History › Our time
World Wide Web and the internet age
web public 1991 (internet roots: ARPANET 1969)· Invention or discovery· Global
Well documentedSolid records, physical evidence or scientific consensus.
A system of linked pages running on the internet, invented by Tim Berners-Lee at CERN and opened to the public in 1991.
See it on the History map →What happened
The internet grew from ARPANET, a US military-funded network that sent its first message in 1969, and in 1983 adopted TCP/IP, the common rules still used. At CERN, the physics lab near Geneva, Tim Berners-Lee proposed the Web in 1989: pages with addresses (URLs), clickable links and a browser to read them. CERN made it free for anyone in 1993. The Mosaic browser that year made it easy, and users grew from millions to over 5 billion.
Why it matters
It made information nearly free to publish and find, transformed commerce, media and politics, and became the platform for smartphones, social media and AI.
How we know
Berners-Lee's 1989 proposal and the first web server survive at CERN; ARPANET records, the published technical standards, and ITU usage data document its growth.
Still debated
Who should control it (governments, a few giant platforms or users) and how to handle privacy and misinformation remain open fights.
Inside this (2)
Connections (4)
▸Space race spurred the research that led to the World Wide Web and the internet ageMostly agreed
The connection
Sputnik's shock led the US to found ARPA, a Pentagon research agency, in 1958. ARPA funded computer networking and built ARPANET in 1969, the ancestor of the internet. The Web itself was invented later at CERN, a European lab, but it runs on the internet that ARPA's work started.
When and how
ARPA founded February 1958; ARPANET's first message went from UCLA to the Stanford Research Institute on October 29, 1969.
How sure are we?
ARPA's founding records tie it to the post-Sputnik panic; ARPANET's design and funding are well documented.
Debates and open questions
A popular myth says ARPANET was built to survive nuclear war; its builders say it was for sharing scarce computers. The space-race link is indirect.
▸World Wide Web and the internet age accelerated GlobalizationWell documented
The connection
The Web made sending information worldwide nearly free. Companies could manage factories and suppliers across continents, outsource services to places like India, and sell online to any country. Money moved at network speed. Alongside container shipping, it became the infrastructure of modern globalization.
When and how
From the mid-1990s: e-commerce firms such as Amazon (1994) and Alibaba (1999), plus offshore call centers and software work.
How we know
Trade in services and cross-border data flows grew sharply with internet use, as tracked by the WTO and the McKinsey Global Institute.
Debates and open questions
Economists debate whether the internet, or cheaper shipping and lower tariffs, did more to drive trade growth.
▸World Wide Web and the internet age went mobile with Smartphones and social mediaWell documented
The connection
Smartphones put the Web in every pocket. Fast mobile networks (3G, then 4G) and the iPhone's full browser made the mobile internet practical. Today most web traffic comes from phones, and in many countries people skipped home computers entirely, going online first by phone.
When and how
From 2007; 4G networks spread from about 2010; mobile overtook desktop web traffic worldwide around 2016.
How we know
Web traffic data from firms such as StatCounter, and ITU surveys, document the shift to mobile.
Debates and open questions
Some argue apps replaced the open web with closed platforms run by Apple, Google and a few others; how much that matters is debated.
▸World Wide Web and the internet age supplied the training data behind the Artificial intelligence boomWell documented
The connection
Modern AI learns from examples, and the Web created an enormous supply: billions of pages, photos and videos. ImageNet, built from millions of labeled web images, enabled the 2012 deep-learning breakthrough. Large language models such as ChatGPT were trained largely on text gathered from the Web.
When and how
ImageNet, released in 2009 by Fei-Fei Li's team at Princeton and Stanford; the Common Crawl archive of web pages, from 2008, fed many language models.
How we know
AI developers' published papers describe their training data, including Common Crawl and Wikipedia.
Debates and open questions
Authors, artists and publishers have sued over whether training on web content without permission is legal; early court rulings have been mixed.