Bruno Lowagie: I asked what we can do to solve this. IBM said, “We have to do it the hard way.” Actuate was a company under the Eclipse umbrella. They entered a research agreement with Gent University. The deliverable was an IP overview.
During 2007, I went to the university to do my job but as soon as I received a report from IBM Canada with some issues, I had to drop everything and fix the issue. Every week, I would get an Excel file with potential issues. For instance, I used some code that was released under a certain license. I was allowed to use it but inside the code, there were some remnants of a project that had started as a close source and was suddenly brought to the open source. But then the source code is set as proprietary to some. We had two conflicting licenses. The lawyers said, “Better be safe than sorry.”I had to refactor everything. I had to change everything that didn’t belong in iText. After one year, all the source was accounted for. I went back to find everyone who had ever contributed code and if it was less than 20 lines, the transfer of IP was implied. If it was more than 20 lines, then we had to ask people to sign a contributor license agreement. We threw away a lot of functionality but once this job was done, we were 100% sure of the IP of iText.
It turned out that Gent University didn’t have any IP in it. We founded our company in 2008. We transferred all of the IP into the company. That was in January of 2008. The company was called IT3XT, which is l33t speak for iText. Unfortunately, nobody knew how to pronounce the name, so eventually we changed the name to iText Group.
We’ve been giving away iText for many years. In order to continue the project, we needed some revenue. iText was taking more and more of my time and I couldn’t do my day job.
Sramana Mitra: Did you quit your job?
Bruno Lowagie: In January 2008, the company was founded. One month after, our son was diagnosed with cancer.
Sramana Mitra: What kind of cancer?
Bruno Lowagie: Bone cancer. He has prosthesis on his left leg. During that year, we were spending more time in the hospital than at work. I almost stopped working on iText. During that period, I received a mail from somebody with a colruyt.be address. As I knew that this was a Belgian company and that most of the developers at Colruyt were Flemish, I replied in Dutch. However, the person said he didn’t understand my answer. It turned out that he was from India!
He had a problem converting invalid TIFF files to PDF. In current iText versions, we are more tolerant towards errors in TIFF files because most image viewers show them anyway, but in those days, we rejected faulty TIFF files. I explained that I didn’t have the time to help him because of my personal situation with my son having cancer. I don’t think he believed me because he got very rude, sending me multiple emails a day demanding that I help him meet a deadline.
Because of this behavior, I searched the web for whoever might be in charge of that project. I mailed the CIO at Colruyt and he explained that the context of this question was indeed a company project. The first idea was to have this project implemented by Dolmen, which is a professional services company owned by Colruyt. As it turned out, the project would have been too expensive if it was done by this subsidiary of Colruyt, so they had decided to outsource the project.
My wife and I were taking shifts at the hospital so that one of us would be with our son at all times. It was my turn to go to the hospital to do my shift when I received this mail from Colruyt’s CIO, but I didn’t have any energy left. I just started crying and I couldn’t stop. My wife phoned me from the hospital asking me why I didn’t show up, and I explained the whole situation. She phoned a doctor to do a house call in order to reboot me. The doctor told me that our son, not iText, was our priority now, and that helped me focus.
This was one of the key events in the lifecycle of iText. I finally understood that open source can lead to a situation that is totally unfair. A Belgian company decided that their own developers were too expensive, so they paid an offshore company to do the project, but in the end, I was supposed to do the work without being paid. This couldn’t be right. Something had to change.
After 10 months and 18 sessions of chemotherapy and several operations, our son was able to return home. At that moment, in December 2008, my wife and I had to make a decision: do we stop iText altogether, or do we take the leap forward?
We wouldn’t be here if we had decided to give up. In February 2009, an American friend, Andrew Binstock, helped us by founding iText Software Corp. in the US. We didn’t have any employees yet, but we worked with sales people on commission. The first year, we had about $300,000 in revenue and about $700,000 in the second year. When we reached the $1 million mark in 2011, we decided to found another iText Software company in Belgium and started hiring employees.
Sramana Mitra: How did you do that? You had a lot of users of iText. Did you give the list to the sales people and have them start calling?
Bruno Lowagie: In the banking and insurance sector, they didn’t want to use open source unless they established some kind of relationship with the company. In 2009, iText was still MPL/LGPL. There wasn’t really a reason to buy a license. Banks and insurance companies have shown the intent to buy licenses anyway. I had this list of potential customers. The sales people who worked on commission would go through the list and that was our first revenue.