MyxerTones Evolution

The team at MyxerTones has been working for a long time on the new release of our website, and we just went live this (last?) evening. I couldn’t be more happy with the process that got us here, and the new features we have in place.

I can vividly remember sitting at the bar of the Wynkoop Brewing Company in Denver with Marsha and Bill, talking about the potential for the company that would become mVisible. We talked about the many confusing and twisted ironies of the internet: borne by the US defense department’s desire to be able to retaliate effectively in the event of nuclear holocaust; yet inherenty decentralized by design; extremely simple in concept; amazingly complicated in practice. There’s been a constant struggle since the first DARPA packets were exchanged, between a centralized command-and-control architecture, and a decentralized, self-correcting, and redundant communitcation system.

At Wynkook, we talked of “taking back the internet,” actually doing something about the fact that something that was designed (by the engineers, anyway) to be completely decentralized was now almost completly controlled by a very few powerful players. PCs too closely shared a bloodline with the DARPA servers of yesteryear to really break free, but mobile phones, now, mobile phones were something different altogether.

We sat at the bar (mmm, Wynkoop had the best menu of appetizers you’ve ever seen, really, I think I could’ve ordered from the appetizer menu alone every day for a week), and talked about using the mobile phone to bring the inherent power of a decentralized, worldwide network that is the internet to individuals. What if there were no ISPs? What if there were no mobile “operators?” What if data flowed from person to person without having to go through any central switching point? Think of the power of this medium if people could freely communicate and exchange information with whomever they wanted!

They were sold. I was sold. It was a magical moment, and I’ll never forget it.

It seems rather melodramatic to talk of lofty topics like this when the company we eventually went on to found is currently…selling ringtones. But the fact of the matter is, we’re taking one step at a time towards our goal of empowering individuals to communicate and exchange information using their mobile phone as they see fit, not as fits the agenda of some global conglomerate. It’s all about the individual, and the amazing potential of the mobile phone. We’re on our way.

WikiPhoneDB

I’ve been thinking a lot about wikis, open source software, and how it relates to for-profit businesses. Specifically, I’ve been occasionally speaking with this little voice in my head that has been whispering that there is a real opportunity for my company to make a difference that empowers the individual in the mobile phone industry.

MyxerTones.com (to hear our home page tell it) is the best way to make and share mobile phone ringtones and wallpapers. What we’ve done is set up a site that allows anyone to make ringtones from audio files, or wallpapers from images, and then send them to their mobile phone. Our most enthusiastic customers have been independent bands – some really small, local bands, some very large acts without major label affiliation – because MyxerTones allows them to connect with their fans in a way that isn’t otherwise possible in the current mobile ecosystem. See, the openness with which MyxerTones allows producers (artists) to get their stuff to consumers (normal people) is in sharp contrast to the way mobile operators have traditionally dealt with mobile content. The operators have used their control over the airwaves to tightly control the catalog of available content for mobile phones.

One of the important components of MyxerTones is our “device database.” This is a SQL database that we use to track information about the thousands of different mobile phones that are in use throughout the world. See, it turns out that each one of these phones is a little bit different. Some have bigger screens than others, some can play MP3 files, some need to have content (ringtones) wrapped in special descriptor files before being delivered, some have limits on the file size for ringtones, all sorts of craziness. I used to think it was a headache to build a website that would work properly with IE, Netscape, etc., but that’s nothing compared to the problem of keeping up with the capabilities of mobile phones.

So we at MyxerTones built this database so that we could figure out how to appropriately deploy content to the myriad mobile devices that request it from our mobile web site. When a request comes in for a ringtone, for example, our web site consults the device database to figure out what output format the requesting phone should be given.  There are a lot of weird cases, so there is a lot of special-case code needed, but ultimately the repository of information in the device database it what makes it possible for us to deliver our service.

The device DB was built by going though manufacturer’s spec sheets, working with MyxerTones customers, combing web forums, etc. And so at this point, having served hundreds of thousands of phones around the world, we have a pretty good idea of what each of these 2,000-odd specific device models require in terms of ringtone format, wallpaper size and format, full-track audio capabilities, etc. But “pretty good” means there’s a lot of holes, and it’s a pain in the ass to manage. New phones come out all the time, some of the information from our old phones turns out to be incorrect, it’s a huge effort to keep up with all of these changes.

We’re a relatively small company. We want to focus what we consider our unique creative energy and traditional product development skills on building products and technologies that help enhance the value people get out of their mobile phones. Having an accurate and dependable device database is a requirement for us to build the products we want to build, but it doesn’t represent very much intrinsic value for our company. It is, after all, just a database.
It has occurred to me that the device database is a perfect candidate for a hybrid open source / wiki project.

My idea is that we take our live device database, running in our production facility, and we expose the administrative interface in Wiki form to the world. So that instead of having one of my guys inside the company being required to hunt down accurate information for every obscure phone that we couldn’t possibly afford to have in-house, we just let the people out there in the real world that have the phone supply the necessary information, whether explicitly or implicitly by saying “yes, that format worked for me”, create entries for new phones that we don’t recognize, and generally use the hive mind to make it the most accurate device database possible.

It will be more than just a Wiki, because MyxerTones.com would eventually offer programmatic interfaces that allow any other company – even competitors of ours – to use the information in the device database however they see fit. This will provide motivation for other parties to help insure the accuracy of the information. I want this to be commodotized. The mobile industry is relying way too much on closed business practices and information hiding to really make it succeed. I don’t want companies to be able to make their money by controlling access to information. I want to force companies to make money by providing valuable and innovative content and services to end users.  And until this basic information is accessible to all, we won’t get the kind of innovation that the mobile phone deserves.

At this point, I have to mention the WURFL project, because that’s an open source style project that is somewhat along these lines. These guys have an XML data file that they update with information about new devices as they become available. Early on at MyxerTones, we used the WURFL XML file as the basis of our device database, but had to move away from it to a traditional DB model because of a variety of architectural reasons.

So, I’m not sure how the logistics of opening up our device database to the world would actually work, but I’m thinking about it…

wikicompany

I edited my first Wikipedia entry today. The article describes scale-free networks (and boy is it a zinger!). It was a strangely intoxicating and simultaneously anticlimactic endeavor. Intoxicating, because about two seconds after clicking the “edit” button next to the heading of the section I want to edit I’m staring at an editor screen. An editor initially populated with text that had been written by, for all I know, Claude Shannon himself. And I’m not exactly what you would call an expert in graph theory (I think of myself as more of a fan of that particular branch of mathematics) And here I am, with that text in my editor, and I can turn it into anything I want – right now – without asking anyone for their permission! That’s intoxicating!

And it was anticlimactic for exactly the same reason: it was so dang easy. I didn’t even have to sign in! How the heck can this possibly work? There must be something wrong here…

Lots of people have been talking about wikis lately, and it’s obviously tied quite closely to discussions about open source software. Wikis and OSS both strive to take the collective input of a large number of people to build something more valuable than what can be built in a tops-down fashion. It’s feed vs. seed in the parlance of The Diamond Age.

To be successful, most open source projects and wikis require a good deal of guidance in the formative stages. This guidance needs to create an environment conductive to eliciting and absorbing the contributions of myriad user who have something valuable to offer, and it needs to launch the project on its initial trajectory, whether that’s by putting together the first version of the kernel code (Linux) or writing the initial articles on Aardvarks and Zygotes (Wikipedia).

Traditional software engineers would probably think of this first stage of a project as building a scaffolding; a framework that encompasses the entire area that is to be filled in later with more specific functionality. But this is a rather limiting and tops-down approach; it’s great for projects with specific functional and scheduling requirements, but it is not at all an environment that is likely to gain emergent value from the masses of people on the network.

If one wants to harness the power of the internet, more correctly the power of the people who use the internet, it seems the most important thing to do is create a hospitable environment for contribution. If we slaughter the epidemiology analogy, what we are trying to do is stock a petri dish with a sustaining auger, and place a few small but established colonies of ideas or structured data. If we get the environment and the initial trajectory correct, we can harness the power of the masses to build upon our first efforts.

Oh, by the way, my contribution to scale-free networks was that the quotes around the definition for “preferential attachment” were in the wrong place.  I fixed them.  How would the world have survived without me?

a lot to say

Sometimes, I have so much that I want to say that I don’t say anything at all.

tipping

The airport newsstand is not where I generally find my reading materials. And the “business books” section of said airport newsstand is one of the specific places you’re unlikely to find me loitering. I rather pride myself on an aversion to the cheesy books one tends to encounter there, and so my first encounter with The Tipping Point by Malcolm Gladwell comes some six years after the rest of the world was exposed to it. I’m intentionally recording my thoughts on the book before doing any internet research on what, if any, consensus has been arrived at by the hive mind.

Mr. Gladwell is best known, of course, as the love-child of Martin Short and Buckwheat:

+ =

…though some people may be surprised to know that he’s had an article or two published in an obscure rag out of New York with which he’s affiliated.

I’m not going to rehash the main points of the book. I’m sure there are a thousand better places to go for a quick overview. But a summary sufficient for my purposes goes thus: the Tipping Point explores the emergence and spread of societal trends (fashion, habits, crime, etc.) by means of a running comparison with disease epidemiology, pointing out unexpected non-linearities and providing speculations as to hidden causes for the explosive spread (or, conversely, failure to launch) of these trends. Though the book is ostensibly aimed at business readers (according to its marketing collateral and genre filing in the newsstand), its most compelling proposals seem to be the novel approaches to problems such as crime, smoking, and disease eradication suggested by example.

First, a coincidence, connection, link, call it what you will. The opening example given in the book is that of the resurgence of Hush Puppies as a stylish footwear choice, in the mid- ’90s, owing to its kitchy-ness being mined by trend setters in New York. I happen to have grown up, K-12, with the son of the CEO of Wolverine World Wide (the company that owns Hush Puppies). We did an incredible number of shockingly stupid things together. Anyway. What’s interesting is that my schoolmate’s name just came up in a conversation I had with my sister the day before yesterday, and that’s the first time I’d thought of him in at least a couple of years. (The context of that discussion is not entirely relevant here. Let’s just say that he and I would provide a rather dramatic counterpoint to Gladwell’s insinuation that people growing up in the same social environment will tend to have similar values and lifestyle choices. Exhibit A: he became a close adviser to Vice President Dick Cheney. I, the record will clearly show, did not.)

The reason this connection between myself and the Hush Puppy, and therefore between myself and this very book, strikes me as so interesting is because I am – according to a test Gladwell presents in his book – not a very connected guy. Like, “please hang up and try your call again” not connected. Briefly, the test consists of a list of 250 surnames, and you’re supposed to go through the list and give yourself a point every time you “know” someone with that last name. Tally up your points, that’s your ‘connectedness’ score. The average score on the test is 41. The scores for those ’super nodes’ in the global social network, people deemed to be those all important “connectors” that figure into Gladwell’s scheme of things, are over 100. I scored 11.

While the talk about the necessity of Connectors, Mavens, and Salesman to make something really ‘take off’ is interesting, and the examples illuminating, the basic concepts seemed rather familiar to me, and therefore didn’t come across as revolutionary thinking. Perhaps my reaction would have been different if I had read the book in 2000 when it originally came out, but popular digests of complexity theory have given even a layman like me a healthy exposure to explanations of emergent behavior in complex systems that exhibit non-linear dynamics. Come to think of it, didn’t a different Malcolm teach us about that in Jurassic Park?

On the other hand, the examples describing environmentally-effected behavioral modification are truly fascinating. The idea that traits we normally associate with (assign to?) people are not constant, but are more accurately described as dominant behavioral patterns exhibited by the person in the environment in which we normally observe them never really occurred to me before. I mean, I’d been exposed to people who espoused, for example, making children wear uniforms to school to make them behave better, but I wasn’t aware of the various ways by which sociologists had studied the phenomenon. And I thought Rudy was just being a hard-ass by cracking down on panhandlers in NYC; I had never seen it mentioned that this was part of a larger effort to basically change the environment such that it would be less likely to incite criminal behavior.

I guess the trick is to understand how the environment can be manipuated to make people behave how you want them to behave. Eeck! Did I just say that?

numberometries

I haven’t ever become obsessed with making sense of the distribution of primes. I do, however, spend about 0.003% of my idle time advancing this half-baked hypothesis that the prime numbers represent the intersection of our world’s concept of natural numbers with the True Numbering System of the Universe. Like, we think of our numbers as this pure straight line through mindspace, with an equal distance between adjacent numbers, but maybe we only perceive one dimension of what is really a gnarled twisting line weeding its way through a space of much higher dimension. So maybe the prime numbers represent regularly, deterministically occurring points (or planes, etc.) distributed in the True Numbering System of 14 dimensions or whatever, and only seem random when seen plotted on our jagged little ‘natural number’ line.

Maybe like mass warping space-time giving rise to what we perceive as gravity, the multidimensional natural number line (of which we only perceive one dimension) is in a kind of “numeric freefall” that gives rise to oddities like the importance of pi and e and the distribution of primes, the cause of which would be obvious if we were to perceive the other dimensions.

And maybe there are experiments one can do to determine if that is indeed the case analogous to how the ant on the surface of the sphere can find out his 2d walking surface is actually curved along another dimension. The ant can carefully construct a really big circle by marching out an equal radius distance in all directions from a center point, and then measure the circumference. c = 2pi*r, all’s flat, otherwise something’s going on. Perhaps there are equivalent experiments for pure numbers. Maybe prime numbers are the key to those experiments.

So just as there are an infinite number of geometries in which the value for (e.g.) pi changes – like the 2d surface of a sphere – maybe there are infinite ‘numberometries’ in which the distribution of primes change. We might be able to use the primes as a tool to untangle this rat’s nest of a number line we mistakenly think of as being pure and straight, and learn the shape of the underlying numberometry of the universe.

Finally, I think we as a society should be spending far more time being blown away by the fact that one can explain some quantum electrodynamics phenomenon only when you admit that there are subatomic particles popping in and out of existance while travelling backwards in time. How cool is that?

beginning

“What are you doing now?” he asked.

“Do you keep a journal?”

So I make my first entry today.

So starts Henry David Thoreau’s journal, which (I hear) went on for approximately 20 volumes before that infamous quarrel with God (of which Thoreau was apparently hopelessly unaware) ended his writing career some years later. I hesitate to invoke the words of such an oft-quoted hero of writers here, being that (1) I am exactly 9 pages into my first reading of Civil Disobedience, and therefore have had insufficient time to elevate the author to heroic status, and (2) I am some 100 words into my first blog entry, and therefore have had insufficient time to establish myself as being, indeed, a writer.

When faced with such daunting tasks as this that lay before me now – the goal of writing my own 20 volumes, perhaps puncuated now and again with a particularly witty essay that might stand on its own and serve some purpose for a Good great or small, hopefully stretching my words across a great many more years that were afforded Thoreau – I often speak silently to myself the words of Confucius (another great thinker of which I am most completely ignorant):

“A journey of a thousand miles begins with a single step.”

So I take my first step today.