Thursday, March 29, 2007

Magnolia - a not-so-brief history

The history of Magnolia really started out in the Summer of 2000. It was internet boom time even in Switzerland, and thus is came that one night, Pascal Mangold, C.E.O. and Founder of Mangoldpartner, and Boris Kraft, C.E.O. and Founder of Kraft & Partner met somewhat unexpectedly in Zurich. They had been invited by another company that wanted to gobble up the two upstarts for their specific know-how with respect to WebObjects.

It turned out that neither Pascal nor Boris were convinced that the concepts presented by the inviting company would lead anywhere. But the evening was not spent in vain: they immediately realized that they would want to work together.

In the following weeks they spent many hours drafting a concept for a new content management system code-named Gamma4 (unfortunately the draft has since been lost) and offered its implementation to a customer of Kraft & Partner for a price tag of two million CHF - a sum that was hard to pronounce for either of them at the time. The offer was rejected, but as so often, this probably helped more than hindered the course of things to come.

A couple of months later, Winter 2000 was approaching, Boris and Pascal decided to merge their companies, and ultimately created obinary Ltd., a company with three parters, and three lines of business. Michael Robertson was responsible for network and infrastructure, Boris Kraft (C.I.O.) for software development, and Pascal Mangold (C.E.O) for Internet projects.

In the coming years obinary realized many internet projects, and started using a commercial content management system called Soloweb by an Icelandic company. The merit of the product was its technological base: WebObjects. This made it a natural fit for obinary, who had been using WebObjects to create customer applications.

At the height of this time, obinary had grown to about 15 employees and revenue of 2 Million CHF. But times started to change. We heard the feedback from our customers regarding usability and realized that things can never be simple enough if it comes to creating user interfaces. We also realized that customization was a big part of any CMS implementation, and that it was getting harder to sell licenses to our customers due to the increased price pressure coming from open source alternatives.

Early 2003 obinary researched open-source content management systems based on Java, trying to benefit from "free software". They realized that of the very few java-based systems out there, none came anywhere near to what they were looking for in terms of usability, ease of customization and ease of installation. It even happened that Michael Wechner presented Lenya to their unbelieving faces (to paraphrase The Matrix: "There is no GUI"). Michael is a great guy, but Lenya was approaching obinary's needs from the wrong end. They were surprised of how much existing systems ignored the needs of their users.

In 2003 obinary started hiring people off of their competition, who was spearheading a new standard that later come to be known as the Java Content Repository (JSR-170). Sameer Charles joined obinary and convinced the management team that they should develop a custom content management system instead of trying to hack around a poorly documented proprietary system that was hard to sell, hard to customize and hard to use. Coming from the company that was busy defining JSR-170, Sameer suggested to build Magnolia on JSR-170, long before anybody else even heard of it - and long before there was any indication that betting on it would pay off.

In summer obinary hired Marcel Salathé to work on the GUI of Magnolia and subsequently released Magnolia 1.0 on November 15, 2003 (by no coincidence the birthday of Boris' brother).

While some people at obinary had doubts regarding the public interest in Magnolia, the first days of releasing Magnolia overwhelmed obinary completely. It was an amazing time (see for some impressions).

So, why the name Magnolia, and why open-source? At the time, obinary already had launched a product called "obinary compass", a tool that later became Magnolia for Business Processes. Our original idea was to keep that naming convention and name our newborn CMS "obinary content". However, Boris could not agree to the name and in the subsequent discussion pulled the name "Magnolia" out of thin air. The name stuck after some initial opposition, and today is considered to be a brilliant name and major asset.

Open-sourcing Magnolia was a very pragmatic choice. Obinary was a small company with exactly zero marketing budget, and the competition seemed overwhelming. Obinary did not see a chance to place another commercial CMS in the market, nor was it their goal to sell software. Their goal was to sell services, as this is what they had done all along.

The Commercialization of Magnolia

Initially, obinary was happy that they could use their own Magnolia to implement projects. It also helped them pitching for services when they were able to tell potential customers how many people had downloaded Magnolia, and that it was used around the world.

Releasing Magnolia into the wild resulted in considerable amounts of unpaid work for obinary - answering emails on the user-list and on info@obinary, building a community, maintaining infrastructure all added up.

At the time, conventional wisdom regarding open-source software was that one could live off support, training, consulting, partnerships etc. Nobody told them that establishing these offerings needs considerable resources. Obinary found a great training partner in the USA and developed a five (later four) day developer training. They defined partnership options and contracts. They defined and communicated their support offerings.

But unfortunately it did not work as expected – at least not fast enough to make a difference to their bottom line. They needed to find a maintainable way of moving forward with Magnolia.

The interest in Magnolia clearly showed them that they were on to something - the benefits of making things simple were starting to be obvious to more and more people. The plan was born to built the best CMS in the world, and sell related products - for instance a yet to be implemented Document Management System. This way the visibility of Magnolia could be used to generate leads for their commercial products.

Magnolia For Documents was released October 12th 2005, delayed by many projects obinary did to finance Magnolia. The need for a java based Document Management was obvious at the time – too obvious, as it turned out. Due to their delays, a new open-source startup had grabbed their niche, and obinary's DMS offering was a commercial failure. While they did sell a few implementations, they had to change their strategy - again.

By the end of 2005, the outlook for obinary was bleak. They had invested in "obinary Compass", then open-source Magnolia, then in "Magnolia for Documents", with little return on investment. They had shed off all employees except for the bare minimum to survive - Pascal Mangold (CEO), Boris Kraft (CIO), Sameer Charles (Magnolia Core Dev), Philipp Bracher (Magnolia core dev) were the only remaining employees by the end of 2005. Obinary's third founding partner, Michael Robertson, jumped ship and founded a new company focused on network and infrastructure services.

At that time, Pascal and Boris had to make a crucial decision: dump obinary or struggle through. The easy way would have been to shut down obinary - everybody of the renaming team was highly qualified and would have had no problem to get well-payed jobs elsewhere. But the company's founders were confident that their investment would eventually pay off - the feedback from Magnolia's users was too encouraging to simply give upon it.

They just needed to find a way to "cash in" on their product's success. The new strategy was to create an open source Community Edition and a commercial Enterprise Edition. They added the previously separate Document Management functionality to the open-source Magnolia offering.

The renaming of obinary to "Magnolia International" mid 2006 was a logical next step. By 2006, Magnolia dominated their lives. Obinary as they had founded it no longer existed - the company had been transformed from a systems integrator to a software producer. "Magnolia International" would be the company focused on Magnolia, and solely on Magnolia.

Magnolia 3.0 was released November 15th, 2006, three years after the first Magnolia release. It was the first commercial open-source software that seamlessly integrated Content Management and Document Management. They had beaten their competition and started to gain significant traction.

The new business concept was easy to communicate. Additional functionality and unlimited support availability quickly established the Enterprise Edition as the way to go for many enterprise customers. By Spring 2007, Magnolia International Ltd. had gained significant numbers of Enterprise customers, had grown to seven full-time employees and enjoyed a splendid outlook of significant growth.


In Spring 2007, Magnolia has received a WebDev award as one of the ten best Content Management Systems 2006.

The company that originally wanted to buy Mangoldpartner and Kraft & Partner went bust in 2001.

It would take Pascal and Boris nearly seven years to make another two Million CHF offer - this time, however, for a system they already had deployed on thousands of systems around the globe - from small semi-static web sites to some of the most demanding sites on the web.

In August 2007, Magnolia founded Magnolia Americas, Inc., a new venture to cater to the needs of the US market. Magnolia has grown to 11 emplyoees and two staffed locations - Basel, Switzerland, and New York City, USA.

Sunday, March 11, 2007

You, inc.

Yesterday in our local bookstore, I stumbled across "You, Inc.: The Art of Selling Yourself" by Harry Beckwith and Christine Clifford Beckwith. I immediately liked it (for its title, its format and quality of paper!) and hence, I bought it.

Its been a while since I last read a book spanning more than 300 pages on a single weekend (especially on a nice, warm and sunny weekend) - this book is wonderful. Everyone should own it, and read it as often as possible. Go and do yourself a favor, buy it. Now.

"Selling yourself" is what you do all your life - from early childhood to old age. You might as well read up on it. The book is full of wisdom, in terse and fun-to-read style, not one of these books that are abundand with motivational stories and little else.

If you read just one book this year, read this one.

Friday, March 09, 2007

Magnolia awarded one of the best CMS of 2006

Magnolia is nearly four years in the making and we finally won our first award. I don't know if this award has any relevance, but the company behind the award claims they have about 1 Million visitors per month, so thats not too shabby. They nominated Magnolia one of the 10 best Content Management Systems in 2006. Hooray!

Magnolia and Scalability

Matt recently asked, how scalable Magnolia is (and other CMS).

First of all you need to ask yourself, what do you need to scale? Number of authors? Amount of Content? Requests served? DIfferent questions, different ways of doing it.

There are many ways to scale Magnolia - one thing is the built-in subscriber mechanism, so its straightforward to have lets say a HW load balancer and two (or more) public facing machines to handle the requests. You can write a custom Apache (or other) cache module (has been done before) or write a custom cache handler (cache is pluggable in Magnolia) if what exists with respect to caching is not meeting your requirements. You can cluster the repository (with the upcoming Magnolia & Jackrabbit 1.2).

You can use Akamai for caching (has been done for, a massive global news website). This site, powered by Magnolia, has content in English, French and Arabic, movies, audio, text, images, is updated simultaneously with the TV channels and has handled 50000 (fifty thousand) simultaneous requests when launched.

Last week we launched, the leading travel site in Australia, with booking integration, white-labeling of content, partial content delivery and other extravaganza. Their requirements are also pretty high regarding performance and scalability (they expect to grow).

(For partial content try this:
Using selectors we deliver parts of the page, so that it can easily be reused elsewhere.

In addition, we do have a portlet bridge in the EE that runs nicely with Liferay, in fact we currently do a joint project for a car manufacturer. It works very well, and scales the same way it would with standalone Magnolia, since the same mechanisms are used (each Liferay instance contains a Magnolia subscriber that handles the content).

Magnolia is now in its forth year of development, its pretty mature and we have a lot of experience with what works, how, when and why - or not. If you have questions, feel free to download our Magnolia Enterprise Edition and make use of our free evaluation services to help you get the most out of your time spent on evaluation. (Details come with the download).