The still-curious case of Bitcoin

Bitcoin appears to be on another run, though it's hard to tell because it's also become considerably more volatile over the past couple of days.  Volume overall is thinner than it was in the last major run-up, which means even more likelihood of funny business in the price -- by which I mean behavior you wouldn't expect to see in something like a Dow or S&P 500 stock, not necessarily outright manipulation, though that's certainly not out of the question.

Clearly Bitcoin is something of value to at least some people, but what kind of something?

First, let's dispense with what Bitcoin is often claimed to be: an alternate reserve currency that could come to supplant "fiat" currencies like the dollar, euro, pound or yen.  Not only is its price hair-raisingly  volatile and its money supply tiny, so far as I can tell no one actually prices in Bitcoin.  People who sell things for Bitcoin figure out a price in a reserve currency and then convert to BTC at the going rate.  For example bitcoinstore.com prominently displays its dollar-to-bitcoin exchange rate, generally substantially below MtGox's quoted price, and shows the dollar cost of every item below the BTC cost.  The dollar prices tend to be stable.  The BTC prices, therefore, not so much.


By the way, what is the Bitcoin money supply?  According to bitcoincharts.com, there are currently about 12M Bitcoin in existence, which currently equates to around $2B.  That's a lot of dosh to you or me, but far smaller than the smallest measure of the dollar supply.  There are currently about 1.2 trillion dollars in circulation, and several times that not in circulation, stowed in checking and savings accounts and such.

It really only makes sense to measure the supply of a currency in that currency, so the BTC supply is currently about 12M BTC.  But which USD supply would that equate to?  I can buy stuff with money in my checking account without ever pulling out paper dollars, and conversely most Bitcoins in existence are not in active circulation, even if there's no formal notion of a Bitcoin money market account.  So probably we want to compare with the M3 money supply, basically anything that's in some account somewhere redeemable for dollars.  M3 is around $10T, or about 5000 times the supply of BTC at the current price.

It's actually pretty remarkable that something someone wove out of pure crypto could have that much value -- 0.02% of a major economy, but it's not really meaningful to multiply the current price by the number of Bitcoin in existence.  If, for example, someone tried to convert all 12M BTC to dollars in short order, that is, sell them, it's highly unlikely that the dollar price of Bitcoin would stay at its current levels.  On its most active days, MtGox was handling about 1% of that.  Flood the market with a hundred times peak volume and it's not gonna be pretty.  So who knows what those 12M Bitcoin really equate to.

The same issue applies, in theory, to reserve currencies.  What would it mean to try to convert the entire US money supply to Euros?  Nothing good, certainly.  But there's a difference.  Reserve currencies are regularly converted, directly, to real goods and services.  We can therefore compare, say, the price of an hour of labor in the US to an hour of the equivalent labor in the Eurozone and get at least a rough cross-check on the relative value of the two money supplies.


But enough amateur economics.  What is Bitcoin?

There would seem to be two major possibilities:

• An anonymous online payment system.  If I want to buy something online anonymously, I can convert money to Bitcoin and send it to the seller, who then converts it back to money.  Though each Bitcoin transaction, by design, is public knowledge, only the exchanges know which real money accounts are involved, so as long as they're not paying attention, or hacked, or subpoenaed or such, there is no way for the seller to know who I am (There are designs for layering real anonymity on top of Bitcoin, to the extent that it would not be possible to tell which particular person participating in the scheme owned a particular Bitcoin in the scheme, but as far as I'm aware no one's actually doing this).
• A speculative vehicle.  It's possible to tell unambiguously who made what Bitcoin transaction, and therefore who owns how much Bitcoin, with the understanding that "who" means "the holder(s) of which private key" and that it's at least theoretically possible for someone or someones with enough computing power to hijack the whole system.  Leaving that aside and thus assuming there's clear ownership, Bitcoin can be traded just like baseball cards, sparkly pieces of rock or mortgage securities.

Ideally an anonymous payment system would maintain a stable exchange rate between the tokens of payment and actual money, but as long as the exchange mechanism is fairly liquid, and the total amount given up in the money - token - money loop is not too much, people will continue to use it, and it appears that people still do.

On the other hand, a speculative vehicle doesn't even require that much.  It only requires that something can be bought and sold and at least some people want to buy and sell it.

That would make sheer speculation the simplest answer.  One crucial question is, how much of the volume in Bitcoin trading is from Bitcoin as a payment system, as opposed to trading Bitcoin for its own sake?

In the last couple of months, daily volume on MtGox has varied from around 5,000 on slow days to twenty times that during major selloffs.  Trading resulting from conversions for payments ought to be fairly steady over the short term, or in other words, it's probably not more than the minimum daily volume, and it's probably less.

That leaves quite a bit of room for speculation.

Welcome back, APOD!

A couple of days ago I went to NASA's Astronomy Picture of the Day (http://apod.nasa.gov/apod/) to catch up on goings-on in the sky.  The site wouldn't load.  Hmm ... is my network connection OK?  And then, I realized that I was trying to access NASA's site, at a pod.nasa.gov, and it was down because, well, the government was down.

Because I don't live under a rock, or at least not all the time, I already knew the federal government was (partially) shut down, and that even with only a partial shutdown, many people were experiencing effects a good deal worse than not being able to load a favorite web site.  What I'd forgotten -- despite the .gov in the URL -- was that APOD was run by a government agency.

That's probably partly because we don't notice URLs so much these days, which is a theme worth revisiting here one of these days, but it's also because a government web site just isn't that much different from any other.  My city's web site, and my school district's, the National Hurricane Center's, and APOD are just web sites, just as government officials from the federal level on down have social media accounts, home pages and so forth just like other people.

I don't think there's any profound lesson to be drawn there, just one of those things you notice every once in a while when the occasion arises.

Lessons or not, it's nice to see APOD again.


[Note: Because APOD is only hosted by NASA, the same content was available during the shutdown from various mirror sites]

Field notes off the web

In St. Louis, Henry Goldkamp spends his weekends on the streets writing poetry for passersby, using a Smith-Corona manual typewriter.

I find this really cool, if only for the very concept, and so did St. Louis's Riverfront Times.

Goldkamp has gone one step further, though, and installed about 40 typewriters, with paper, at various places around the city, inviting St. Louisans to answer the question "What the hell is St. Louis thinking?"

Now, I have to admit, something in the back of my tech-saturated head was thinking "Hey, that would make a cool project, put a bunch of keyboards around the city, pipe the results back to a server somewhere and show them on a web page in real time ..."

Um, no.  We already have that, more or less.  It's called Twitter.

What Goldkamp is doing is getting people to interact with technology that, to many of them, comes from another age, almost as though from another planet.  "How do I work this thing?" seems to be a common response.  Having learned to type on some combination of IBM Selectric, Smith-Corona and my grandfather's manual Underwood (I loved that thing), I have to chuckle a bit, but by the same token I can understand why it might be daunting at first.

Besides having to fuss with paper and the carriage return, and mash on the keys to get the typebars to move, the most distinctive feature of a typewriter as opposed to a computer keyboard is that you can't delete anything.  There's a backspace key, but that just moves the paper one space to the right.  About the best you can do is type Xs over what you already typed (on the other hand, you can have fun combining letters to make little icons).  If you were writing professionally, you'd generally type double-spaced (i.e., with a blank line after each line of words), mark up the results with a pencil, literally cut and paste to rearrange, and then retype the whole thing for the next draft.

So if you're sitting at one of Goldkamp's typewriters, expressing your thoughts, you're making a record of every typo and every false start,  all in the order it came out, on a nice, tangible piece of paper that you produced yourself.  At least at first, it will be the only copy of those words in the world, and if someone wants to read it, they'll have to come see it.

Pretty much the antithesis of publishing on the web, and pretty much the way things were for around a hundred years until the "Personal Computer" came along.


Goldkamp is by no means a technophobe.  Among other things, he runs the web site freshpoetrystl.org,  where you can see "A selection of poems that turned out okay", such a typically Midwestern way of putting it, and one of the best page titles I've ever seen.   For my money, the poems I looked at did come out okay, if not better.

1,259,997-way tie for last

Honestly, I'd forgotten all about Technorati.  Back in 2007, a friend suggested I create a Technorati profile, and so I did.  For a while I watched closely for signs that I was rising in the ranks.  Sure enough, I soon reached a Technorati authority of ... two.  At the time, there were several million blogs listed on Technorati, and that was enough to lift me out of the bottom million or so.  Eventually the link that had lifted me to authority two fell away and I was back at the bottom.  Sic transit gloria mundi.

In 2010 I removed the Technorati widget in favor of something else, which has since been itself replaced  (I now have the usual suspects: email, Blogger, Twitter, FB and g+).  And then I forgot all about it.  At one point, I thought, I'd heard that Technorati had been acquired or shut down or such.  More on that in a bit.

Looking through a search for what links to Field Notes (not very much at all), I saw that it was still listed on Technorati, which, of course, meant that Technorati was in fact still around.  Even more surprising, Field Notes had rocketed up to a rank of 71,375 out of 1,331,372, with an authority of ... one.  As I said on the other blog, I don't put that much stock in ranking statistics, but ... interesting.

Hmm ... I thought.  There must be a lot of blogs with authority zero below a bunch of blogs with authority one.  But no.  There are lots and lots of blogs with authority one.  Everyone from 71,375th place on down.  That's one long tail.

I did a little binary searching.  The rank of  71,375 first appears in the middle of page 2885, with 25 listings to a page.  I make it to actually be 72,145, but maybe I missed something.  Right before the rank 71,375 blogs start is a block of rank 57,611 blogs with authority ... 79.  So basically rank one is "we have no idea who you are and no one else seems to", and 79 means "you're in our universe, but at the bottom".

What changed from the "number of blogs linking to yours in the last six months?" days? Actually, the half-rumor in the back of my head about Technorati's acquisition or demise was really my misremembering that they'd done a redesign sometime in 2009.  They changed their look a bit, but they also changed how authority was calculated.  Instead of being a count of links in the past six months, it was a number assigned from 1 to 1000 based on link activity in the past month.  This change was designed to make Technorati more responsive to trends in the blogosphere.  Presumably, recent activity is weighted more heavily in the calculation.

This fits with the rankings I found.  Evidently, almost all of the blogs in the Technorati database are islands unto themselves, and Technorati only tracks about 70,000 blogs in any meaningful way.

Hey, I don't really care.  I've made myself immune to the vagaries of blog ranking entities by the simple expedient of not making any money off this blog and instead being happy that people read it at all.  Amateur status has to be good for something hasn't it?  But it does seem a bit underwhelming that a site that claims to be keeping its finger on the pulse of a millions-strong blogosphere really just seems to be tracking how a much smaller number of blogs, many of them already well-known major players, point at each other.

Considered as a portal into a collection of blogs, with an entertaining front page and "what's hot" listings, Technorati seems fine, but as a measure of where a particular blog is in the world, not so much.  For that, I'd probably turn to Alexa.  Field Notes is #28,898,459 there, out of "over 30 million websites" (not just blogs).  That I can believe.

Spam links and the economics thereof

I can't remember the word for this, but I know I don't like it ...

You go to a more-or-less reputable news site, say for a mass-market but not-quite-tabloid newspaper, and read an article.  To the side of the article is a list of related articles.  Towards the bottom of the list, or maybe in a different section that looks pretty much identical to the first list, are links to a few more articles, with catchy headlines and/or pictures.

Except they're actually external links to infomercial-style ads, or fluffy articles with ads in every direction, or fluffy articles that want you to wade through an extra page of ad links before you can even see them.

Nothing illegal or immoral going on here, just annoying, and somehow more annoying because there's generally some hint on the original page -- fine print, an "elsewhere on the web" section header or such -- to let you know that these are not from the same source.  If you're looking.

It's a trade-off of reputation for revenue.  Roughly, the levels are

• Top news sites (at least "top" in my estimation), just don't do this.
• Many sites link to sites with somewhat more ads and somewhat less substantial articles, and make it clear that "that's them, not us, but try this if you're looking for mindless fun"
• Some sites link to outright crap and make only a minimal "see, it says right here" effort to disclose it.
• And then there is outright phishing, which is immoral, and often illegal.  Sort of an informational bait-and-switch.

The interesting territory is in the middle.  You can probably tell something about how a site is faring if it shifts one way or another on this scale.

Still here, not much to report at the moment ...

... but it has recently come to my attention that Wikipedia has an article devoted to fictional actuaries.

Can your online crowdsourced reference work say that?

Just when I thought I understood this "e-commerce" stuff ...

... I was looking for clothes on a major retailer's web site.  Scrolling down, I see a bunch of "sponsored links" at the bottom.  For other stores.

For the same items I'm looking for.

For less.

Huh?

A piece of history

Saw this on my g+ feed: CERN has re-posted the first web page ever (or at least the first one they put up for public consumption), 20 years after the fact.  I'm not sure if it's under the original URL, but http://info.cern.ch/hypertext/WWW/TheProject.html sure looks like it could be.  Anyone remember "hypertext"?

Like a lot of efforts in the early stages, much of the content is about the project itself.  For example, there's a "Line Mode" browser:

The LineMode Browser is suitable for use on dumb terminals, requiring no control sequences except for carriage return and line feed.

"Dumb terminals"?  "Control sequences"?  "Carriage return"?  What language are they speaking? (Of course, Lynx is still in business.)

In any case, it says something that CERN can bring back the original web page from the archives twenty years later, in all its textual glory, and it still works, and people really can see it, world-wide.

Vires in numeris, or salsipuedes?

After the last few days, it's pretty clear there are some significant obstacles to Bitcoin becoming a widely-adopted currency.   Recapping a bit:

• Its exchange rate against reserve currencies is extremely volatile.
• It is fairly illiquid.  I've watched several Bitcoin exchanges now.  I've seen significantly slow servers, extremely thin order books, huge spreads and significant price differences across exchanges (arbitrage, anyone?).  This is simply not a mature market by any stretch.
• There does not appear to be very much of a pure Bitcoin economy.  For the most part, a price quoted in Bitcoin is actually a price in some reserve currency converted at the going rate, and quite likely the seller is going to turn around and convert the proceeds to a reserve currency at the next opportunity.
• While the Bitcoin protocol appears reasonably secure, there are plenty of other ways to lose Bitcoins, and once they're gone, they're gone.
• While Bitcoin aims to be a decentralized currency free from the influence of banks and governments, most Bitcoins are held by one of a small number of accounts.
• Right now The Man does not seem particularly interested in trying to control Bitcoin use.  That could change.
• Bitcoin's deflationary model, under which the money supply will never be more than about twice what it is now, has drawn considerable fire from economists.
• Bitcoin is far from the first attempt at establishing a private currency, or an internet currency (anyone remember Flooze, or Beenz?).  It may well be the first, or one of the first, with its particular technical characteristics, but it's not clear at all that that buys anything.  Private currencies as a whole do not have a great track record.

And yet ...

A currency works if enough people want it to work.  If a solid cadre of core Bitcoin users arises (or has already arisen), doing its business entirely or almost entirely in Bitcoin, then (as the Economist's Free Exchange blogger points out) they would not need to care much how many dollars a Bitcoin was or wasn't worth.  In the ideal case, it becomes more and more convenient for someone who accepts Bitcoin in payment to turn around and use it directly for something else of value, rather than exchanging it, the economy grows and we're off.

The acid test is whether there can be stable Bitcoin prices, that is, whether someone offering an item for X Bitcoins today will be offering the same item for X Bitcoins tomorrow, and roughly X next week, and next month.  This would allow people to save or invest Bitcoin like an ordinary currency.  This probably still requires stable exchange rates with reserve currencies.  If Bitcoin appreciates, that X Bitcoin price starts to look expensive and buyers will be strongly tempted to buy from someone who takes dollars or Euros.  Likewise, if Bitcoin depreciates, sellers will be strongly tempted to raise the Bitcoin price, lest outsiders take advantage of the low prices.

To make all this work, we need some combination of

• Stable exchange rates with reserve currencies
• Significant ecosystems of items available only in Bitcoin
• Enough people with a strong determination to use Bitcoin no matter what, regardless of the financial consequences

My feeling, personally, is that the odds are stacked pretty heavily against this, but I've been wrong many times before.

(Vires in numeris, Latin for strength in numbers, is Bitcoin's semiofficial motto.  Salsipuedes is informal Spanish for "Get out if you can.")

A note on Bitcoin foreign exchange

Bitcoin has been extremely volatile of late.  That doesn't speak well for its potential as a currency, but the  Economist's Free Exchange blogger makes a good point:  Volatility is only a problem to the extent you have to exchange with other currencies.  If the virtual nation of Bitcoinia can transact its business in pure Bitcoin, then it matters much less how many dollars people think a Bitcoin is worth.  Even so, the examples of volatility cited in that article, as much as 20% in a matter of weeks, are far, far below what Bitcoin has experienced over the past month or so.

Right now we don't seem very close to such a scenario.  There don't seem to be a lot of places where goods or services are priced purely in Bitcoin.  Rather, the real price is in some reserve currency and the Bitcoin price is based on the going rate.  For at least some period of time, transacting in Bitcoin will generally involve converting to and from reserve currencies.  Bitcoin foreign exchange, if you will.

That's a bit of a problem.  To use one of the several existing Bitcoin exchanges, you have to provide an account in conventional currency, subject to conventional banking regulations and so forth, which is pretty much the opposite of what Bitcoin is supposed to accomplish.  Besides that, having a single point of failure that can bog down or fall over at any point (and is more likely to just when the need for it is most urgent), seems silly.  Finally, I've seen complaints that exchanges can tend to charge fairly heavy transaction or withdrawal fees, which Bitcoin is supposed to eliminate.  Can we do better?

It shouldn't be a problem to track bids, offers and trades in a distributed manner.  Bitcoin already does much the same with its blockchain, and there are various flavors of distributed hash table running around.  I'm sure someone could come up with something.

The bigger problem, I think, is escrow, since this is all supposed to be pseudonymous.  Suppose I call myself Mr. Blue and I own a Bitcoin wallet.  Ms. Green is willing to give me $X for one of my Bitcoins.  We can both make sure the world knows this without going through a central exchange.

But how do we actually settle the trade?  I need to give Ms. Green one Bitcoin.  Bitcoin makes that easy.  I initiate a transfer and few minutes to an hour later the Bitcoin world agrees that that Bitcoin is hers (or, more precisely, is now in the wallet whose identity she gave).  Ms. Green then disappears and I'm one Bitcoin poorer.

To make this work, Ms. Green needs to put $X in a safe place.  At that point, the dollars are no longer hers, but neither are they mine.  When the Bitcoin transfer settles, and only then, I get access to the dollars and transfer them to my account.

What we need, then, is is a way of moving currency from one conventional bank account to another, securely, anonymously and cheaply, without having to trust any particular third party.

I don't know if that's technically feasible or not, but if it is, it's not clear why we would need a virtual currency.

Dollars accepted here

Having grown tired of waiting for MtGox to update, I had a look at Bitfloor [which folded not long after I wrote this].  Its order book seems to work considerably better.  From that, it looks like the Bitcoin market is moving again, which does not come as a great surprise.

Then I had a look at Bitcoin Store, where you can use your Bitcoin to buy goodies like digital cameras, memory and other such.  Prices are quoted in dollars and Bitcoin.  The Bitcoin price is simply the dollar price converted at the going rate from MtGox.  That is, if Bitcoin moves against the dollar, it's the Bitcoin price that changes.  The dollar price stays the same.

While it's perfectly legitimate for Bitcoin Store to say that it's accepting Bitcoin, and it is taking on some risk in the doing so, if it's pricing in dollars, it's effectively accepting dollars.