The history of web content management can be told in many different ways. I chose to go through my own personal archives and tell it by visiting the various different stages of our product CQ that I have been involved with my entire professional career.

One could say that I am still in my first job after 18 years -- however, I feel like the internet market in general and the web content management market specifically has reinvented itself so many times that it feels more like 4 or 5 different eras, and in that sense, like 4 or 5 different jobs. 

1994: In the sausage factory

Before the WCM market was born we started in a sausage factory on the outskirts of Basel. We were a mixed group of futurists that were adamant about using modern technology (for example, CD-ROMs at the time) to help our customers communicate. We described ourselves as a high-tech communications agency. In the anglophile world that we lived in, we purposely chose a French name for our company to illustrate that we were not about technology, but about using technology as a communication vehicle.

When we realized that there was something more interesting than CD-ROMs, games or screensavers we broke the bank and got ourselves a "leased line" to the "internet" with a whopping bandwidth of 64kbits/s and a really interesting green Silicon Graphics Indy as our web server.

Learning HTML to manage their web sites was not an option for our customers, but getting hundreds of phone calls to make changes to their web sites was not an option for us. So we built something that we called an "admin tool" to allow business users to change web content using simple web forms and to get us out of the boring business of making simple updates for our customers. We didn't know that this would be called a "web content management" system in the future.

1996: Openweb is born

A couple of years later we realized that our "admin tool" was a unique selling proposition for our web design business -- "if you let us design your website, you will be able to update it yourself". Since it could be sold as a stand-alone product, we thought it at least deserved a proper name, so we called it "openweb".

openweb1.jpg

"Openweb" was written in C as a webserver plugin (NSAPI initially) and fit onto a 1.44m floppy disk.

One of our early customers came back from a conference in the U.S. and told us that there was a market in the U.S. called "web content management (WCM)" which was very well aligned with what we were doing. We heard distant rumblings of Vignette in the U.S.; however, the market was dominated by a large number of local players in every single European country -- many of them have since vanished.

1999: CQ2 takes the market with bakers and fryers

In 1999, our WCM was still in C, still fit onto a 1.44mb floppy drive, featuring our own ECMAScript -- aka server sided JavaScript (huh?, seems really fashionable again…) interpreter, and the server started up in less than a second on the hardware back then. We realized that the software business we gradually entered was a global market. We got more and more competition from the U.S. and decided that we wanted to enter the U.S. market and rename our company to "Day Software" as our French name really was not going to be a great asset in the U.S. market. At that time, we also decided that instead of a French company name, we'd rather have a French product name so we called our "communiqué" CQ and since nobody wants a 1.0 product we started our numbering scheme at CQ2.

At the same time, the WCM market started to shape up, there were primarily two different models of understanding WCM, known as "baking" and "frying". Both of those were represented and marketed by individual U.S. companies. The chief baker was Interwoven and the chief fryer was Vignette. While baking as represented by Interwoven was an automation to produce static html more efficiently with solid versioning and different tools, frying -- as proposed by the Vignette feature stack -- looked like an early stage Tcl-based application server backed by a relational database and some reverse proxy caching.

The market fragmentation was increasing and it seemed like every single week there was a new WCM vendor added to the space. Switzerland, with a very small domestic market, maxed out at a good dozen domestic vendors.

2002: CQ3 -- pure Java

After going public in 2001 and a peak at the Internet bubble, we had a crazy ride where our stock price took a plunge from chf780 down to 2.9chf. We were in the middle of a turnaround.

purejava.jpg

Under the influence of our enterprise customer base and its growing affection with Java and J2EE we looked at rewriting our entire stack on a Java/J2EE platform. Our natural Swiss conservative and distrusting nature protected us from swallowing the J2EE bait and protected us from leveraging pretty much anything else from the J2EE stack other than the servlet/JSP interface and allowed us to remain portable between application servers as well as building our own container.

Other vendors got in with J2EE so deep that their market dominance was shattered by their leap into a full product rewrite in Java without offering their customer base an upgrade path and without offering support for multiple application servers.

This was the high time of J2EE in the enterprise. CIOs spent millions of dollars buying app server infrastructure and it became mandatory to be able to run a WCM on "whatever our CIO just bought".

Rewritten from the ground up for the enterprise market with a heavy focus on integration, as deemed appropriate for "enterprise software", CQ3 no longer fit on to a floppy disc and seemed to take ages to start up.

ents.jpg

2005: CQ4 with JCR inside

An effort that started in 2001 called JSR-170 or JCR (content repository for Java technology api) came to fruition and we managed to use the Java community to get everybody who had a vested interest in content (with the notable exception of Microsoft who was in a lawsuit with Sun over Java) to agree on a standard specification. This was an evolutionary step in the broader content management industry. The breakthrough was not so much the 300 pages of fine specification (if I do say so myself) but the fact that we had a platform for the industry to come together on a technical level and discuss matters of interest for the content management industry.

Together with the release of the JCR specification in 2005 we released our fully compliant product, in the ill-guided belief that it matters to be "first" at something. So CQ4 represented "the first WCM" that was based on a fully standards compliant repository but, more importantly, that meant that we completely rewrote the infrastructure foundation of our product stack.

As a side effect of starting the specification process of JCR, we were able to get Roy Fielding interested in our company. He joined us in 2001. Roy had a material impact on the architectural style of our entire product stack as well as on the web as a whole and, even more fundamentally, on our open development paradigm. The SOAP::WSDL hype came and stayed for a few years.

2008: CQ5 complete revamp

The WCM market consolidation seemed to move forward as many bigger WCM companies were acquired by big software portfolio companies. At the same time, it felt like new WCM systems (either open source or closed source) started to emerge. It seemed that the ECM bubble that wanted to swallow WCM had finally given up and Gartner brought back the WCM magic quadrant.

Right after releasing CQ4, we realized that as we standardized and modernized the infrastructure foundation, it was now finally time to work on the front-end authoring experience. At this point, we still had front-end code written for a Netscape 3 browser. So we decided to rebuild the entire front-end experience from scratch using modern (2008) browser techniques (Ajax was one of the major buzzwords), which turned out to be a major effort but really put our user experience far ahead of what customers were used to in comparable products.

In a sense we "arrived" from a product perspective with a clean, overhauled architecture and a powerful, beautiful user experience. So what was left? Well, we were still a small Swiss vendor in a rapidly changing market.

At this time, we were very lucky to find an awesome management team and put together a great sales team to put us on the map and take us to the next level from a business perspective.

hist.jpg

2012: CQ5.5 solid architecture brings a lot of features

Based on the success of the CQ5 platform, we were acquired by Adobe in 2010, which turned out to be a very successful combination not just from a business perspective but also as a technology symbiosis.

I never felt like I had to "let go" of our "baby", even after such a long time. It feels like CQ, Day and the WCM market as a whole have grown up and have found their destinies as an integral part of the broader digital marketing space. All the fun that we had building the world’s best research and development team is just amplified and applied to a broader space.

We live in a time where social, mobile and cloud are three mega trends that are coming together at the same time. There has never been a more exciting time in technology -- there has never been more room for innovation, and the web (and with that web experience management) is at the heart of all these developments.

Web content management has a vibrant past history, though it feels like nothing is settled. With the massive paradigm shift impacting mobile, social and cloud we will see an even more interesting next 15 years of web content management evolution.

Note: This article was originally published on the Adobe PEEK blog here.

Editor's Note: Watch for David’s second post on the "The Future of WCM"