DMT Week 2: Networks

First thoughts …

The main impact on me from this week’s lecture was that I started thinking about how the giant ‘network’ of the internet fits together, in particular what hooks it together, specifically:

  1. Common (if numerous) protocols; and
  2. Distributed ownership.

The internet’s growth over the past 10-15 years has been phenomenal, but was it the first network? I’d argue no … whilst not in communications, railways and the telephone are just two examples of a new technology that took decades longer to become mainstream, despite being similarly groundbreaking for their time. So, allowing for advances in technology, how is the internet different?

Interoperability and a common platform

The internet is not solely “owned” – or operated – by large corporations. A typical result of such ownership (see my thoughts on intranet centralisation) would be said corporation(s) segmenting by different audience groups, cutting service to the less profitable, preapproving (at least at a macro level) content and so on. So the ideas and contributions of, say, the non able-bodied or low-income, would be automatically excluded unless there was a dollar to be made … and of course this would also be influenced by what content they were driven toward by those owning the system.

Going back to the railway example, after Federation in 1901 it made sense that a train could travel on any railway in Australia … and yet this was not possible, owing to each individual state having their own rail gauge (protocol!) that did not match any of the others. This is an example at the middle level, but gives an illustration of how giving ultimate control at a certain level can affect all users.

The internet, on the other hand, is controlled by both corporations, government and the audience, with no one group (despite the best attempts of certain parties) gaining an edge over any other. As a result, the medium to date has evolved similar to any other community – a great deal of differences, lots of variations in use and expertise etc – but with enough similarity and common conventions to provide access for all. Anyone can, within reason, own either a client or server piece of a network, meaning they are not constrained or dictated to by the rest of the network.

Role models

So if not the railway, what is the best role model for the internet as a network that is accessible to all?

  1. Languages are a great example. English or French both have grammatical rules and a vocabulary honed by the masses over hundreds of years, and yet still manage variation by individuals – be it at the macro level of defined regions (eg. the Québec dialect of French) or at the individual through slang or common expressions. The same root language, but many different ways of use.
  2. Likewise, road signs all over the globe vary in terms of language and specifics, but at their core retain many conventions in most countries – for example the red octagon of the English-language Stop sign (Arrete in French) or the Triangle for Give Way (Geill Sli in Irish). Again, commonly understood conventions for the masses, but differences for the individual.

Thus both the masses and each individual are served by the same network, with both having their objectives met to at least some extent.

BUT does this stifle innovation?

Keeping the masses together at a similar level can also lead to difficulty: if an individual or company builds a new protocol, it’s natural that adoption amongst the rest of the user community will take a lot of time and effort.

The ‘browser wars’ of the 1990s were primarily between Microsoft and Netscape – both added proprietary (ie. product-specific) features in the hope that they would eventually dominate the market, therefore defining user needs themselves. However, as no permanent domination was achieved, eventually all players gave at least lip-service to following the standards laid out by the W3C, ensuring – at least for now – general commonality of the protocol. But with more innovations such as Ajax and Silverlight continuing to be developed, this will continue to be an evolving area.

Going forward …

As most technologies evolve and reach maturity, protocols become embedded, and the rate of innovation tends to decrease. With the internet however, the nature of the network and protocols – being defined by potentially every user – seems to be ensuring a much longer lifecycle. This is illustrated by the approach of the W3C to Web Accessibility Guidelines, where the relatively fixed approach of version 1.0 ten years ago has been revised for version 2.0 to focus on general principles that can be broadly applied to a range of technology – not without issues.

Perhaps a good comparison is building a street and road network. Early on, the flattest ground is built on, so a grid or straight highway is easily constructed. Later, as the network expands, less hospital terrain such as mountains makes the work much more challenging in itself, as well as having to match the precedents and expectations of users set previously.

Leave a Reply

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.