Daily Bits – distributed storage, tweetstorms and cities – June 22nd, 2017

I’ve been pondering for a few days now the question of why decentralized file storage is better than centralized file storage with a trusted third party that can be held accountable if things go wrong.

And then this morning I come across this article published on Coin Center’s blog, written by Juan Benet, Jesse Clayburgh and Matt Zumwalt of the Protocol Labs team (a startup that focuses on data storage problems – so they know a thing or two).

These are the arguments against centralized cloud storage:

“This architecture makes the web brittle, undermines privacy, allows the price of storage to remain artificially high, and creates bottlenecks that prevent innovative new uses of data.”

Okay, I get that.

I’m also intrigued by the idea of locating files by what they are, not where they are. If a webpage calls a file now, it references it by where it’s stored. That, as the authors say, does make the web “brittle”. Move the file, and the call no longer works.

The authors helpfully compare this to pulling library books by their location, not their title.

Add to that the economies of scale of harnessing all that unused computing capacity (just on my laptop I probably have a fair chunk).

So, I’m liking the idea. But I still don’t see how this system will get enough of a network effect going to be practical. Website code will need to be re-written, and given the vast size of the ecosystem, the incentives will be tricky. If indeed possible.

But, just because something is “too big to change”, doesn’t mean we shouldn’t try, right? Look at the valiant cryptoeconomists working on imagining a new central banking system.

A daunting task, though. But one that I am now more interested in.

— x —

An epic 37-tweet thread from Naval Ravikant that, although it’s long, showcases the discipline of having to work within 140-character limits:

It includes gems such as this:

— x —

Diverging from blockchains for a second, I came across an illuminating article on the three different types of cities: suburban (built for cars), hypertrophic (built for transit) and traditional (built for walking).

This reminded me of something I read recently by urbanist and philosopher Jane Jacobs, who asked: “Are we  building cities for people or for cars?” I know which type I prefer (hint: I don’t even have a car.)

I do, however, recognize the impracticality of having a totally walkable city. Pity.

city traditional
Traditional city


Hypertrophic city
Hypertrophic city


Suburban city
Suburban city

— x —

Finally, an article on digital identity that recognizes that the incumbents may have a say in how this turns out. But, they may be amenable to a business model mutation: “Hey, so you lose all this income from the identities you control, but just think about all the extra users you’re gonna get!”

Okay, it needs some work, but it’s a start.

Moral: the incumbents aren’t going to let go of their stronghold on our identities without a good reason.



Leave a Reply

Your email address will not be published. Required fields are marked *