Monthly Archives: August 2013

Quicklisp new packages, August 2013

warren-teitelmanBut first, I was sad to see that Warren Teitelman passed away.  I first encountered the idea that all the basic operations could record how to reverse their actions, and hence you could easily implement undo, in his Programmer’s Assistant.  I’ve wondered if he actually invented undo.

Programmer’s Assistant first in BBN Lisp and then in Interlisp made a huge impression on me.  It is curious how his invention of DWIM became something of a joke over the years.

There are videos of Warren and his dog competing on his Google+ page only weeks before his heart attack.

Now back to our regularly scheduled programming…

cl-autowrap 23K -- BSD-2-Clause
  Import c2ffi specs and generate CFFI wrappers
  author: Ryan Pavlik
  git: https://github.com/rpav/cl-autowrap.git
  more: https://github.com/rpav/cl-autowrap#readme

cl-date-time-parser 10K -- MIT License
  Parse date-time-string, and return (values universal-time fraction). Parsable date-time-format: ISO8601, 
     W3CDTF, RFC3339, RFC822, RFC2822, RFC5322, asctime, RFC850, RFC1036.
  author: Takaya OCHIAI
  git: git://github.com/tkych/cl-date-time-parser.git
  more: https://github.com/tkych/cl-date-time-parser#readme

cl-ledger 253K -- No license specified?
  Port of the Ledger accounting system to Common Lisp.
  git: git://github.com/jwiegley/cl-ledger.git
  more: https://github.com/jwiegley/cl-ledger#readme

cl-paymill 7K -- BSD, 2 clause.
  CL-PAYMILL is a common lisp interface to the Paymill
    payment service API.  See https://www.paymill.com/
  author: Peter Wood, email: pete_wood at runbox.com
  git: https://github.com/a0-prw/cl-paymill.git
  more: https://github.com/a0-prw/cl-paymill#readme

cl-string-match 106K -- BSD
  Provides implementations of the standard sub-string search (string
   matching) algorithms: brute-force, Boyer-Moore, Rabin-Karp, etc.
  mercurial: http://hg.code.sf.net/p/clstringmatch/code
  more: http://clstringmatch.sourceforge.net/

epigraph 18K -- BSD
  A library for representing and processing graphs (nodes and edges)
  author: Cyrus Harmon
  git: https://github.com/slyrus/epigraph.git
  more: https://github.com/slyrus/epigraph#readme

trivial-tco 3K -- MIT
  A Common Lisp library to assist in ensuring certain code is executed with tail
    call optimizations enabled.
  author: Ralph Möritz
  git: https://github.com/ralph-moeritz/trivial-tco.git
  more: https://github.com/ralph-moeritz/trivial-tco#readme

vgplot 9K -- GPL
  Interface to gnuplot
  author: Volker Sarodnick
  git: https://github.com/volkers/vgplot.git
  more: https://github.com/volkers/vgplot#readme

Hourly Cost of Healthcare?

When you see a gas station with cheap gas are you actually observing?   Consider for example a gas station owner who hires people to pump his gas.  His costs are lower if he can skip providing health coverage.  Examples like that are one reason I support universal provision of healthcare is the powerful way that market competition works to eliminate it.  It’s an ethical puzzle, eh?

The national minimum wage is $7.25/hour, while spend $6,815 per person on health care.  If we presume a 40 hour work week, 52 weeks a year, that’s $3.40/hour.   Or to look at it another average premium for health insurance is $215/month, e.g. $1.29 an hour.

Setting standards, for things like provision of healthcare is one way to address the problems ethical and otherwise that market’s present.   But standards come in many forms, for example labeling.  Maybe the gas stations that have uninsured employees could be required to display a large shaming sign.  Or if you into techno-utopian solutions maybe we could spin up an App for your phone that let’s you look this info up.

The Times has an article today about how the people who pick our vegetables lack health insurance and providing it might cost a dollar an hour.  Well yeah.

prices are forward looking

One of the old languages, Aymara, of South America has a unique feature.  Events in the past are said to be in front of the speaker, i.e. where he can see them; while those in the future are behind him.

If you think about it a little it makes sense.  You can’t see the future.  But, you can call up memories of the past.  So in sense that is all you can see.

But, yeah, doesn’t that treat memory as more valid your imaginary futures, your plans.  It hardly seems fair to treat one tenuous mental model as more real than the other.   It makes me wonder is Aymara the perfect language of nostalgia?

I was reminded of all that by this essay which digs into the common presumption that you deserve your income.

“Free market prices are essentially forward-looking.  Current prices send signals to producers as to where the demand is now, not where the demand was when individual producers decided on their production plans.”

I like to say that there are two simple naive ways to think about the price of something: that it cost to make v.s. what it value the buyer will extract from it.  If $cost < $value then we can negotiate.  There are uncountable reasons why this model is too simple, but it’s still useful.  In part because it’s nice and symmetric, the seller doesn’t want the buyer to know $cost and vis-a-versa.  They keep their secret, but it looms large in their mind.  During the negotiation the buyer strives to dampen the seller’s nostalgia and his own valuation; while the seller does the opposite.

If I have labored a life time to acquire some skill then I know that my $cost is high, but the market doesn’t care.  The market is just as likely to have a high $value for things that skills that were trivial to acquire.   This is a line of reasoning that I am familiar with for most products, but I’ll admit I’d never seen it so boldly stated for transactions involving skills.

The Modern Gentlemen’s club

I inherited from my father an affection for cartoons.  I still have his books of Punch, and a complete set of New Yorker cartoons.  One of the standard tropes in these collections is set in a gentlemen’s club.  Two large cigar smoking elderly gentlemen recline in leather arm chairs.  One says something to the other.  Such as: “I think I’ve acquired some wisdom over the years, but there doesn’t seem to be much demand for it.”

old_fartsMen’s clubs seem to be rare in America.  We used to have lots of them.  The suburbs and television sucked the life out of the, or so I’m told.  And the term “gentlemen’s club” now has become a euphemism.

But, for the rich, the men’s club solved a problem.  It got them out of the house.  Which is a serious problem if your a type A over achiever, since chances are you married a type A over achiever.  Imagine the trouble.  You sell your startup and decide to spend more time with the family only to discover there isn’t really room at home for two over achievers 24×7.

In the good old day’s you’d join a good club.  And added bonus: you could sit around and whinging about who the youth of today are going to hell in a hand basket.

But of course capitalism is very good at filling demand, particularly when those who have the need happen to have disposable income.  And thus we have the venture capital firm.  It’s actually better than the classic gentlemen’s club.  You can still have the comfortable digs, the high end dinning room, the subscription to all the daily rags.  But it’s better!  Instead of complaining about the young you get to invite them in and give them advise!

That’s an old joke of mine, though it’s not entirely clear if it’s a joke or a deep insight into some aspect of the venture capital firm’s value proposition for it’s partners.  Memory is an untrustworthy beast, but I’m pretty sure I came up with this joke after hearing of a VC firm in the valley that had a wall down the middle; one one side it was pure luxury and on the other it was standard spartan office space.  As I remember the story the partners would always meet the entrepreneurs on the spartan side of the wall.

And so, it was with great delight that I read an article in today’s paper.  A friend of mine, having recently exited from his last of a series of successful entrepreneurial activities appears to have taken my insight to heart.  He is setting up a startup incubator here in Boston.  And make no mistake: that’s is honorable work.  But this was the sentence that delighted me:

English says he’s planning an “outrageous” workspace that will transform into a club at 6 PM, with regular events that “celebrate creative people” like dancers, sculptors, and clothing designers.

That is definitely an improvement on my original insight.

Whinging about PGP

I find PGP very lame because the community has failed to do even the most basic product management.

There are audiences that need to be drawn in.  Who: email users, email client authors, major email users (i.e. commercial entities), even the intermediaries.  Each of these audiences needs guidance, best practices, models.  And usability sucks.

For example there is no guidance for how to signal that I support PGP and or that I would like my correspondents to use PGP if possible.   There should be a norm about how to signal it in a footnote.  There should be a standard header so counter party clients can automate the song and dance.

The state of all this has only slightly improved over the many many years PGP has existed; but the name change around GPG was the worst kind fragmentation of branding.  Pure navel gazing.

It’s all a terrible shame.

 

Claiming-race Taxation

An amusing  tax scheme, framed as a Claiming-race, via Underbelly.

The idea is that the owner must declare what he thinks the the property is worth.    The state then get’s to pick one of two options.  A) collect the tax on the declared value, or  B) buy the property.

The term claim-race comes from horse racing.  It’s apparently common.  Horses in a give race can be purchased for a price tied to that race.  Owners accept that as part of the risk they take when they enter the race.

In the horse race it helps to get horses of about equal quality into the race and relieves the race organizer of assessing quality.  In the tax scheme the state is relieved of assessing the value of the property being taxed (unless they want to exercise option B).  In the example underbelly found of the property in question was cargos passing a toll booth.

interviewing & scores as noise

MatchmakerDVD2A recent interview on hiring in the New York Times[1] included these little snippets:

“we did a study to determine whether anyone … is particularly good at hiring. We looked at tens of thousands of interviews, and everyone who had done the interviews and what they scored the candidate, and how that person ultimately performed in their job. We found zero relationship. It’s a complete random mess, except for one guy who was highly predictive because he only interviewed people for a very specialized area, where he happened to be the world’s leading expert.

G.P.A.’s are worthless as a criteria for hiring, and test scores are worthless — no correlation at all except for brand-new college grads, where there’s a slight correlation …”

While these are consistent with my experience, it’s a reasonably bleak conclusion.  First off just imagine you’d spent the last few years expediting a huge hiring pipeline, twisting arms to get interviewers, forcing them to fill out score sheets, etc. etc.  And now it all turns out to have been totally useless.

Or what if you want to make the decision to commit to a relationship.  It’s presumably like other shopping exercises: You collect information.  You then distill into “signals.”  Those signals help trigger the decision to commit.   Apparently all that interviewing and dating is useless.

What then?  Seek other signals presumably.  Height?, rosy complexion?, ficco score?  The article does mention: “what works well are structured behavioral interviews.”   I wonder if that has a consensus meaning.

It’s fun to realize that they probably have the data which would allow them to realize some unfortunate hiring patterns.  For example they might discover that candidates interviewed in building 3 just after afternoon tea on a Tuesday early in the month by a senior employee are much more likely to be hired.  Remember that one of the five ways to get rich is to marry well.  Candidates would pay for that information.

Of course it’s, no surprise that the match making (commitment) problem is so hard.  Evolution you know.

Luis writes:

Open source and open standards are similar because both are generally created by communities of collaborators who must trust each other across organizational boundaries. This is relevant to the right to fork, because the “right to fork” is an important mechanism to help those communities trust each other. This is surprising – the right to fork is often viewed as an anti-social right, which allows someone to stomp their feet and walk away. However, it is also pro-social: because it makes it impossible for any one person or organization to become a single point of failure. This is particularly important where the community is in large part a volunteer one, and where a single point of failure is widely perceived to have actually failed in the past.

Which is a nice way to frame things.  I particularly like the phrase “trust across organizational boundaries.”  It’s a phrase that get used in interesting contexts.  My fellow open source enthusiasts tend to thing that the trust problem is between individuals, but in the end individuals are just tiny organizations.

The phrase “single point of failure” is an interesting way to frame the larger problem of contingency plans.  Yeah!   Right to fork is a kind of prenup.