If The Internet Had Been Created By Committee … We Would Possibly Still Be Waiting and Need A…

If The Internet Had Been Created By Committee … We Would Possibly Still Be Waiting and Need A Licence To Connect

For innovation: vision, mission and gut instinct often triumph over committees

The Internet is possibly the greatest creation in the history of humankind. It has remodelled our society and has broken down the barriers to knowledge, and where virtually every person on the planet has the ability to access knowledge that had previously been hidden away in libraries. But, it was created by people who had a vision and a mission to succeed.

I have never been a great belief in running innovation through a committee. To me, innovation needs leaders and people who — with all the data that they can gather — will go with their gut on key decisions.

To me, if the Internet had been created by a committee, we would probably still be waiting for it to be created. Leadership comes from people who have a vision and a mission. And it is to people like Larry Roberts who was willing to stand up against the normal practice of giving contracts to the major computer and defence companies and give the contract for IMPs (Interface Message Processors).

We would know these IMPs as network routers these days, and they are the devices that know how best to get from one place to another — and where “best” might just mean that they are able to quickly find a route. When Larry was planning for ARPANET, he focused on building an architecture which was non-centralised and where an IMP was used on each site to route data in and out of the organisation. At the time, different computer systems were unable to speak to each other and used different operating systems. The IMP aimed to simplify things and focused on the routing of data away from the existing computing infrastructure, and thus build a gateway into and out of the network.

For this, he sent out an RFQ (Request For Quote), and where many of the major companies — such as IBM — thought that it was just not possible to build such a device and make it work for a network that spanned the East to West Coast of the US. The RFQ was a work of technical genius and was inspired by the work on packet switching by Donald Davis at NPL.

While the major players such as Raytheon and DEC bidded for the ARPANET contract, Larry decided to go with a company that was often defined as the “third university in Cambridge” (alongside MIT and Harvard): BBN.

Overall, BBN has their roots in acoustic consultancy and had been created by Bolt, Beranek and Newman. While their acoustic work struggled in places, it was their leading in using computers that became the core of the business. While IBM thought that their large System 360 was the solution and DEC with their PDP-1 computer, BBN focused on the ruggedized Honeywell DDP-512 minicomputer for the gateway:

In fact, it was so ruggedized that Honeywell gave a demo of it being attacked with a sledgehammer. The impressive thing about the BBN quote was that it not only met Larry’s specification but exceeded it by a good margin. The key advantage of the Honeywell computer was that it was significantly cheaper than the IBM and DEC equivalents.

But, where most companies would keep their secrets to themselves. The Internet moved away from standardization by committee towards the RFC (Request For Comment) method of quickly developing a standard. And the first RFC was RFC-1, and defined the IMPs [here]:

With RFCs, others could follow the basic design of the IMPs and interface into ARPANET. These documents are now published by the Internet Engineering Task Force (IETF).

Conclusions

Without leadership, innovation can struggle. Larry went with his gut, and his gut proved to be right.