I awaken this morning feeling optimistic. I’d seen last night that a big announcement from Coinbase was coming, and the WallStreetJournal appeared to confirm that the announcement was to be a fully regulated U.S. exchange. The price shot up immediately and was still rising when I went to bed.
Now, I look at the price widget on my phone and my optimism vanishes. If Bitcoin price is any measurement of enthusiasm, it’s clear before I get downstairs that Coinbase’s big revelation hasn’t quite lived up to expectations.
I head to the new exchange site, log in with my Coinbase credentials and…
I suspect this explains the lack of enthusiasm.
I turn to Reddit, Twitter—all my usual sources—and hear many other voices telling the same story: Coinbase’s Exchange isn’t available in their state. A friend on Twitter links me to a map from a Coinbase blog post (http://blog.coinbase.com/post/107618322282/usd-wallets-in-7-more-states)
That’s their service area.
To be fair, this looks worse than it is. Our population isn’t evenly distributed here in the U.S. and they did get California, the most populous state…but they also missed Texas and Florida, Nos. 2 and 3 on that list. A Wikipedia article and an Excel spreadsheet later, I discover that your odds of being in Coinbase’s service area are basically 50/50. I seem to have lost the coin toss.
Thankfully, the world is a connected place these days and it isn’t hard to find someone in one of those green states who will let me digitally shoulder-surf. I’m irritated, but if I really cared I could incorporate in California or something. There are always ways around these things.
Access problems (sort of) bypassed, I begin to feel more optimistic. The interface is simple, elegant and clean. There aren’t an overwhelming number of options presented, but that’s not a bad thing. Simplicity is the cornerstone of user experience, after all. But even this small positive experience would prove ephemeral.
Coinbase accounts and Coinbase Exchange accounts are separate entities. I suspect this is for regulatory reasons, and you can transfer funds between the two with a few clicks, but it does add a little friction to the experience—and you’ll probably have to make transfers often. More on that later.
My friend and I throw a little bitcoin into the exchange account and make a few test trades. The interface is simple, minimalist and obvious in all the right ways. The trades execute precisely as you’d expect them to. The order book continues the trend of simple design. For all its failings, it is abundantly clear that a lot of design work went into this interface. “Responsive” and “flat” are both buzzwords that apply here. Unfortunately, so is “minimum viable product.”
I don’t want to put this thing down too much. It’s a work of art, and as far as I can tell, it has a rock-solid infrastructure. Thus far it has chewed through more than 5,000 BTC worth of trade volume without any noticeable hiccups. The front end appears to be socket-driven, meaning it updates in realtime as events occur, not every n seconds. That feature is bound to be a day trader favorite. The foundation of this system is excellent—but there’s not much there besides foundation.
For example, this is what the buy/sell interface looks like:
Two big tabs, two textboxes, one big button. That’s simplicity, which is a good thing. But wait, what if I want to place a market order?
(A quick aside, for the non-traders in the crowd. There are typically two kinds of orders you can place on an exchange: market orders and limit orders. A market order is what most people are referring to when they say “buy” or “sell—you’re telling the exchange “Here’s $500, buy bitcoins until it runs out.” But to do that there have to be outstanding sell orders. Where do those come from? Limit orders. A limit order has conditions. On the buy side, it says, “Here’s $500, buy bitcoins with it if the price goes down to $250.” On the sell side, it says, “Here are 10 bitcoins. Sell them when the price goes up to $250.”)
I hunted for a market order option. There were none to be found. Remember when I said you’d be transferring funds a lot? That’s because if you want to place a simple market order, you have to do it here:
As previously mentioned, Coinbase and Coinbase Exchange appear to be separate-but-connected systems. Market orders and limit orders are the two most fundamental operations on any exchange, and Coinbase has split this functionality across two websites. This is where “minimum viable product” (MVP) comes into play.
MVP is a product development strategy. In simple terms, it means you build only the core features necessary to deploy. Since Coinbase already handled market orders, the MVP for Coinbase Exchange wouldn’t duplicate that feature. MVP is a great way to create a “starting point” that you build on later—but you should never release an MVP to a general audience. MVPs are usually released to limited audiences of devs and early adopters—beta testers, essentially. These audiences know they’re looking at unfinished software, and they tend to be a lot more forgiving than the general public.
I highly suspect, for these reasons, that we’re actually looking at unfinished software to match the incomplete licensing. Some big names have invested a lot of money in Coinbase, and it got a lot of media attention as a result. My best guess is they released early to preserve that media “momentum” (maybe a touch too early).
O.K., so it failed to live up to the hype—but there was a lot of hype. At the end of the day it’s a pretty solid offering. I traded on Mt. Gox back when Dwolla was the only funding option; I know what bad looks like, and this isn’t it. This is a hot rod engine and a gorgeous paint job on a car with no seats or doors. It’s the first half of a beautifully written but entirely unfinished symphony. Not bad, just…incomplete.
For more posts from David, visit Coding in My Sleep.