20 Sep 2020
I'm a professional software developer and architect for more than two decades.
I have done a lot of talks and presentations during the years.
All the time, there was this tention between business-oriented and engineering-oriented approach to presentations.
The commonly-used presentation techniques are usualy business-oriented.
But that thas not work that well for engineering.
Then there are young engineers that often do not get any format training in presentation techniques at all.
However, they have a lot to say to the community.
It is quite obvious, that they would benefit from a good presentaion guide.
Back in 2002, I have created a short presentation guide for our local Linux User Group (SkLUG).
The HOWTO was very useful back then.
I was quite surprised to realize, that it is still useful, almost two decades later.
Therefore I have decided to update the guide.
This somehow got out of control and ended up with (hugely) expanded version.
The result is "How To Make Technology Presentations" guide.
I hope that this will be useful to a broad technology community.
17 Sep 2020
I have spent quite a lot of my time in last 6 months on one interesting project: Data provenance prototype. This was an experimental project aimed at metadata functionality in identity management. There were many interesting results of the project that I have summarized in my Evolveum blog post.
This was also my first project that was mostly funded by European community money. I have to say that it was quite a nice experience overall. There was some paperwork, but not too much. There is kind of tradition in my coutry that the European should be spent instead of used, therefore I did not know what to expect in this project. But this was international NGI project and not an ordinary national project. It was different. We felt a lot of pressure to deliver tangible results, which I undestand and fully support. This atmosphere produced excellent results. We even ended up using our own money on top of EU funding to improve the results, even though we were not expected to. The project was a success on all fronts. I just wish that more EU-funded projects could be like this one.
11 Mar 2020
Open source is not a business model. This may sound like a strange statement from a founder of a successful company which is firmly based on open source. But it is true. Open source is not a business model. But there are many ways how to build a business around open source software.
Evolveum is an open source company which I have co-founded in 2011. We focus our energy to build and maintain midPoint - open source identity management and governance system. Open source approach is embedded deep in the fabric of our company. MidPoint was an open source product since the day one. MidPoint is pure open source, without any paid features, enterprise editions or any other limitation. Everything is open. Despite that, we are profitable and midPoint is stronger than ever.
But making that happen was no walk in a rose garden. The beginning was very hard. The technology was not the problem. There were good engineers in Evolveum team even in the early days. We were used to deal with problems of technology. That part went quite well. The business side was the problem. How can one make a living by developing an open source software? We had some idea, but it was not working as well as we expected. We were struggling, experimenting, trying and failing. But after all those years the solution was clearly in front of us. We just could not see it.
The problem was that we were trying to monetize open source software. How do you monetize something that is freely available to everybody without any serious limitations? We were struggling with that question for years. Yet, I guess that if you ask young kids they can easily answer the question. Because in fact, the answer is trivial: you don't. You can't monetize open source software.
But do not worry. Open source developers are not going to starve to death. There is a nice way how to earn money. The solution is quite simple. Do not monetize on software. Monetize on services.
Software is not a static thing. Software is changing all the time. There are bugfixes, new features, improvements and documentation updates. Software has to adapt to new environments, new business requirements and regulations. Software is a living thing that changes all the time. A software that cannot change is as good as dead. It may work today, but will it work tomorrow?
There is always a lot of work to maintain the software. Therefore software maintenance is a service. And it looks like the people are willing to pay for that service. People will pay to get a fix for a problem that affects them. People will pay to get a feature that they need. People will pay for an improvement that makes their life easier. People will pay for the knowledge. They will pay for a training, for the privilege to gave their questions answered quickly, for ad-hoc documentation updates. Or they will just pay to have their backs covered. People will pay for services.
That is the answer that is perfectly obvious, but we could not see it for such a long time:
Our software is free. Both free as in speech and free as in beer. You can take it, deploy it, run it, modify it and you do not have to pay anybody to do that.
But our services are not free. Do you want a single minute of our time? You have to pay for that. Do not expect any free lunch here. Not even a free snack.
We are not selling the software. We are selling services. In fact, open source software is pretty much just a byproduct of those services. Open source is not a business model. But a viable business can be built on services around open source software.
(Reposted from Evolveum blog)