Myself and my fellow digital archaeologists, Kalle Everland and Jesper Lycke, were very happy to welcome Dan Noyes and his colleagues from CERN to our Error 404 exhibition yesterday.
Dan’s leading a project to restore the first website, document early browsing experiences and archive documentation related to the birth of the Web. When he heard we had reunited the first webpage with the first browser, he decided to come and check it out for himself.
Whilst at CERN, Tim Berners-Lee had a state-of-the-art NeXT Cube. Consequently, his browser only runs on the NeXTSTEP Operating System, which is now obsolete. Machines running this O/S are increasingly rare. We got our Cube from Rob Blessin, who knows as much about NeXT as anyone in the world. Rob recreated the environment TBL would have had at the time on our machine. He installed a version of TBL’s WorldWideWeb browser from 1991, a version of the updated and renamed Nexus browser from 1994 and an archived version of the first website from 1992.
Both browsers are buggy. The navigation panel doesn’t work at all on the WWW browser and doesn’t operate as you would expect on the Nexus. After a significant amount of trial and error, Jesper, Kalle and I worked out its idiosyncrasies. The most significant of which, is that the active page is not recorded in the browser history. Combined with the fact that navigation panel is designed to operate the links on the preceding page, this results in a very confusing browsing experience. In short, the navigation panel operates links from two pages back, rather than the page you are on! Throw in multiple windows, dead links and anchors acting like links and it makes for a fun time.
Once you get your head around the fact the navigation panel operates the preceding page and the active page doesn’t exist (@_@), you can really enjoy Berners-Lee’s original vision hands on. There are lots of details that I will go into at a later date but the most obvious ways in which the Nexus browser differs to a modern browsing experience are as follows:
1) The browser menu is independent of the browser window. Opening a link, opens a new window, creating a multi-window environment, very different from the single page browsing common today.
2) WWW/Nexus is a browser editor, it can be used to create and edit webpages as well as view them. What became apparent on this visit, where we set-up a web server rather than accessing the files locally, is that you can only edit local files. This makes me question my assumption that the Web was envisaged as a multi-author environment from the start. Further investigation required.
3) Finally, and perhaps crucially, the Nexus browser accommodates two types of browsing. The user can create their own user journey, by opening links as they please, much as today. They can also use the navigation panel to take a document-centric journey, following links sequentially. In other words, the site can be navigated in a linear fashion, in a series determined by the author.
Dan is also responsible for CERN’s current website and brought along Mark Boulton (no relation) of Mark Boulton Design. They noted how the Nexus browser addressed some of the recent requests they’ve had for the CERN site. Who knows? Twenty years on, the guiding hand of the author could make a return to the browsing experience. Watch this space.