Computers are large, expensive and slow. They're used for engineering or science and at this point it's more likely to have one computer serving multiple people than having multiple computers. This is a process called time-sharing and it means that computing resources can be utilised by people with their own terminals as if they did have their own computer.

However groups did build their own networks to facilitate communication. Since they were adhoc with their own system of handling data there had to be a common protocol for intercommunication between the different networks. This internetwork came to be shortened as the internet.

One issue in networking is reliability. This means that when you send a message you expect it to actually be sent and not randomly get lost which can happen if you just normally send requests. Previously the primary way to prevent this was to have the network guarantee reliability by keeping track of messages and making sure they arrive as intended. TCP works by instead having the sender and receiver send and track messages to ensure that messages keep to a few principles:

• Reliable - messages get to where they should
• Ordered - messages arrive in the correct order
• Error-checked - if there are slight discrepancies in the messages they will be found

It also implements congestion controls so that hosts shouldn't use the whole network bandwidth. ^[0]

People aren't going to remember addresses like 192.0.33.8 so instead we want a system for assigning names to these IP addresses. There used to be a method of just having someone update the hosts file with a list of all the domains and you would call them by telephone to get them to update it. However this is obviously impractical so a new system based on distributed servers that map the domain names was created. ^[1]

Managing information was a problem Tim Berners-Lee was thinking about while working at CERN. ^[2] The proposal introduces a bunch of historic things such as noting that the organisation is an interconnected "web". It details why hypertext is a useful application to the information management problem. It uses the term "browser" for a program that accesses the server. Most of the principles as detailed in the document are still applicable to the web today.

It's interesting to think of how things might have gone differently. Alternative histories are difficult to imagine since we have so much hindsight of seeing the ways things did transpire and there's so much chaos in the world that small changes could have caused differences in many ways. No doubt something would have formed if it wasn't the world wide web. However from the proposal:

There will always be a large number of information management systems... However, we will lose out if we try to constrain them, as we will exclude systems and hamper the evolution of hypertext in general.

I do wonder if the web has essentially monopolised the system of the internet. It has been extended to fit every purpose; maybe we would be better off having different systems for different purposes?

Even though they were coming out at around the same time, the Gopher protocol is seen as a pre-web internet system. Looking at it now, it looks archaic but so did HTTP ^[3] at one point.

Gopher servers still exist today, although there aren't many of them and the protocol hasn't changed much. There's also alternative protocols such as Gemini

Although there were previous browsers, the Netscape Navigator browser is one that grows with and grows the internet itself. Netscape was created by Marc Andreessen ^[4] who had worked on the Mosaic browser at NCSA. During development the Netscape browser was known as Mozilla, a portmanteau of Mosaic killer but that's a teaser for the future. Mosaic had introduced advancements to the web basically single-handedly such as the img tag for displaying images on the web.

The people at Netscape realise that Microsoft are going to ship some kind of VBScript in their Internet Explorer browser and also realise that a scripting language of their own would be a really powerful feature that would be compelling for developers and users. Unfortunately the language they decided on wasn't entirely driven by practical considerations but also commercial ones as Netscape had started a deal with Sun Microsystems meaning that they wanted a language similar to Sun's Java but more suited to the web and thus mostly from Brendon Eich, JavaScript was born.

Flash forward 30 years and JavaScript is one of the most popular programming languages in the world. It's used in many, many places far beyond the browser. It powers applicaions, servers and more. It is extended to support far more than just the small scripting it was initially intended for; people even create other programming languages that compile to JavaScript so that they can run on the web.

One advantage of HTML is that it declares the meaning rather than just pixels on the screen which means that users can display the content in whatever way they want, indeed one of the original browsers was a line mode browser and was very simple. You can view a simulation of it by CERN here

However, people want more than just text and images clumsily laid out next to each other without any colours. Therefore there needed to be a way for a developer to declare the way they want their page to look. If the user wanted to have their own styling then they should be able to as well. The language for achieving this was CSS or cascading style sheets. It's cascading because the styles cascade down the document and is the language for laying out and styling elements.

XML had been the standard way to structure data, it was what HTML was based on and seemed to be everywhere. However it did have some flaws. One was that it was too complex, another that it was too verbose so Douglas Crockford created a minimal format based on Javascript called JSON or JavaScript Object Notation.

Sometimes its minimalism hurts it in some departments. For example when it's used as a configuration language it really should have comments and it often allows trailing commas. But as a simple, human-readable data format it works great, especially since it's easy to use with Javascript, the programming language of the web.

The W3C (World Wide Web Consortium) was the custodians of the web standards but the browser owners, especially Opera and Mozilla were frustrated by slow progress. Thus they formed their own group with an even worse acronym, WHATWG (Web Hypertext Application Technology Working Group) which would push the development of HTML forward themselves. This shows that while web standards may be important; ultimately what's actually implemented by the popular browsers is what matters. Furthermore, control over the browser market share is equivalent to control over the direction of the web.

Apple announced the iPhone in 2007 and it quickly became huge, revolutionising communications, games, apps, streaming; having enumerable services rely on them. It changed the web as it became commonplace to design websites around the smaller screen sizes, having to lay things out in a different manner.

The browser wars were back and this time it was Google wanting to take over the browser market. The benefits of this to them are enormous since they can serve Google as the default search engine and also push the web towards a direction more amenable to their interests which are many since they want to make and support complex web applications such as Google Docs/Excel, whereas someone like Microsoft might prefer to keep Microsoft Office/Excel the main application people use.

With Google Chrome, they focus heavily on performance, keeping memory usage low. They make Javascript run faster with the V8 Javascript engine. With all these features as well as heavy promotion on their dominant search engine, Chrome comes to be the biggest browser in the space.

Flash was added as plugins to the browser and during the early 2000s was a pivotal part of the web, especially as part of games, animations and applications. However just as Java applets were once killed off, so was Adobe Flash. Probably the key point was Apple refusing to have Flash on their iPhones for a few reasons explained in this document written by Steve Jobs in 2010 Ultimately it was killed off but you can still view old Flash content through emulators such as Ruffle

This saga shows how even though backwards compatibility is one of the web's principles, it doesn't come above some of the others. The security issues and propriety nature meant that Flash was doomed even though it did help to define what web applications could be.