Don’t want to lose this:
You can disable ldb … at runtime …:
(funcall (sb-alien:define-alien-routine
(“disable_lossage_handler” cl-user::disable-sbcl-ldb)
sb-alien:void))
Don’t want to lose this:
You can disable ldb … at runtime …:
(funcall (sb-alien:define-alien-routine
(“disable_lossage_handler” cl-user::disable-sbcl-ldb)
sb-alien:void))
While doing a bit of work helping The Echo Nest get their developer network rolling I got to observe an amazing outrageously cool example of what can happen when you open up your technology.
This bends one of my blogging rules: no blogging about the job. This time it’s a consulting client. But the gig is all done and I’m not revealing anything proprietary.
I treasure examples of why relinquishing control of your technology is a good move. Because bewilderment is often the first reaction when I suggest it. And then, most technology owners don’t seem to like the explanation, which seems straight forward to me.
My favorite answer for why this can work: searching for cool applications demands skills and attitudes that the firm lacks. These are on the demand side. They are close to the problem the user needs to solve. I love this answer because it’s symmetric – scarcity on both sides. The firm should not horde it’s options because the knowledge to act on those options is scare.
You can frame this answer as a search problem. Searching the option space created by the new technology requires all the usual stuff: capital, talent, knowledge, an appetite for risk, and intimacy with a high value problem. Delegating the search problem to the third parties works well because they bring increased knowledge, because they understand the problem being solved. The firm only understands the technology being applied. The developer in your developer network brings a heightened appetite to solve the problem, because it’s their problem. It is perversely fun to note that the 3rd party will take risks the firm would never take; they might be small, foolish, impulsive, or very large and self-insured.
This isn’t the only workable model for a developer network (there are, just to mention three: commoditizing, standardizing, and lead generating models).
But if this is the model your using you can begin to set expectations. A successful developer network must create surprises. If the search created by the developer network does not turn up some surprising applications of your technology it’s probably not working yet.
When it works the open invitation to use your technology creates a stream of surprises. Expect to be bewildered. Curiously the somewhat bewildering decision to relinquish control, if successful, leads to yet more bewilderment. But, surprise comes in many flavors. You may be envious because the third party discovers some extremely profitable application of your tech, as Microsoft was when the spreadsheet and word process emerged in their developer network. You maybe offended, as some of us in Apache were when violent or pornographic web sites emerged in the user base. You maybe disappointed, as I was when the market research showed that most spreadsheets had no calculations in them. You are often delighted as I suspect the iPhone folks were when this somebody invented this wind instrument based on blowing on phone’s microphone.
Dealing with the innovations created in the developer network can be quite distracting. It’s in the nature, since the best of them take place outside the core skills of the firm. That means that comprehending what they imply is hard. Because of that I seem to have developed a reflect that treasured these WTF moments.
So, One aspect of managing a developer network is digesting the surprises. To over simplify there are two things the developers bring to your network: a willingness to take risks, and domain expertise. The first means that you often think, golly that’s seems rash, foolhardy, and irresponsible.
Consider an example. It is very common to observe a developer building a truly horrible contraption. They use bad tools, in stupid ways, even dangerous, ways. And just as your thinking “oh dear” they get a big grin on contented face. If that happens inside an engineering team you’d likely take the guy aside to discuss the importance of craftsmanship. Or, if you’re a bit wiser, you might move him into sales engineering. That kind of behavior is not bewildering; it’s a sign of somebody solving a problem, creating value. Value today, not tomorrow. It’s a sign of an intense need. Now intense need is not enough to signal a high value product opportunity, for that you also want the need to be wide spread. Once you get over yourself, and learn to appreciate the foolhardy, you can start to see that it is actually a good sign.
But developers don’t just bring a willingness to take risks. They can also bring scarce knowledge that you don’t have. I love these because it’s like meeting somebody at a party who’s an expert in some esoteric art you know nothing about. It’s a trip to a foreign country. It’s the best kind of customer contact – they aren’t telling you about their problems they are revealing intimate information about how to deal with those problems. Like travel to a foreign land it is, again, bewildering.
We caught one of these last fall at The Echo Nest. I love it because it is so entirely off in left field.
The folks at The Echo Nest have pulled together a bundle of technology that knows a lot about the world of music. They have given open access to portion of that technology in the form of a set of web APIs. So they have a developer network. They have breadth of music knowledge because their tools read everything on the web that people are saying about the world of music. They have in depth understanding by virtue of software that listens to music and extracts rich descriptive features about individual pieces. It is all cool.
The surprise?
Last fall Phipip Maynin (a Libertarian, a theoretical finance guy, a hedge fund manager, and one of the developers in The Echo Nest developer network) figured out how to use the APIs to guide stock market investments. How bewildering it that! He started with a time series – hit songs – and ran the music analysis software on that series. He then gleaned out correlations between the features of those songs and market behavior. He reports that work in this paper: Music and the Market: Song and Stock Volatility.
It is a perfect example of how the developers in your network bring unique talents to the party. I doubt that anybody at The Echo Nest would have thought of it.
I often get asked were the money is in giving away your technology in some semi-open system. The question presumes that hording the options the technology creates is the safest way to milk the value out them. If you start from that presumption it’s a long march to see that other approaches might generate value. What I love about this example is how it is a delightful value counter point to the greed implicit in that hording instinct. What’s a more pure value generator than market trading scheme?
Of course now I’m curious: anybody got any examples of trading schemes based on the iPhones, Facebook, or Romba platforms?
These are some three of a kind examples accumulated over the years. Much thanks to my various correspondents. I ought to sort these out a bit.
Many are the top three of power-laws; i.e. Hertz, Avis, Budget. Once you’ve noticed that you can use any sharp power-law to generate three. For example the three top words in english: spoken: The, You, I; written: the, of, and; adjectives: other, good, new. And many of the geographic ones are like that, just forced onto the landscape: England, Scotland, and Wales.
Many of just regions along some natural scale: federal, state, city for example. There are lots that are on linear or cyclic time.
There are a number that are triangles; and then you can create a plane, or balance your three legged stool. Ordering the triangle – you have three objects and you link them into a circle with arrows rather than mere lines. Then you can play rock paper scissors; or polish your mirror.
There are number that are pairs with a middle: buyer, seller, middleman; man, woman, relationship; in, out, door; etc. In this context I find the triple: reflective, transparent, opaque thought provoking.
Apparently in some languages there are three words: one for a thing near me, a second for a thing near you, and finally a word for a thing distant from both of us. But foreign languages are not my thing. In Japanese? – koko/soko/asoko
I’m particularly amused by this group:
Well, here goes:
About once a year I dabble with trying to setup a video streaming server so I can watch movies on my Mac which doesn’t have a DVD reader on it. Usually this all falls apart because the debug loop is the scale of a DVD, the free tools are all full of personality, and the world of video encoding is quite confusing. This time I think I got it. So here some tricks I picked up along the way.
Darwin Streaming Server isn’t hard to set up. I have it serving up files who’s extension is m4v, which I assume stands for mp4 video.
I use HandBrake on the Mac to rip the DVDs. In fact I use HandBrakeCLI to rip them. A command like: HandBrakeCLI --preset Universal --input /dev/disk3 --output foo.m4v rip and convert DVD into a single file that is 1 to 2.5 Gig in size.
While that foo.m4v file can be served by Darwin Streaming Server your better off if you add so called hints to the file. These are apparently nonstandard mark up that Apple invented to make it possible to seek to arbitrary points in your movie. While there are signs that “QT Sync” can do hinting I ended up using mp4creator to do it. In addition to hinting you can optimize, so while I have no idea what that does, I do that as well. Hinting is a pain because you need to do one pass to hint the video track and another pass to hint the audio track. So you end up doing something like this: mp4creator --hint=1 "foo.m4v" ; mp4creator --hint=2 "foo.m4v" ; mp4creator -optimize "foo.m4v"
In that example the numbers 1 and 2 are denote the track number to hint. Those are reasonably dependable; but to be save you can use mp4creator -list foo.mp4v to be sure. I wrote a bit of perl to glean out the right numbers and gin up the commands to do the hinting.
Usually when you stick a DVD into a mac the operating system launches the DVD player, but you can change that behavior in the CDs and DVDs Preference Panel. Using that I launch an apple script. Since I’m more fluent in shell scripting I have that turn around an launch a shell script. That script devolved from other examples I found around the net. The essoteric bit-o-clever in it are useful; so it appears below. It has an unfortunate flaw that the heuristic for guessing which disks are DVDs sometimes gets the wrong answer; but it’s good enough for now.
My shell script then: 1) runs HandBrakeCLI, 2) ejects the DVD, 3) hints, optimizes, and 4) moves the result into appropriate directory for the Darwin Streaming Server. A little cgi script then enumerates what’s available to machines on my household’s local network.
global dvdPath
try
tell application “System Events” to set DVDs to name of disks whose capacity is less than 8.589934592E+9
set dvdPath to quoted form of (POSIX path of item 1 of DVDs)
do shell script “/Users/bhyde/bin/snarf_dvd.sh “ & dvdPath
end try
“The Barter Theater first opened its doors in June 10, 1933, providing relief and diversions for Depression-era audiences. It was founded by Robert Porterfield, a young actor who suggested that audiences barter homegrown produce for admission. Its motto was “With vegetables you cannot sell, you can buy a good laugh.” Crowds were receptive to the idea of ‘ham or Hamlet,” and an estimated 80 percent of the audiences paid with fruits, vegetables, and livestock, or dairy products.” [1]
I’m surprised I’d not recognized that bubbles of the business cycle has numerious analogies to a mast year [1,2]. Which has got me to wondering if there are cases where an industry might strive to synchronize the equivalent of a mast year so as to disrupt the populations adjacent to them in the supply chain. No doubt it happens by mistake, but intentionally?
Here’s an interesting cognitive science puzzle.
Seth is seeking a cognitive task which he can use to test how his brain is working. He want’s something quick, something he can self administer. But this is the part I find particularly thought provoking, he want’s something that the brain won’t learn how to get better at.
This is like one of those cryptographic puzzles. He needs a puzzle generating black box. The class of puzzles mustn’t be too hard and they need to have difficulty metric delivered with htem. But, no matter how many of these puzzles the brain solves, it should never manage to learn anything about the class of puzzles the black box spits out.
I wonder if there is something deep there?
They can only sell fruits and vegetables at Boston’s Haymarket and there is a guy who wanders around and fines the occasional vendor who tries to put on something else. I’ve seen people try to sell batteries, olive oil, cleaning products. Around the corner from
Haymarket are the remains of the old fish market and one maybe two vendors occasionally show up there; selling only fish. The stores the line the sidewalk adjacent to Haymarket are mostly meat markets. They aren’t as tightly regulated and they put on a larger variety of goods. No doubt you could write a whole book about history of the negotiation between all these vendors and their landlords, e.g. the market owners.
The homogeneity of the vegetable market is an interesting contrast to the shopping mall where the landlord, it seems, strives to maximize the heterogeneous diversity of the tenants. No doubt the tenants prefer the absence of competition and I assume their leases spell that out explicitly.
I don’t doubt that the vegetable vendors prefer that Haymarket’s regulator stamps out the occasional attempt to change the rules of the game. The homogeneity creates externalities that the can be observed at larger scale in other geographic concentrations.
Which is why the stifling nature of zoning laws tends to be pretty uncontroversial. No doubt the folks who live in NYC’s fabric district are peeved when some innovator decides to open a clothing store.
It is a mystery to me why there are so few commercial malls with a homogenous rather than a heterogeneous collection of stores. Of course there are examples. Tourist trap malls that sell nothing but trashy collectibles. Antique furniture malls, that seem take over failed malls. And of course mall landlords to work to achieve some kind of homogeneity; filling their space with vendors whom will all appeal to the same demographic. So maybe some of the mystery can be resolved by saying that the homogeneity is on either the buyer side, or the seller side. Maybe you can’t get high producer homogeneity until you achieve a sufficient population density.
Nice example of how tit-for-tat is often exactly the wrong approach.
It’s a curious saying, “Render unto Caesar.” Advice to the oppressed, give it up, accept what you can not change. Don’t organize a resistance. Don’t fight back. Cave. No matter that he’s evil, illegitimate, foreign, and couldn’t careless about your culture. Of course in context the entire quote is about surrender “…and unto God the things that are God’s” so it is really pretty consistently in the give it up and submit school of philosophy.
But there are times when it’s probably good advise: fighting city hall, gravity, death and taxes, etc.
I’ve spent a lot of calories over the years declining to submit to Microsoft’s monopoly. Even had some success in doing so. But rolling the clock back I’m comfortable that I didn’t expend too much effort evangelizing that other people decline to submit to that Caesar.
That’s all water under the bridge, more or less, but this issue arises again and again. A dominate commercial monopoly exercises their power in offensive evil ways and the choice arises, submit or sign-up for the extremely abrasive alternative of fighting or shunning the network?
What brings this all to mind? Today’s example – Facebook v.s. Google