, aggregate prices had to multiply out to that specie hoard. Holders of specie made out, but the rest of the economy suffered in debt that grew in real terms. Not a process that aided Main Street. Not to mention tech/science was creating truly new forms of goods and commerce. And, no, today's cybershit isn't remotely as revolutionary to this era as coal fired steam traction and petroleum introduction and air flight and telegraphy and on and on to its.

Should bitcoin, and any number of other fiat currencies (why, one might wonder, is it OK for private issue fiat currency to exist, but not sovereign?), displace national currencies, batten down the hatches, Bro! The Dark Ages will have returned. The Winklevossen will be the liege, and the rest of us serfs.

26 November 2013

Life and Death on the Nile

One might ask, "why is it that CMS would opt for a twelve year old, still private, still VC slurping outfit to 'run' healthcare.gov?" Why would such a tactic make sense? I mean, they could go belly-up at any moment; although having insinuated themselves into CMS might get them to Too Critical To Fail status. Well, the answer appears to be, inertia.

More than a couple of decades ago I worked for Optimed Medical Systems (no longer exists separately, and last I knew had been inhaled by a competitor after stops elsewhere). This was a Progress shop, and was known for the cadillac pre-qualification software of the time. VT-X00s connected to *nix and the Progress "database" run C/S in-a-box style. We made fitful attempts at a sort of AI module. Prolog would have been the proper language, but that's another episode. HL7 was then relatively recent, and beginning to be widely adopted.

What I found odd, and never reconciled, was the medical community's love affair with hierarchy. The embodiment of this affair is HL7. The Wiki article is reasonably comprehensive. Here's a cautionary tale. IBM flowchart support diagrams, (an example on page 29 that looks just like a pyramid!) dating from the 1950s, are perhaps the earliest existing example of hierarchy blessed by The Smartest Guys in the Room. Dr. Codd faced significant opposition, and IBM paid him back by appointing an IMS cowboy (Chamberlin) to devise the query language.

As with any hierarchy biased datastore, finding *relations* among all that data is hard. And it's relations that we care about. R folks, and quants generally, take a probabilistic view of data relations. There's X% of people who buy diapers and whiskey together, so let's use an animated baby in our ad for Uncle George's Fine Moonshine. And get Uncle George stocked in grocery stores, and diapers in liquor dispensaries. RDBMS folks view relations as deterministic: order lines *will have* an associated order. The pyramid folks get all bent out of shape with this kind of assertion, blithely ignoring that their hierarchical datastore forces all data into *but one* deterministic structure. Some people...

From a diagnostician's point of view, the IBM 1401 flowchart (or its railroad equivalent) establishes a reasonable model for decision making: is the patient breathing? And so forth. As a datastore "model", not so much. The medical community had a serviceable hammer, and made all else into nails.

So, in a world of folks who insist that all data be pyramids, forcing MarkLogic on the developers isn't all that surprising. As usual, policy beats the crap out of data when there's a disagreement.

24 November 2013

Marky Mark and the Funky Bunch

So, who or what, the hell is MarkLogic? For one thing, it's sometimes Mark Logic and other times MarkLogic. Since I left my Clark Kent memorial fedora, complete with press pass, at Superman's phone booth, what follows (modulo my speculation) is gleaned from the InnterTubes. In sum: not so sure I'd bet the farm on these guys for an HIE style application. Somebody, whom I've not identified, is responsible for choosing them to run that "core" of healthcare.gov. Them new duds the Emperor's got sure impressed somebody.

Its site says it's been around since 2001, yet slurped $25 million more in VC money recently. Twelve years in being the greatest thing since high button shows, and still living off the VCs? That and the named managers, from CEO on down, turn over about as fast as "want fries with that?" voices at Mickey D's. Something's not quite right. Still private, so financials aren't available. By comparison (and MarkLogic's promotional lit goes after them), when Oracle was ten years out from its first release (1989) it was a public company with $584 million (in 1989 dollars, mind) revenue. "And you don't mess around with Jim."

Let's start with some promotional slides, circa 2011, here. The slides are marked "confidential". Hehe.

Slide 23 - they're touting the CMS Health Insurance Exchange: "Massive amounts of data at high velocity; highly transactional" not so much, when the rubber hit the road

Slide 34 - they admit (though likely not bright enough to know it) that they've re-invented IMS: "Information is stored as XML (hierarchical)" Kiddies just can't stop building pyramids; perhaps we should rename these Madoff Information Systems (MIS, rebooted from the 70s)

What the Kiddies don't get is that hierarchies (whether in semi-binary as IMS, or clear text as xml) are built to support *a single*, hard coded, optimized access path. And that's for reading. Any other path may be impossible, and will be at least difficult. Changing the structure of the hierarchy requires changing any code which uses said hierarchy, since some (even most) Stations of the Cross have been moved. Dr. Codd figured out the problem: with an easily fungible relational structure (and sensible queries) adding tables or columns will be transparent to existing queries; those that don't need the added data may ignore it. Orthogonal is a very good thing. Too bad the Kiddies never figured that part out. The Kiddies insist that re-inventing IMS is progress. Yeah, right.

The tenor of the deck (hard not to insert the obligatory r) is MarkLogic as a Master Data Management implementation, although that term is kind of passe' most places these days.

Another of their buzz sentences: "Database and search engine are the same" as if this were any different from real RDBMS, or IMS for that matter

Since I'm willing to risk being accused of ad hominum behaviour, here's what the founder, Christopher Lindblad, did before MarkLogic
Chris founded MarkLogic while working as an architect on Ultraseek Server, Infoseek's enterprise search application, which is now one of Autonomy's principal products following the acquisition of Verity.

Here's the dirt: Infoseek was crap, Ultraseek got passed around like a plate of week old baked beans until Autonomy ultimately ended up with it, and who was the mess that H-P way, way overpaid for. And, "principal product"?, no it isn't: "Under Autonomy, Ultraseek continues to be developed and marketed as Autonomy's entry-level keyword-based site search offering." Not something to brag about, no. "Try not! Do, or do not, there is no try" The point: HIE is an OLTP application, not a Google. If the only tool you have is a hammer, everything tends to look like a nail. Not do OLTP from a doc store; there is no try.

As I posited in an earlier post, Kiddies assume that search *is* computing and that hierarchies are the bee's knees of datastores; and they aren't. While there may be some part of ACA implementation that will benefit from blob/clob data bits, the guts are OLTP/POS. And they ain't nothing better than an industrial strength RDBMS for that. If it starts in Organic Normal Form™, morphing the schema as you go isn't such a big deal. Lots of schema migration apps out there to help, if you feel the need. Remember: orthogonal is your friend. Cleave to orthogonal; as the adverts used to say, "Calgon, take me away!" Hierarchies ain't, they're as mixed up as a Christmas fruit cake.

The really big deal here is the federation with Federal, state, and insurance company systems. That's the crunchy part. There are multi-database federation protocols out there, including in SQL-2003; it's called Foreign Data Wrappers in PG terms, SQL/MED (Management of External Data) officially. DB2 provides the most robust, cross-vendor federation (likely because it has to!). Oracle does too, of course using its own semantics (took them a while to wean themselves off CONNECT BY for CTE syntax, of course). Progress (you remember it, right?) offered federation in the early 90s.

This is a classic distributed database application. Full stop. No need for NoSql. There is no try.

23 November 2013

Never Send A Boy to Do a Man's Job

About a month ago, I mused that the Kiddies were the cause of healthcare.gov's abject failure.
I haven't yet found more than vague descriptions of the architecture of healthcare.gov (herein, hc.gov), but I'm willing to bet a quid that it's a labyrinth of xml data streams exploding all across the InnterTubes.

Turns out to be worse than even I surmised. CGI has been an Oracle based application developer for a long time. Turns out that some part (don't know the extent from this reporting) of the site is run on MarkLogic, a *xml/NoSql* "database". Here's their site. Nailed it!!

Another sore point was the Medicare agency's decision to use database software, from a company called MarkLogic, that managed the data differently from systems by companies like IBM, Microsoft and Oracle. CGI officials argued that it would slow work because it was too unfamiliar. Government officials disagreed, and its configuration remains a serious problem.

Come on guys!! One knows that the NYT isn't a techy journal, but a bit more specificity would help citizens understand how bad this idea was. CGI has a marginal reputation, but this just tied both hands behind their backs. To repeat: reporting so far doesn't describe where in healthcare.gov the MarkLogic "database" sat. We also don't yet, from reporting, know whether this MarkLogic store was the central repository; although "yes, yes it is" seems a reasonable inference.

According to this reporting before 1 Oct:
It is also the core back-end database powering the new federal healthcare exchange that is set to open October 1. That system has been called the world's largest information integration project, says Aaron Rosenbaum, MarkLogic's director of product management.

No definition of what "core" means here, but since we're talking NoSql, I'll assert that "weakist link" covers most of what matters. I'd gloat, but hc.gov is way too important for that. It also means that Kiddies are easily flummoxed by Emperor's New Clothes. Back before the Triage piece, I was invited to the DNC IT folks to talk about databases and stats. This was preceded by phone conversations about databases and stats. When I got there, I suffered through a couple of 20-somethings who only wanted to jabber about Ruby on Rails. Didn't go well, and I wrote up Triage in a fit of pique. DNC proceeded to blow it in the state elections just as they did in 2010, although the absolute number wasn't as bad. Given the Obama coattails, I'd argue that DNC blew it worse. Off years are always difficult, but presidential years such as 2012 should be a springboard. Not 2012.

Then, this:
"We're doing grown-up, dull sorts of things," Todd says, "but also cool things that other NoSQL players aren't able to start thinking about because they're still playing catch up on the enterprise stuff."

Um. No, not grownups. Never send a boy to do a man's job.

20 November 2013

Party Like It's VT-100 Time

Those handful of folks who've followed this desultory philippic know that I've been banging a drum for Organic Normal Form™ databases running on multi-processor/SSD boxes from the beginning. As connectivity and bandwidth have progressed from dial-up to today, the infrastructure for VT-100s connected to *nix databases over a kind of long RS-232/RJ14 wire gets ever closer to reality. Whether COBOL indentured servants or Kiddie Koders (for the same reason, but different paths to that reason) will ever embrace the paradigm still isn't a lead pipe cinch.

Which brings us to the news of the ECMAScript development plan.
Citing other Web-related developments, Eich said WebRTC, a specification equipping browsers with real-time communications apps via JavaScript APIs and HTML5, would become a standard. The work of Google, Mozilla, and the IETF will make this standardization happen, said Eich, currently CTO at Mozilla. "It's really cool, and you can use it for data communications, too," he said, noting that browser support for WebRTC varies, with Mozilla's Firefox a bit behind Google Chrome.

While progress requires change, not all change is progress. In particular, the dumb data on a client with smart code (also on the client; fie!) paradigm embraced by both COBOL dinosaurs and java/PHP/fooBar Kiddie developers. Resistance is futile. You will be assimilated. Render unto Server that which is Server's.

The Hard Way

Just saw this post from r-bloggers. Since Heroku is Postgres based, I've left the following comment:

Why not just use PL/R? Would be a lot less work. I've built a small project using PG and PL/R, and it works as advertised.

17 November 2013

Not Such a Nobel Man

Lars Peter Hansen is the Other Guy who got the Nobel in economics this year. Mostly been quiet about it, although it was generally understood from the outset that he was somewhere in between Shiller and Fama (get it?) on the ideological spectrum. Well, today Jeff Sommer in the NYT has a bit of an interview. I'm disappointed. This is a Nobel Man? He labours at Univ. of Chicago, which makes him, a priori, a Freshwater economist. (That's another way of saying Right Wing.)

As with most econometricians of late, he ignores policy (or, as some do, provides cover for his preferred version), and toes the stats line.
Earlier in his career, along with Professor Sims and especially with Professor Sargent, he provided some of the mathematical underpinnings for what is known as the rational expectations theory -- the notion that people use all available information in making economic decisions.

Nonsense. Capital, in particular, makes decisions based on gaming the rules of engagement. To posit otherwise is foolish. If they really were "rational" there'd never have been a Great Recession, since no rational Bankster or CDO insurer would have touched those mortgages with a ten-foot pole. The mortgage companies wouldn't have made them, either, since such instruments were bound to fail. While the macro folks in New York and Washington might have willfully ignored the fantastic unsticking of median house price to income ratio, local real estate agents couldn't. The numbers were in their faces every day. Early on in the run up to the crash, I read/saw some reporting in which the reporter asked one of these how the buyer could possibly carry the mortgage he just sold them. His response: "I don't care". He didn't care because he'd gotten his gelt, and from then on the problem belonged to someone else. Such a country!!

One might, and some have, argue that everyone from the local real estate agents on up to the CEOs of Merrill and AIG and so on rationally expected to get a Washington bailout when it all imploded. But that's both specious and tautologous. No one who hews to the homo economicus myth can argue that it is rational to depend on welfare.
It suggests that people may not "be fooled by policy makers" into making decisions against their own self-interest, he said -- for example, by spending all the proceeds of a one-time tax cut if they understand that the windfall is only temporary.

This is an astonding assertion, and is contradicted by actual behaviour. Folks used these "one time" gains in house appreciation as ATM withdrawls, particularly as median income continued to stagnate. And people do so every spring, with that tax refund they get. Never mind that 99.44% of those getting a refund have no clue that they've lent the money to Uncle Sugar at 0% interest. Come on!!
Prevailing economic models do not adequately explain the financial crisis, the severe recession or the weak global recovery, he said. "Systemic risk" is a buzzword for politicians and financial regulators, he said, but "the truth is, we really don't know how to measure it or what exactly it is."

Again, more willful ignorance. The data were out there and available to modelers. But, just as the Banksters took the notion of musical chairs ("while the music plays, we all have to keep dancing"), or Russian roulette, with the real economy, so too did a large proportion of quants who claimed to be analyzing the economy objectively. Since the behaviour didn't square with both their bias and models, i.e. house prices are always correct, they made no attempt to incorporate conflicting data. And so we got a black swan event. Except that it wasn't.

As my Momma used to say, "what would the world be like if everybody behaved like you?" Ignorance of this question remains the flaw in the "macro is just the aggregate of all the micros" approach to all of economics, both theory and quant. It ain't, just as a local water supply can't be sustained if all residents treat all the water as all theirs. The fact that this remains unspoken by the micro folks just means it will happen again. Most quants, be they econometricians or hedgies, earn their pay from entities which ignore or actively dissemble the externalities they create. If Adam Smith (the real one) were right, then there would be no unpaid externalities. No fracking way, of course.

14 November 2013

Obambi, Meet Gresham

Gresham's Law is long and widely known in econ circles, though not mentioned often enough among the punditocracy. Obambi faces it now, and once again, he's blinked (as of the time I'm typing).

In essence, in the absence of symmetric true information, bad goods drive out good goods (hard to say, and it's not just coinage). There was a reason that the scam health insurance policies, almost all aimed at uninsured individuals, weren't grandfathered into the AHA: stupid poor-ish people are easily duped into buying same. They look only at the monthly premium, not at the expected value vs. the premium. Such policies will drive those most in need of real insurance into faux insurance.

As I type, the news states only existing (and those canceled, allegedly, due to AHA) policies will be allowed to continue. New buying is not permitted. We'll see the Right Wingnuts crying: "let the market decide!!!" once again. As if health is a market good. Ayn Rand would be proud.

10 November 2013

RIP, Bob

Bob's dad died.

You might remember Bob, that yakking paperclip that was supposed to make Windows easier to use? His pop was Clifford Nass, and the creation of Bob should be enough to ignore his passing. I mean Bob???

On the other hand, according to the recitation in the Times obit, his principle occupation was at Stanford, and he had, at best, a gingerly view of human/computer interaction.

I mean, who wouldn't applaud this:
Clifford Nass, a Stanford professor whose pioneering research into how humans interact with technology found that the increasingly screen-saturated, multitasking modern world was not nurturing the ability to concentrate, analyze or feel empathy

While the obit doesn't include a cite, I expect he'd agree with the legendary studies which showed that GUI-fied applications, Word in particular, distract users from content to focus on the easy facade. Playing is always more fun than working, so it's not surprising to discover that Word users would expend more time and energy on font selection than on more mind taxing content.

Nass' research centered on the effects of multi-tasking. Again, I'm a member of the choir.
"We all bet high multitaskers were going to be stars at something," he said in an interview with the PBS program "Frontline." "We were absolutely shocked. We all lost our bets. It turns out multitaskers are terrible at every aspect of multitasking. They're terrible at ignoring irrelevant information; they're terrible at keeping information in their head nicely and neatly organized; and they're terrible at switching from one task to another."


"We worry that it may be creating people who are unable to think well and clearly."


Dr. Nass found that people who multitasked less frequently were actually better at it than those who did it frequently. He argued that heavy multitasking shortened attention spans and the ability to concentrate.

Is it any wonder that some of us question the au courant meme of Agile, for example? Read Nick Carr in memorium.

Kumite', Again

Anthony Bourdain is among the most interesting of reality show personalities. He's now on his third show, and third network: "Parts Unknown" on CNN, Sunday at 9 PM. (As with his previous shows, the title is both metaphor and allusion: he goes to places tourists seldom do, and indulges in eating animal bits that suburban Americans wouldn't, and likely don't even know about.) Somehow, I missed last week's episode, but I saw a promo for it, in which he talks about doing the episode. It's even edgier than usual. Which is more than usual for basic cable.

The episode is Tokyo, but not the touristy parts. It repeats again Sunday, the 10th, at 8 PM. You should see it. This Tokyo is disturbing.

Well, it's Tokyo and Tony visits with a sushi chef he'd known for years in New York, Yasuda. At about 20 minutes into the piece, we see that Yasuda has been practicing karate for some years, and we see a few minutes of footage from the dojo he uses in Tokyo. An earlier missive in this endeavor made metaphor of karate and kumite'; not all practitioners have the skill and discipline to fight as they practice. There's some kumite' on display, and some, but not all, of the roundhouse kicks are by the book.

We find that Yasuda does, both in kumite' and sushi.

Yasuda, not too surprisingly, is a stickler for doing sushi the right way. His knuckles remind me rather a bit of George's.

09 November 2013

The Human Pretzel Factory

Never been a big fan of Agile, which should come as no surprise, since any relationalist subscribes to the notion that one needs first decide on the nature of persistent data. There are lots of tools for schema migration these days, so don't give me any shit about that. Changing schema is no more difficult than changing code. And, lest we forget, an Organic Normal Form™ schema is the constraints of record, and thus can be used to generate client side code. Lots of folks use xml monsters to do just that.

So, I came across this piece, via Artima. There's a link to the prequel in the piece.

As he concludes, there is no silver bullet. Well, there is, it just isn't some codified process. Carpenters live by it: "measure twice, cut once". The whole agile thing asserted that it isn't possible to understand an application a priori, so don't bother and just code sling until the cows come home. Eventually, if you don't wear out your debugger in the process, you'll get something to work. It should come as no surprise that healthcare.gov got built this way. Here's the autopsy. For the record: the previous sentence's contents, I just discovered, while the one next previous I just winged it: seemed obvious given the Kiddie Koders who've been the public facing talking heads.

Measure twice, cut once. Repeat until you understand that.

05 November 2013

Frankly My Dear, I Don't Give a Damn

Let's, for the moment, assume that Obamacare works much as it was designed: provide affordable healthcare. Next, let's follow Einstein's lead and conduct a thought experiment. He devised the notion, although not the algebra, for the theory of special relativity whilst sitting on a tram looking at a receding clock tower. He imagined that, were the tram moving fast enough, the clock would look to him, on the tram, to have stopped. And the rest is history and a Nobel. The proceeds went to his estranged wife, whom he didn't like all that much. Oddly, still more: she did most of the algebra underpinning special relativity. Albert was a bit of an ingrate. Didn't believe in quantum theory, either.

What, then, can we imagine would stop if Obamacare works as designed?

Mostly, bad marriages. Very Einstein, don't you think? And why would one come to such a conclusion? It's been well documented that having access to healthcare is a powerful force, keeping otherwise disgruntled people in bad marriages and bad jobs. House ownership, too. But Obamacare doesn't deal with that. Yet anyway.

So, we should see an increase in divorces. Whether that increase will be above sampling noise, is another issue. But unintended consequences have a habit of occurring. Just ask Greenspan; he crashed interest rates without much of a clue what would ensue.

Bookmark this post, and come back in five years. If the globe hasn't exploded, of course. Never know what the Jews and Arabs might do.

04 November 2013

Pretty Boy Floyd

Floyd Norris' column this Saturday was unusual, in that there was no cite to the QWERTY Analytics (or some such), just his byline, NYT, and BLS data. Let's see.

Start with the title:
Changes in Labor Force Mask Gains in the Jobs Situation

With such a title, one would rationally expect that he's going to present data which contradicts both the Left Wing and Right Wing Nuts, who point to the continuing drop in the size of the labour force as the main reason the stated unemployment rate has fallen. The lack of progress by Obambi is something both camps agree on. They disagree as to why, of course. The Left asserts that the Tea Party House (and the Senate, but less egregiously) eviscerated the stimulus effort; they got to spike the recovery and shift the blame to Obambi. The Right disclaims any responsibility for watering down the stimulus, claiming it wouldn't work at any amount of money; the key is to get all those welfare cheats back to work at starvation wages, and then all will be well.

The record is clear: the Right did spike the stimulus. What might have happened with either a more focused stimulus, or none at all can't be known now. We do have much historical data to demonstrate that robust stimulus works. It's called World War II. It would be helpful to get the stimulus effect without all the dead bodies and razed cities, of course.

So, he continues:
But the actual employment picture may be better than those statistics would indicate. Over the last few years, the labor force has changed in important ways because of demographics.

The accompanying charts attempt to adjust for those changes. They do not show a strong recovery, but they do indicate that the overall employment situation has improved.

One might reasonably infer from this that, despite a drop in the size of the measured labour force (usually referenced as the Labour Force Participation Rate, rather the count), unemployment has really gone down, i.e. employment has really gone up. Let's see.

The makeup of the working age population has changed substantially in only a few years. When the recession began at the end of 2007, 54 percent of the people considered to be working age were in the prime working age range of 25 to 54. Now, the figure is 51 percent. The proportion over 55 went to 34 percent, from 30 percent. With more people at or over traditional retirement age, it should be no surprise that fewer are working.

While I don't have a cite immediately to hand, data since The Great Recession began have measured an *increase* of older workers working past the "average" retirement age of the years previously. In other words, to the extent that older workers make up the demographic, there is data indicating that they've not gone off to the Life of Riley Hammock, sipping gin. And if we look at the linked graphs, we see that old folks continued to work. If you examine the middle cohort, in middle dense blue, you can see that the unemployment rate and participation rate track exactly in The Great Recession period, lending credence to both the Left and Right that things ain't all that much better.

In sum: the oldest cohort has stuck to the labour force in greater numbers than in the past, from which those with rose colored glasses conclude that the unemployment situation is getting better, overall. For those not so festooned, the middle cohort has, even with Norris' "adjusted" numbers taken it the gut. The youngest cohort, worse still.

As I read the data, the oldest cohort retains due to coercion, not choice. Their positive effect on overall employment status isn't really a positive: they can't afford to retire and leave their jobs to younger folks. This is not progress. Note that the Right Wing Nuts constantly bray about forcing the old farts to work until 70 or 75; after all in 1935, 65 was the average life expectancy so we should obviously raise today's retirement age to average life expectancy. Makes sense to me.

01 November 2013

First Thing We Do Is, Kill All the Biostats

Sequenom is a diagnostic/testing company that devised a way to test for Down syndrome non-invasively. They've been in a patent fight over their tests for some time. Today's Times has some details. What's both interesting, and perhaps disturbing, about the case is exemplified in some contradictory text in the article.

On the one hand,
The new tests offered by Sequenom and several competitors take advantage of the fact that some DNA from the fetus can be found in a pregnant woman's blood and can be analyzed with new genetic techniques.

Note the "new genetic techniques". But later on
In her ruling, Judge Illston said that the presence of the DNA in the pregnant woman's blood was a novel discovery but could not be patented because it was a natural phenomenon. And claims in the patent over how to detect that DNA were not eligible for a patent because they involved conventional genetic techniques.

So, we have it both ways. At least in this write-up.

But more to the point, as quoted in the article
"It's hard to imagine patents on diagnostics surviving if that approach is taken," Christopher M. Holman, a law professor at the University of Missouri-Kansas City, said of Wednesday's ruling.

And, that I would agree. All those biostats out on the sidewalk, destitute, having to hit the dusty trail to New York to work as a financial quant. One question: how large a part of quants generally are the biostats? This piece from a few years ago says, about 1/3, if the data have the biostats counted with the general stats, 1/4 if they're really disjoint. And half of the remaining 2/3 work in the vampire squid colony in lower Manhattan.

While I'm by no means a fan of software patents, and most bio patents, in the latter a case can be made that such diagnostics/tests can be novel physical inventions, and thus meet the criteria for patenting as set down by The Founding Fathers. Unlike Myriad, in my opinion, what Sequenom has is different from attempting to patent the "natural phenomenon". One of the arguments against a patent is the notion of obviousness to a competent practitioner. Where that exists in statute, I can't find, but is repeated endlessly. Let's suppose that it is really in law. Then, in the case of medical diagnostics (and the jobs of all those poor biostats), the issue is obviousness in how the "conventional genetic techniques" (if truly "conventional") were structured. Ultimately, all new genetic techniques derive from conventional techniques, at some arm's length. Standing on the shoulders of giants, and all that. By this logic, no patent on a VOM could be awarded, since, after all, electrical current/resistance/voltage are all natural phenomena and lots of conventional meters existed. If Apple can get a patent for a rectangle with rounded corners...