Great piece indeed, thanks for sharing.

A friend of mine said he couldn't read it though, because he doesn't
have a LinkedIn account?

In case anyone else is encountering the same problem, here the flat text.

Cheers, Aaron.

==========

Why Bitcoin is and isn't like the Internet
Joichi Ito – Director, MIT Media Lab

In the post that follows I’m trying to develop what I see to be strong
analogues to another crucial period/turning point in the history of
technology, but like all such comparisons, the differences are as
illuminating as the similarities. I'm still not sure how far I should
be stretching the metaphors, but it feels like we might be able to
learn a lot about the future of Bitcoin from the history of the
Internet. This is my first post about Bitcoin and I’m really looking
more for reactions and new ideas than trying to prove a point.

Feedback and links to things I should read would be greatly appreciated.
I’m fundamentally an Internet person -- my real business life started
around the dawn of the Internet and for most of my adult life, I’ve
been involved in building layers and pieces of the Internet, from
helping start the first commercial Internet service provider in Japan
to investing in Twitter and helping bring it to Japan. I’ve also
served on the boards of the Open Source Initiative, the Internet
Corporation for Names and Numbers (ICANN), The Mozilla Foundation,
Public Knowledge, Electronic Privacy Information Center (EPIC), and
been the CEO of Creative Commons. Given my experiences in the early
days of the net, it’s possible that I’m biased and everything new
looks like the Internet.

Having said that, I believe that there are many parallels between the
Internet and Bitcoin and there are many lessons from the Internet that
can help provide guidance in thinking about Bitcoin and its future,
but there are also some important differences.

The similarity is that Bitcoin is a transportation infrastructure that
is decentralized, efficient and based on an open protocol. Instead of
transferring packets of data over a dynamic network in contrast to the
circuits and leased lines that preceded the Internet, Bitcoin’s
protocol, the blockchain, allows trust to be established between
mutually distrusting parties in an efficient and decentralized way.
Although you could argue that the ledger is “centralized”, it’s
created through mechanical decentralized consensus.

The Internet has a root -- in other words, just because you use the
Internet Protocol doesn’t mean that you’re necessarily part of the
Internet. To be part of THE Internet, you have to agree to the names
and numbers protocol and root servers that are administered by ICANN
and its consensus process. You can use the Internet Protocol and make
your own network, using your own rules for names and numbers, but then
you’re just a network and not The Internet.

Similarly, you can use the blockchain protocol to create alternative
bitcoins or alt.coins. This allows you to innovate and use many of the
technological benefits of Bitcoin, but you are no longer technically
interoperable with Bitcoin and do not benefit from the network effect
or the trust that Bitcoin has.

Also like the beginning of the Internet, there are competing ideas at
each of the levels. AOL created a dialup network and really helped to
popularize email. It eventually dumped its dialup network, its core
business, but survived as an Internet service. Many people still have
AOL email accounts.

With crypto-currencies, there are coins that don’t connect to the
“genesis block” of Bitcoin -- alt.coins that use fundamentally the
same technology. There are alt.coins that use slightly different
protocols and some that are fundamentally different.

On top of the coin layer, there are various services such as wallets,
exchanges, service providers with varying levels of vertical
integration -- some agnostic to whichever cryptocurrency ends up
“winning” and some tightly linked. There are technologies and services
being built on top of the infrastructure that use the network for
fundamentally different things than transacting units of value, just
as voice over IP used the same network in a very different way.
In the early days of the Internet, most online services were a
combination of dialup and x.25 a competing packet switching protocol
developed by Comité Consultatif International Téléphonique et
Télégraphique, (CCITT), the predecessor to the International Telecom
Union (ITU), a standards body that hangs off of the United Nations.
Many services like The Source or CompuServe used x.25 before they
started offering their services over the Internet.

I believe the first killer app for the Internet was email. On most of
the early online services, you could only send email to other people
on the same service. When Internet email came to these services,
suddenly you could send email to anyone. This was quite amazing and
notably, email is still one of the most important applications on the
Internet.

As the Internet proliferated, the TCP/IP stack, free software that
anyone could download for free and install on their computer to
connect it to the Internet, was further developed and deployed. This
allowed applications that ran on your computer to use the Internet to
talk to other programs running on other computers. This created the
machine-to-machine network. It was no longer just about typing text
into a terminal window. The file transfer protocol (FTP) and later
Gopher, a text-based browsing and downloading service popular before
the web was invented, allowed you to download music and images and
create a world wide web of content. Eventually, permissionless
innovation on top of this open architecture gave birth to the World
Wide Web, Napster, Amazon, eBay, Google and Skype.

I remember twenty years ago, giving a talk to advertising agencies,
media companies and banks explaining how important and disruptive the
Internet would be. Back then, there were satellite photos of the earth
and a webcam pointing at a coffee pot on the Internet. Most people
didn’t have the imagination to see how the Internet would
fundamentally disrupt commerce and media, because Amazon, eBay and
Google hadn’t been invented -- just email and Usenet-news. No one in
these big companies believed that they had to learn anything about the
Internet or that the Internet would affect their business -- I mostly
got blank stares or snores.

Similarly, I believe that Bitcoin is the first “killer app” of The
Blockchain as email was the killer app for the beginning of the
Internet. We are in the process of inventing eBay, Amazon and Google.
My hunch is that The Blockchain will be to banking, law and
accountancy as The Internet was to media, commerce and advertising. It
will lower costs, disintermediate many layers of business and reduce
friction. As we know, one person’s friction is another person’s revenue.

One of the main things we worked on when I was on the board of ICANN
was trying to keep the Internet from forking. There were many
organizations that didn’t agree with ICANN’s policies or didn’t like
the US’s excessive influence over the Internet. Our job was to listen
to everyone and create an inclusive and consensus-based process so
that people felt that the benefits of the network effect outweighed
the energy and cost of dealing with this process. In general we
succeeded.

It helped that almost all of the founders and key technical minds and
technical standards organizations that designed and ran the Internet
worked together with ICANN. This interface between the policy makers
and the technologists -- however painful -- was viewed as something
that wasn’t great but worked better than any of the other alternatives.

One question is whether there is an ICANN equivalent needed for
Bitcoin. Is Bitcoin email and The Blockchain TCP/IP?
One argument about why it might not be the same is that ICANN
fundamentally had to deal with the centralization caused by the name
space problem created by domain names. Domain names are essential for
the way we think the Internet works and you need a standards body to
deal with the conflicts. The solutions to Bitcoin’s centralization
problems will look nothing like a domain name system (DNS), because
although there is currently centralization in the form of mining pools
and core development, the protocol is fundamentally designed to need
decentralization to function at all. You could argue that the Internet
requires a degree of decentralization, but it has so far survived its
relationship with ICANN.

One other important function that ICANN provides is a way to discuss
changes to the core technology. It also coordinates the policy
conversation between the various stakeholders: the technology people,
the users, business and governments. The registrars and registries
were the main stakeholders since they ran the “business” that feeds
ICANN and provides a lot of the infrastructure together with the ISPs.

For Bitcoin it’s the miners -- the people and companies that do the
computation required to secure the network by producing the
cryptographically secure blockchain at the core of Bitcoin -- all in
exchange for bitcoin rewards from the network itself. Any technical
changes that the developers want to make to Bitcoin will not be
adopted unless the miners adopt them, and the developers and the
miners have different incentives. It’s possible that the miners have
some similarities to the registrars and registries, but they are
fundamentally different in that they are not customer-facing and don’t
really care what you think.

As with ICANN, the users do matter and are key for the network effect
value of Bitcoin, but without the miners the engine doesn’t run. The
miners aren’t as easy to identify as the registrars and registries and
it’s unclear how the dynamics of incentives for the miners will
develop with the value of bitcoin fluctuating, the difficulty of
mining increasing and the transaction fees being market driven. It’s
possible that they will develop into a community with a user interface
and a governance function, but they are mostly hidden and independent
for a variety of reasons that are unlikely to change for now. Having
said that, one of the first publicly traded Bitcoin companies is a miner.

The core developers are different as well. The founders of the
Internet may have been slightly hippy-like, but they were mostly
government-funded and fairly government-friendly. Cutting a deal with
the Department of Commerce seemed like a pretty good idea to them at
the time.

The core Bitcoin developers are cypherpunks who do what they do
because they don’t trust governments or the global banking system and
are trying to build a distributed and autonomous system, one that is
impervious to regulation and meddling by anyone at any time. At some
level, Bitcoin was designed to not care what regulators think. The
miners have an economic interest in Bitcoin having value, since that’s
what they’re paid in, and they care about scale and the network
effect, but the miners probably don’t care if it’s Bitcoin or an
alt.coin that ends up winning, as long as their investments in
hardware and plant don’t disappear before they make a return on their
investment.

Regulators clearly have an incentive to influence the rules of the
network, but it’s unclear whether the core developers really need to
care what the regulators think. Having said that, without some sort of
buy-in by regulators, it’s unlikely to scale or have the mainstream
impact that the Internet did.

Very much like the early days of the Internet, when we saw the power
of Internet email but hadn’t yet invented the Web, we are just
imagining the potential uses of concepts such as crypto-equity and
smart contracts … to name just a few.

I believe it’s possible that over-regulation could cause Bitcoin or
the blockchain to never achieve its full potential and remain a
feature of the side-economy, much in the same way that the Tor
anonymizing system is extremely valuable to people who really need
privacy but not really used by “normal people”... yet.

What helped make the Internet successful was the lack of regulation
and the generally inclusive and permissionless nature of innovation.
This was driven in large part by free and open source software and the
venture capital community. The question I have is whether the fact
that we’re now talking about “money” and not “content,” and that we
seem to be innovating at a much higher speed (venture capital
investment in Bitcoin is outpacing early Internet investments), the
dialog in popular media is growing, and governments are very
interested in Bitcoin makes this a completely different game. I think
ideas like the five-year moratorium on Bitcoin regulation proposed by
US Representative Steve Stockman are a good idea. We really have no
idea what this whole thing is going to turn into, so a focus on dialog
versus regulation is key.

I also believe that layer unbundling and innovation at each layer,
assuming that the other layers will sort themselves out, is a good
idea. In other words, exchanges and wallets that are coin-agnostic or
experiments with colored coins, side chains and other innovations that
are “unbundled” as much as possible allow the learnings and the
systems created to survive regardless of exactly how the architecture
turns out.

It feels a lot to me like when we were arguing over ethernet and token
ring -- for the average user, it doesn’t really matter which we end up
with as long as in the end it’s all interoperable. What’s different is
that there is more at stake and it’s moving really fast, so the shape
of failure and the cost of failure might be much more severe than when
we were trying to figure out the Internet and a lot more people are
watching.

On 22-01-15 17:15, Amir Taaki wrote:
> wow very enlightening. thanks for the post.
>
> On 01/22/2015 09:16 AM, Ben Vickers wrote:
>> https://www.linkedin.com/pulse/why-bitcoin-isnt-like-internet-joichi-ito
>>
>>
>> Joichi posits an interesting point of leverage which might be
>> useful to some people here for repositioning things.
>>
>>
>> _______________________________________________ unSYSTEM mailing
>> list: http://unsystem.net
>> https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/unsystem
>>
> _______________________________________________ unSYSTEM mailing
> list: http://unsystem.net
> https://mailinglists.dyne.org/cgi-bin/mailman/listinfo/unsystem
>