The History of Collaboration Tools

Turn this article into takeaways for your work.

Each assistant summarizes the article only for you and suggests best practices for your work.

Picture work that involved more than one person before any of this. It ran on carried paper and on meetings. It ran on one person remembering what the others had said.

Collaboration tools are software that lets several people work on the same thing. They began on December 9, 1968, when Douglas Engelbart's team at SRI showed a live computer system built for shared work. Networked email followed in late 1971. A research field was named in 1984, and Lotus Notes sold the idea to companies in 1989. Every tool in this story tries to make the shared record visible to everyone who needs it, when they need it. The throughline is that each generation made sharing cheaper and created a new problem: too many places where the work lives.

Timeline at a Glance

Year Milestone Why It Mattered
About 506-497 BCE Administrators at Persepolis record rations, travel passports and dispatches on clay tablets A precursor: a shared written record of who owes what to whom
October 1962 Engelbart's Augmenting Human Intellect report Sets the goal of tools that raise a person's problem-solving ability
August 1965 MAIL added to MIT's CTSS Messages between users of one shared computer
December 9, 1968 Engelbart's NLS demonstration in San Francisco Live editing, hypertext and a video conference with colleagues at SRI
Late 1971 Tomlinson sends mail over the ARPANET Messages between people on different computers
June 1973 RFC 524, a proposed mail protocol from SRI-ARC Early thinking about distribution lists, acknowledgments and replies
Around 1984 Greif and Cashman's workshop coins "computer-supported cooperative work" Collaboration becomes a field of study
1989 Lotus Notes is released Groupware reaches large enterprises
1995 Ward Cunningham creates the wiki Any reader can edit a page
2006 Writely becomes Google Docs after joining Google A document lives in a browser
April 15, 2010 Google publishes "A rebuilt, more real time Google documents" Editors begin to see each other work
Late 2013 Slack launches publicly, according to TechCrunch Team chat replaces part of email
March 2020 Zoom reports more than 200 million daily meeting participants Video becomes the default meeting room

The Paper Precursors (c. 500 BCE to the 1950s)

Shared records, kept on clay and paper, before any screen

Start in a storeroom in Persia. The Persepolis Fortification Tablets, as Livius describes them, run to between 20,000 and 25,000 tablets, written between 506 and 497 BCE, mostly in Elamite, the language of the Persian chancellery. They record economic transactions "in kind," such as a receipt for 130 liters of barley, and also cover passports, the dispatch of accountants, caravans and teams of laborers. A clerk in one place wrote down what another place was owed or owed out, and the tablet traveled or stayed as the proof.

That was a precursor, not a collaboration tool. Nobody edited the same tablet at the same time, and the record served an administration, not a team. But it solves the same problem every later tool does: two people in different places need to agree on what was promised, and a written record beats memory. For the next 2,400 years, shared work meant paper in some form, moved by hand or by post. The postal side of that story is in our history of email, and the growth of the people who did this work is in our history of knowledge work.

The new problem arrived with the computer. Early machines were shared by many users but operated by few, and the people who used them had no easy way to talk to each other through them.

What it teaches teams today: a shared record is older than any software. If your team can't say where the single record of a decision lives, a new app won't fix it.

The Engelbart Years (1962-1971)

A lab imagines the computer as a team room, and email appears

In October 1962 Douglas Engelbart, then at SRI, wrote a report titled Augmenting Human Intellect. It defined the goal as "increasing the capability of a man to approach a complex problem situation". The report focused mainly on the individual, though it also noted human cooperation made possible by speech and writing. Collaboration was the next step, and his lab spent the 1960s building toward it.

Meanwhile, ordinary messaging started small. Tom Van Vleck and Noel Morris wrote a MAIL command for MIT's CTSS time-sharing system, and, per Van Vleck's history, it was added in August 1965. It only reached users of the same machine, but it taught a generation of researchers to leave messages for each other instead of waiting at the terminal.

Then came the stage. On December 9, 1968, at the Fall Joint Computer Conference in San Francisco, Engelbart presented to an audience of about 2,000 for roughly 90 minutes, according to the Engelbart Institute, while members of his lab teleconferenced in from SRI to demonstrate features of the system. The institute's account of the session lists a mouse, hypertext links, switchable views of the same information and a video link to colleagues in Menlo Park, including a live software tour by Jeff Rulifson and a video conference appearance by Bill Paxton. It also notes that the ARPA Network was due online the following year.

In late 1971, Ray Tomlinson of BBN modified the TENEX mail program SNDMSG to send messages over the ARPANET to users on other TENEX systems, and chose the @ sign to separate the user name from the host name. The feature shipped in the next TENEX release in early 1972. Messages now crossed machines, and the network gained its first widely used application.

The problem this created was scale. Email gave everyone a direct line to everyone else, and the shared record became a pile of private inboxes.

What it teaches teams today: Engelbart's demo showed shared screens, links and video before most offices had a terminal. The tools were early; the questions, such as who sees what and when, were already the right ones.

The Groupware Years (1973-1994)

Researchers name the field, and Notes sells it to companies

Even in 1973 people were trying to organize the flood. RFC 524, written by J. E. White of SRI-ARC in June of that year, proposed a mail protocol with access lists, distribution lists, acknowledgments and serial numbers for linking replies. It was a draft for comment rather than a standard, but it shows how early the group-mail problem was recognized.

The field itself got a name around 1984. Jonathan Grudin's account, published in IEEE Computer in 1994, says that about ten years earlier Irene Greif of MIT and Paul Cashman of Digital Equipment organized a workshop of twenty people from different fields, and that the workshop coined the phrase "computer-supported cooperative work," soon shortened to CSCW. Other accounts name different venues for the workshop, so the year is firmer than the place. The point is that designers, social scientists and engineers now sat in one room to ask why groupware so often failed.

The commercial answer came from Ray Ozzie. According to the Computer History Museum's oral history, he had been inspired by online collaboration on PLATO, the education system created at the University of Illinois in 1960. After building Symphony for Lotus, he got funding from Lotus cofounder Mitch Kapor ($1.3 million) for Iris Associates, which built Lotus Notes. Released in 1989, it offered email, discussion groups, calendars, chat and to-do lists, and became the defining groupware product in large enterprises. IBM later absorbed Lotus, and by the museum's account pushed Notes to 125 million users worldwide. Our history of office software places Notes beside the suites it competed with.

The new problem was administration. Notes needed specialists, and shared databases had to be designed before anyone could use them. The next era would remove that barrier by letting anyone edit a page in a browser.

What it teaches teams today: a tool that needs a specialist to set up each use will stall. Check how many people can start a shared workspace without help.

The Wiki and Web Turn (1995-2009)

Anyone can edit the page

Ward Cunningham built a website whose pages readers could change themselves. In correspondence on the etymology of wiki, he confirmed 1995 as the year he created the technology and said that he had written similar programs earlier, but they did not run on the web and were not called wiki. He took the name from "the wiki wiki bus" he met on a trip to Hawai'i, where the airport shuttle bore that name, and the word means quick. The idea was a page anyone could fix, with no approval step. For the web that carried it, see our history of the World Wide Web.

Documents moved into the browser next. Google says Writely launched in 2005, became part of Google a year later, and was officially launched as Docs in 2006. For the first time a colleague could open the same file with nothing to install. The servers that made this affordable are covered in our history of cloud computing, and the hardware before it in the personal computer story.

The new problem was versions. Documents lived in the browser, but chat, tasks and files still lived in many other places, and people copied between them.

What it teaches teams today: the wiki and the browser document both won by removing a gatekeeper. Look for the step in your own process where only one person can publish or edit.

The Real-Time Years (2010-2019)

Everyone sees the cursor, and chat becomes the workplace

In April 2010 Google published a post titled "A rebuilt, more real time Google documents", dated April 15. Google's own anniversary post says the 2010 update added the ability to see others editing and writing in shared documents. The shared file had become a shared moment. Engelbart's 1968 idea of a screen several people could watch together had reached anyone with a browser.

Chat followed a stranger route. According to TechCrunch's account, Stewart Butterfield founded Tiny Speck in 2009 to build an online game, Glitch. It was released in September 2011 and closed about 18 months after launch. The company's engineers had built a real-time messaging tool for their own distributed team, and when Glitch failed, Butterfield told investors the firm would become an enterprise software company called Slack. TechCrunch dates Slack's public launch to the end of 2013 and says that within months it claimed 60,000 daily users and 15,000 paid seats.

The new problem was noise. Email had been a pile of inboxes; chat was a stream that never stopped. Our article on async communication best practices deals with the habits that came out of that, and our team collaboration tools ranking compares what is available now.

What it teaches teams today: real-time tools raise expectations of instant answers. Agree on which channels need a reply within the hour and which can wait a day.

The Video Room (2020-present)

The meeting room moves online, and an assistant joins it

In its April 1, 2020 post, Zoom said the maximum number of daily meeting participants, free and paid, was approximately 10 million at the end of December 2019, and that it had passed 200 million in March 2020. Video calls were not new, since Engelbart's team had run one in 1968, but now whole workforces were using them at once. Our history of remote work tells the wider story of that shift.

The most recent turn is the assistant that sits inside the meeting, the document and the thread, summarizing and drafting. That belongs to the history of AI productivity tools. The problem it inherits is the oldest one on this timeline: someone still has to own the record.

What it teaches teams today: a tool can carry a conversation, but only a person can own the decision. End each meeting or thread by writing down who decides what, by when.

Key People

Name Role Contribution Year
Douglas Engelbart Researcher, SRI Augmenting Human Intellect report; the 1968 NLS demonstration 1962, 1968
Tom Van Vleck and Noel Morris MIT programmers MAIL command on CTSS 1965
Ray Tomlinson Engineer, BBN Mail over the ARPANET; chose the @ sign 1971
J. E. White SRI-ARC Author of RFC 524, a proposed mail protocol 1973
Irene Greif and Paul Cashman MIT and Digital Equipment Organized the workshop that coined CSCW About 1984
Ray Ozzie Software architect Built Lotus Notes at Iris Associates 1989
Ward Cunningham Software engineer Created the wiki 1995
Stewart Butterfield Founder, Tiny Speck Turned an internal messaging tool into Slack 2012-2013

Key Facts

Key Facts: Collaboration Tools History

The Pattern Behind the History

We call this the Shared Record Pattern. It is our own synthesis of this timeline, not a published model.

  1. A shared record comes first. Persepolis clerks, CTSS users and Notes databases all began by agreeing on one place to look.
  2. Messaging gets cheaper, so the record fragments. Email and then chat gave everyone a direct line and scattered the decisions.
  3. A gatekeeper falls. Wikis and browser documents let anyone edit without a specialist.
  4. Presence arrives. Seeing others work, in 1968 on a screen and in 2010 in a document, became ordinary.
  5. Ownership stays human. Every era still needs someone to say which version counts.

What It Teaches Teams Today

  • Name the single record. For each kind of decision, say where the final version lives.
  • Lower the cost of starting. Tools that anyone can open without help spread; tools that need a setup specialist stall.
  • Set response expectations by channel. Real-time tools make every message look urgent unless the team agrees otherwise.
  • Practice the human skills. Software only carries the work; our article on collaboration skills covers what teams still have to do themselves.
  • Place this in context. The long arc is in our history of productivity.

Collaboration tools have always sold the same promise: everyone sees the same work at the same time. Teams that decide where the record lives, and who owns it, get the benefit whichever tool is fashionable.

About the author

Victor Hoang

Victor Hoang

Co-Founder, Rework.com

Victor Hoang is Co-Founder and CMO of Rework. He spent 12+ years scaling B2B SaaS growth, building a lead engine that generated over 1 million leads and $10M+ in annual recurring revenue. Today he builds AI agents and MCP servers into Rework's products to empower customers across growth and operations. He writes about what actually works.