The evolution of the web: from the document web to the application-like web

Objectives

By the end of the lesson, you know the main milestones in the origin of the web, you can distinguish the characteristics of the document web, the participatory web, and the application-like web, and you understand why the newest approach is not always the best.

Why is it worth knowing the history?

Many oddities of how the web works today are historical legacy. The structure of the URL (Uniform Resource Locator), the request–response nature of HTTP, and link-based navigation all come from the era when the web was still built to connect documents. Whoever knows this understands better why today's complex applications are also built on the same foundations, and why a new network system does not have to be invented for every new need.

The beginning: linked documents

The web was born at CERN (the European particle physics research institute). In March 1989, Tim Berners-Lee wrote a proposal for a distributed information system that would connect researchers' documents with links. By the end of 1990, the first browser-editor, the first server, and the first web page were ready. The web software was presented in August 1991 on internet newsgroups, and on 30 April 1993 CERN declared it freely usable (public domain). This decision also contributed to making the web a standard that anyone can implement. In October 1994, Berners-Lee founded the W3C (World Wide Web Consortium), one of the central communities for developing web standards.

timeline
    1989 : Proposal for a distributed information system (CERN)
    1990 : First browser-editor, server, and web page
    1991 : Public presentation of the web software
    1993 : The web is placed in the public domain
    1994 : Founding of the W3C
    2004 : Web 2.0 conference

The essence of hypertext is that a link inside a document points to another document or resource. This is also the basic operation of today's web: a click starts the request of an address.

The document web ("Web 1.0")

"Web 1.0" is a retrospective collective term, not an official version. It is customary to call early services this way if they mainly published documents: a company introduction, a university information page, a personal homepage.

• the content was produced by the provider;
• the visitor read, rarely wrote;
• pages changed rarely;
• every click loaded a whole new page;
• the browser was mainly a document viewer.

The participatory web ("Web 2.0")

The term "Web 2.0" was popularized in the mid-2000s by Tim O'Reilly and his team: the first Web 2.0 conference was held in October 2004, and O'Reilly's article published on 30 September 2005 explained in detail what they meant by it. (There are earlier uses of the term too; what matters is how it spread.) It is not a new protocol or the second version of the web, but an approach: the web as a platform, where users create content and the value of the service comes from data and community.

Typical examples: social sites, blogging platforms, video-sharing sites, collaboratively edited encyclopedias. The user is reader, author, and data source at once. Characteristics:

• user-generated content;
• personalized interfaces;
• services built on databases and APIs;
• fast, partial page updates;
• network effects: the more people use it, the more valuable it is.

The application-like web

Today many services behave not as a series of pages but as applications: an online mailbox, map, or word processor continuously reacts to actions, loads and saves data, notifies, and synchronizes across several devices. This is made possible by program code running in the browser, asynchronous (in the background, without reloading the page) data communication through APIs, browser-side data storage, real-time communication, responsive interfaces, and technologies that support offline operation. We deal with these separately in later chapters.

The three forms did not replace each other, they live side by side:

flowchart LR
    D[Document web] --> M[Today's web]
    R[Participatory web] --> M
    A[Application-like web] --> M
    D -. basis of .-> R
    R -. basis of .-> A

The dashed arrows indicate that the later form builds on the earlier one (the same addresses, the same HTTP), and the solid arrows that all three are present today.

The price of progress

More complex applications have a price:

• the amount of data to download and the loading time can be larger;
• more personal data is created and must be protected;
• accessible use is harder to ensure;
• more attack surfaces appear;
• dependence on certain platforms can become stronger.

Service	Typical approach	Rationale
University regulations page	Document-centered web	It is for reading and rarely changes.
News portal	Content-based, partly dynamic web	Frequent updates, search, personalization.
Online mailbox	Application-like web	Continuous interaction, data handling, synchronization.
Social site	Participatory platform web	Users create a significant part of the content.

Worked example: the three faces of Kurzuspont

Kurzuspont can take all three forms, depending on which of its functions we are talking about.

• The regulations and information pages are documents: they change rarely and are made for reading. A simple, fast page is the best choice here.
• The student reviews attached to courses are already a participatory feature: students create the content, so it must be moderated, protected, and managed.
• The enrolment interface is application-like: it gives immediate feedback if a course is full, does not reload the page while saving, and stays consistent across several windows.

So within one system there is no need to choose a single model. Which function gets which approach is a design decision (discussed in detail in chapter 6).

Common misconceptions

Statement	Clarification
"Web 2.0 is a new internet or a new version of the web."	It denotes an approach: user content, platform thinking. The protocols are the same.
"A static web page is obsolete."	In many cases it is a faster, cheaper, and more secure solution.
"Every modern web page is a single-page application (SPA)."	Many modern services use other rendering and navigation models.
"The early web was simple, so it has nothing to do with today's."	The address, the link, and the request–response pattern are still the basis of the web today.

In brief

• The web was born in 1989 as a proposal at CERN, was made freely available in 1993, and the W3C was founded in 1994.
• The document web, the participatory web, and the application-like web exist side by side, they are not eras.
• "Web 1.0" and "Web 2.0" are labels of approach, not version numbers.
• More advanced applications have a price (size, privacy, accessibility, security), so the technology must be chosen for the task.

Review questions

1. What was the original purpose of the web, and where was the first server completed?
2. What does "Web 2.0" mean, and what does it not mean?
3. Which approach suits a university regulations page better, and why?
4. What disadvantages can it have if a simple information page is built as a complex application?

Learning resources

• A short history of the web (CERN) — from the proposal to the public domain decision.
• History of the W3C — the founding of the W3C.