A friend asked: “So you don’t think smart contracts are ever going to be … secure and useful?”

Here’s my response:

“Ever” is a long time, but … let me tell a quick story:

When I raised venture capital during the dot-com bubble, if memory serves, the combined legal bill for my company’s Series A paperwork was in the neighborhood of $70,000. Wilson Sonsini and the like. Good lawyers. In spite of all this money and time and expertise, various documents signed by the parties had a company name in them that was not my company’s. Probably because the lawyers copied and pasted from a previous similar company.

Now, among all the bugs those legal papers might have contained, spotting that This Is Not The Right Company Name should be the easiest of all. Nobody did, except me because I’m a stickler for reading the fine print if I’m supposed to sign at the dotted line. And I’m sure there were lots of other “bugs” that I didn’t see because I’m not a lawyer after all.

How much would the paperwork for the Series A have cost if it had to be correct at the level of an automatically executing algorithm that cannot be revoked with “oh, oops, I just noticed …”?

(The wrong company name problem here of course was not really one, because nobody could conceivably argue in court that they thought it didn’t apply to them — it was part of a package all signed at the same date by the same parties. Not so on the block chain.)

So: at the very minimum, smart contracts are only ever going to work for very large transactions, or a very large number of identical small transactions. Which excludes most transactions. Ergo: they may have applicability, but as a niche application.