WorldWideWeb: The First Web Browser Was Also an Editor
How Tim Berners-Lee's 1990 NeXT application combined browsing and authoring, which Web components it implemented, and why later browsers narrowed the model.
Tim Berners-Lee wrote the first Web browser at CERN in 1990 on a NeXT computer. Named WorldWideWeb—later renamed Nexus to distinguish the program from the project—it was also an editor. Users could follow links and directly create or modify hypertext documents. The read/write idea was present in the first working Web client, even though browsing and publishing later became separate mainstream workflows.
The browser was one piece of a larger implementation
Berners-Lee’s March 1989 proposal described linked information management for CERN. A 1990 revision with Robert Cailliau refined the project. By the end of 1990, Berners-Lee had implemented key components: URLs to identify resources, HTTP to retrieve them, HTML to represent hypertext, the WorldWideWeb client/editor, and the httpd server on a NeXT workstation.
The first website at info.cern.ch explained the project and how to run a server or browser. This self-description mattered: the system could distribute its own documentation and invite remote sites to join without a central publisher creating every link.
NeXTSTEP supplied an object-oriented development environment and rich text capabilities that helped one developer assemble a graphical application quickly. WorldWideWeb displayed formatted text and links and could edit documents in place. Its editing features were tied to the NeXT environment and early HTML’s limited structure; it was not a collaborative cloud editor in the modern sense.
Portability initially reduced the interface
The NeXT application could not run on the many terminals and workstations CERN collaborators used. Nicola Pellow developed the Line Mode Browser, a text-oriented portable client. It lacked direct graphical editing but greatly expanded who could access the Web. Other browsers followed on X Window, Macintosh, Windows, and additional systems.
This introduced a recurring tradeoff: a rich integrated authoring environment on one platform versus a broadly portable reader. As public Web use grew, browsers optimized navigation and rendering, while document creation moved to text editors, server tools, content-management systems, and later Web applications.
HTTP methods and HTML forms did not by themselves preserve universal in-place editing. Servers needed authorization, concurrency control, content formats, and write protocols. Later work such as WebDAV addressed remote authoring, while wikis and hosted editors built application-specific read/write models.
CERN’s release enabled open expansion
On April 30, 1993, CERN made Web software available on a royalty-free basis, and later issued an open release. That policy reduced licensing friction as browsers and servers spread. It did not mean every later browser, page, or Web technology was public domain; it made the core CERN implementation and protocols accessible for adoption.
CERN’s restored WorldWideWeb emulator lets modern users explore the early interface, but restoration is an interpretation running inside contemporary browser technology. The project’s source and documentation are stronger evidence for exact capabilities and dates than screenshots without provenance.
The first browser-editor is historically important because it reveals the Web’s original social model: following a link and contributing information were adjacent actions. Mass adoption favored universal reading first. Today’s collaborative editors, wikis, and content systems recover portions of the writeable Web through layers the 1990 prototype did not yet need.
Related:
- Public-Key Cryptography: From Classified Prehistory to Diffie-Hellman and RSA
- Unix at Bell Labs: From a Rejected Timesharing Project to a Portable Operating System
Sources: