Roland Dürre
Friday March 11th, 2016

“Inspect-and-Adapt“ Your Estimates!

td-logoI always like visiting the Munich Techdivision. For me, this event, organized by my PM-Camp accomplice Sacha Storz, is very attractive. The next event is on March, 16th, 2016, between 7 and 8.30 p.m. They finish earlier than usual, because they want the visitors to have the chance to be back home in front of their TV sets for the Bayern Soccer Game. The location is the TechDivision offices at 73, Balanstr. (building 8, 3rd floor) in 81541 München.

Here is the official call for participation:

According to estimations by User Stories, project planning still plays a central role in agile software development. Imprecise estimations will, for instance, cause higher costs than planned or a longer production time than originally agreed upon.

During a TechDivision study, we investigated the User Story Estimates in four different projects. The goal was: identify problems and propose improvements. It is a rather well-known principle – after all, there is always an “inspect-and-adapt” framework in Scrum.

But: how can it be adapted to estimates? And how are the results of the study related to the #NoEstimates discussion?

I find the topic rather interesting. After all, I am a supporter of the provocative theory: “don’t estimate“– and I can find numerous reasons for this theory. In addition, I find it quite motivating that my own son Rupert will be the speaker.

About the speaker:
imagesRupert Dürre is a consultant at the Swedish IT company Netlight. His interests are various areas of agile software development in requirement engineering and development practices, such as also the question how to organize teams in order to make an effective cooperation possible.

Note:
There is no admission fee for the event. Please register at the website: event.

Well, next Wednesday at 7 p.m., I will be at Balanstrasse and I look forward to seeing many friends!

RMD
(Translated by EG)

Hans Bonfigt
Monday December 28th, 2015

(Deutsch) Ganz agil vorbei am Ziel

Sorry, this entry is only available in German.

Roland Dürre
Saturday November 7th, 2015

Founding my Own Company in 1983 – A Retrospective Analysis

logo-HITLooking for a HIT logo, I found this article. It is based on an interview I gave in August 2014. It was perhaps during one of those (totally confusing) many state-sponsored contests for founders of medium-sized firms. Well, one of those many sponsoring opportunities. I no longer know the name of the person who interviewed me at the time.

The article was published in:

header-1

The basic message of the article is:

My own company founding process went well because many necessary requirements were met. Yet we had not planned them a-priori and rationally. Instead, we only found out about them a-posteriori.

It is a personal report on how InterFace was founded. I will modify it a little for publication in if-blog.de, because it is part of my personal life story. With this report, I would like to encourage young founders and also prepare you for the almost finished fourth instalment of my series “vintage project management”, which is soon to come. Here is the text:

Ever since the early 1980ies, I was tempted to become self-employed. One of the reasons was that I (like many founders of today I know) wanted to take responsibility for an enterprise and decide its future, thus also enjoying my work more. Another reason was that I wanted to earn more money. Strangely enough, this is a motive that does not seem to be so important to many founders of today I know – perhaps because today the restrictions at the work-place are a lot more than they used to be and many people are no longer prepared to sacrifice their private life for their career.

For starters, I set out looking for the “ideal partner” as early as 1982 (I did not look for the “ideal business idea”, because even then I firmly believed that there is no such thing as an ideal business plan). Mind you, it was not easy to find the “ideal partner”, either, but after a little more than a year, I was lucky enough to find Wolf Geldmacher.

He brought enormous entrepreneurial power and was down-to-earth, just like I was. With Wolf, founding the “InterFace Connection Gesellschaft für Datenfern-verarbeitung und Entwicklung von Software mbH“, which was the predecessor of InterFace AG, did not take long. Our topics were IT and Unix. On the then new Unix platform, we wanted to create a successful product. It was important for us to actually have a product, because we assumed it was harder to scale a service. Besides, in 1983 (before the foundation in 1984), we were not sure if body-leasing was going to be a business that would endure. Even then, we believed that Body Leasing (“Letting of Workers” AÜG) would be a sector that, from the legal point of view, occupies a grey zone at the least.

Consequently, it was clear that we wanted to build a product. After various ideas (data base, networking,…), we decided in favour of an office-compatible typing system on Unix. As a name, we chose what we wanted it to become: a HIT: Seen in retrospective, we actually succeeded with something rather risky. It took us only a few years to become the most successful text system by far on Unix in Europe. It was like a dream!

Thinking about it later, I identified persons and important requirements or events without which it would never have worked. We simply were moving in the right time and had enormous luck that made so much just right.

The duo “Wolf & Roland“
Even in the early 1980ies, the two of us firmly believed in agile, lean and open. We were in favour of self-organization and self-determination, formulated our ideas and our requirements and then let our teams do the job. It was all done in a profound sense of belonging together.

Necessary “skills“
In 1960, my parents had managed to get me a place at the Augsburg “wirtschafts-wissenschaftliche Oberrealschule Jakob Fugger” after the four years of primary school at Wittelsbacher Volksschule. This had by no means been an easy task. Later, this school was re-named “„wirtschaftswissenschaftlichen Gymnasium“ (until 1960, it had still been a business school). Book-keeping and business administration, both of which I was taught at the “wirtschaftswissenschaftlichen Gymnasium”, were certainly useful skills when it came to founding a company, but they were not an absolute necessity. The things I had learned at TUM in computer science were basically also not much use. I had learned programming as a student at Siemens. In the Siemens Laboratory, I had learned team work, in the sales department communication, and at Softlab: business.

“The method“
We had developed and lived our own private method of software development. It was a little like what they call SCRUM today. Wolf was the “SCRUM-Master“ (and more than that). He was responsible for the technology and the people. He pointed the colleagues towards quality and made it clear to them that they had to live and give quality, first and foremost for themselves. And I was something like the “product owner” and business director.
There are a few persons who hugely contributed to our success. I would like to give my special thanks to them:

Anton Böck
Stenography and typing had been obligatory at the Jacob Fugger school until 1960, and later, during my time, elective courses. My father forced me to learn both, because he considered these two techniques important advantages when it came to fighting for your professional position.

Anton Böck was my teacher of these two elective courses. I always did quite well in stenography. Whenever I was forced to learn at home, I spent hours drawing stenographic symbols. For me, this was just like calligraphy. Beautiful. And my parents thought I was learning. In reality, I fled into some kind of “meditative drawing”, thus finding solace in my dreams.
Mr. Böck was a strict teacher and he rather liked me because I was an excellent stenographer. But he also forced me to sit behind the typewriter. I hated the typewriter and even when I was sixteen, I dreamed of what a “nice” and “beautiful” typewriter would have to look like. It may sound a little ridiculous, but I am sure that, without this early negative experience of mine with the generation of text, the InterFace Connection would never have developed a text processing system. Which means it would never have become the rather successful product enterprise it was.

Hans Strack-Zimmermann
Hans was my mentor and the person who had made Unix big, both in Europe and at Siemens (here under the brand name Sinix). He awoke my enthusiasm for his vision and believed in our team. And he helped us a lot. This is how, in the end, our success proved him right.

Dr. Peter Schnupp
Peter was a second generation IT pioneer (I see the generation Zuse as the first and myself as part of the third). IT expert, column writer in the Computer-Woche and also quite famous for many other activities. He had an excellent reputation as an expert.

Peter managed to convince the strategic decision-maker of a huge agency that the future of IT lies in Unix and that there is a great local product for text on the market.
Without this stroke of luck, the project CLOU/HIT would never have been a success.

My projects
Even as a young software developer for Siemens AG in the mid-1970ies, I had worked on a great task. In the course of the development of Transdata, I had developed the “Connection Handling“ and also been part of the team that developed “APS“ (user programming language). “Connection Handling“ has a central importance for “Remote Data Transfer”, as it was then called. With APS, it was possible to decentralize computing power to local “data station computers” (operating system PDN) and thus to be the first to break the centralized main frame principle.

With this “superior knowledge”, it was easy for me to shine during big projects and as a logical consequence of this, I was able to change into the department “special projects sales” at Siemens AG. In this department, my most important project was DISPOL, a central project of the Bavarian Police Force – which, early in the 1980ies had given itself the task of changing from paper filing (data), typewriters (documents) and teletext machines (communication!) to electronic data processing.

I accompanied this project until I founded my own company and in the process, I understood how market, customers and in particular authorities work.

Without this personal history of mine, HIT/CLOU would never have become a successful product.

The people at InterFace 
We employed very young persons for production. More often than not, they were still students when they first came to work for us. And in (almost) all cases, they were the right persons to hire. The speed with which these people became overachievers and took upon them a high responsibility was truly amazing.

The right principles
Along with product development, almost as an accidental by-product, a high-quality counselling and co-operation with Siemens in the field of “Unix Operating Systems” developed. Technologically, we were at the source and we learned many extremely helpful things from our operation systems partner. For instance, we were among the first who started using tools that were not at all common in Europe. And we created many novelties, such as for instance the “National Language Support (NSL)”, which later even made it into XOPEN, becoming the basis for all Unix systems.

We used methods (or rather, we invented them intuitively) such as the four-eye-principle of programming, peer2peer-reviews, “extreme programming“, developer rotation and much more. Those methods did not even exist at the time, or else were not known to us. But it just made sense to do it in this way. In fact, it brought us more than fundamental advantages when it came to developing speed, user-orientedness and quality.

Our developers were always in direct contact with the customers. For instance, they taught the HIT courses for our customers themselves and thus understood what the customers wanted. All these factors contributed hugely towards the quality of our product.

Finances
We were well aware of the difficulties. Consequently, we shared the IF computers during our initial founding phase. For the second phase, the marketing, we had planned to share tasks. The InterFace Connection developed the product and looked after our big customer Siemens. InterFace Computer did the porting to the many other Unix systems and the sales for more hardware producers and partners. Until, later, the InterFace Connection took over the entire affair.
Developing a product necessitates a huge “man power”. And men cost money. In 1984 and the year after that, we solved this problem in a very simple way. Wolf Geldmacher and yours truly worked as consultants. Evenings and sometimes also Saturdays were used to look after the product and the team.

As consultants, we charged 150.- DM per hour. That was a terrific rate and the only reason we got away with it was that US consultants with equal know-how were a lot more expensive.

Now you can do a simple calculation: A good month will bring you 200 man hours (we were rather industrious). Multiplied with 150.- DM, that would earn us 30,000 DM in a good month. Subtracting our salary of 5,000 DM each, after taxes, which makes around 6,000 before taxes, this left us with 18,000 for hardware, Heidi (our assistant, who worked for us right from the start) and our students, the product developers.

As early as a few months after our foundation on April, 1st, 1984, we were able to recruit, and immediately use as consultants, some young computer scientists. They brought us similar money to dispose of. Money we used exclusively for product development. And from late in 1985, the product itself saw to it that we quickly had increasing marginal returns.
Events 
Moreover, there were some rather lucky circumstances that helped us very much.

For instance, Siemens had started a very huge project, wishing to develop their own text system for BS 2000 and Unix. Regardless of the fact that these projects had several times more people working for them than our development team, and also regardless of the fact that the Siemens project developers were all grown software developers, none of these projects seemed to get anywhere. They all turned out more or less complete failures.

Yet the concern Siemens needed such software in order to achieve what they wanted to achieve. Consequently, they had to buy licences from two providers – we were one of them. This is how we became the provider and licenced producer of the then market leader in Germany on Linux.

The “technological time frame” worked in our favour: At the time, Unix replaced the numerous varying computer systems of the “medium-sized data technology” MDT. Our product HIT could not have come at a better time.

Those were also the times when using databases became more and more common. As a brand new definition, you had SQL as a “query language“ based on natural language. There even was a German SQL version! So what better than to extent CLOU, the 4GL (programming language of the fourth generation) that enabled HIT to program text elements, to also include an “embedded SQL“ which would suddenly make it possible to send dynamically generated queries to a database during the running of an element program and to then automatically use the received data for creating a document? That was really sensational, and it also came at exactly the right moment.

Lots of luck and just a little ill luck
It was certainly something special that a federal agency was courageous enough to base its future technology on a totally new technology created by a very small provider. A wonderful market development favouring UNIX. Then there were many more courageous and, for us, lucky customer decisions. And a super team…

There were also some problems
Buying the hardware for the development was extremely expensive. As early as 1985, we had to buy MX500 – which at the time was listed at more than 300,000 DM. For us, this was an unbelievable sum. But it was clear that, without this system, we would never be able to manage the necessary speed of development. Two years later, the machine was already totally outdated. We worked on SUN computers and quasi overnight, new fast PC-s with diverse Unix variants that were considerably less expensive came onto the market.

InterFace Computer backed out, the strategic cooperation was no longer working. Consequently, we were forced to buy the rights to the product. This was a huge investment and a hard decision, yet, in retrospect, it was a decision that paid off.

Other necessary requirements
I am sure there are several more causes and incidents without which the enterprise HIT/CLOU would never have become a success. Some of these might be things I no longer even remember or things I am not really consciously aware of. But without all the aforementioned factors, the InterFace Connection as the producer of HIT/CLOU would never have existed. Many special circumstances and coinciding events taken together caused our huge success.

With this article, I wanted to use my own example to show that many requirements must be met if you want to succeed. And that there are things you cannot plan. I also wrote this to encourage you towards having a (reasonable) readiness towards leaving things undone. Yet I also want to show you that it is not all that easy to found a company and that a pragmatic approach is fundamentally important.

RMD
(Translated by EG)

craftsmanshipIn 2013,we at InterFace AG had a beautiful “Technological IF Forum”. It was about “Software Craftsmanship”. We had great guests and competent speakers. The discussions were about questions such as: So how do you become a champion? How to get the necessary experience? How can you get motivated towards perfection? How to best achieve optimality and quality in a team?

One comment in this workshop, in particular, remains unforgotten: at the time, Bernhard Findeiss related (perhaps he also meant it a little provocatively) that a good “craftsman” who wants to become a real champion in his field will have to invest up to 20 hours  for his continuing education. And that usually this will not be possible during working hours. Instead, a significant part of your leisure time will have to be sacrificed for it.

Initially, I was surprised to hear the number. I had to think of how many people strictly divide their time between leisure and work. And I remembered quite a few discussions I had with my colleagues. For instance about how much of the time you spend on your continuing education program can be considered official work time. Consequently, I did a lot of thinking on the topic during the last two years. On the one hand, you certainly need twenty hours each week for practicing and learning if you strive towards championship.

I totally agree. On the other hand, you still need plenty of time in order to work towards success. And then, you also want enough time for your family and private affairs. And I think it can be done. Many of my friends – experts, managers and entrepreneurs, both male and female – live and love their jobs. They are true “champions” and basically spend all their time tackling topics they consider important. And still they are good spouses and mothers/fathers.

I am now a “retiree”. And still, I learn and practice twenty hours each week. Except, I doubt if I am a champion. But I will continue practicing …

RMD
(Translated by EG)

P.S.
For video recordings of the IF Forum, click here: Craftsmanship.

P.S.1
For all articles of my entrepreneurial diary, click here: Drehscheibe!

Sorry, this entry is only available in German.

The work of art by Wolf Nkole Helzle which was created during our IF Forum CRAFTSMANSHIP has already been submitted to all of you as a form of saying “Thank You” (Danke Schön).

Besides, the first two IF Forum presentations (Vorträge) by Wolfgang Menauer and Kristin Block were made available live by me a few days ago. Today, you get the next two!

Firstly, we see Bernhard Findeiss, “Technology Evangelist” at InterFace AG. He tells you about “a day in the life of a software craftsman”:

Dr. Elmar Jürgens of CQSE reports about his own experience and the good team results with Peer2Peer reviews in the interest of quality:

Thanks to Bernd, Elmar and Friedrich!

RMD
(Translated by EG)

IF20x_5172Today is the day!

At 13:00 hours, our Workhop “Craftsmanship” will start!

For more information, click here!

There are still a few vacancies!

And for those who cannot come, we broadcast the presentation live on the internet.

On http://www.ustream.tv/channel/IF-Forum , you can see the video stream of the presentation, starting around 13:45 hours.

In addition, we will – as usual – make a video recording and publish it on youtube (Kanal InterFace AG).

RMD
(Translated by EG)

Dr. Elmar Jürgens is one of the speakers at our Craftsmanship workshop on June, 13th, 2013. His presentation will be about:

Knowledge Transfer through Lightweight Reviews
Experiences from six years of working in a heterogeneous team

Here is the abstract of the presentation he will give for us:

Experience, competence, culture, quality and knowledge are the focal issues of the IF Forum on craftsmanship. How can we generate a software development culture that promotes mutual exchange of experience, competence and knowledge in such a way that the software quality will benefit?

For me, the answer is a culture of lightweight peer reviews. There is hardly any quality control mechanism the usefulness of which has been more thoroughly investigated than peer reviews. Moreover, they give us an effective tool for knowledge transfer. We have been using it for many years now. Regardless of this, many teams still do not conduct any peer reviews during their development phases.

In my presentation, I will introduce a lightweight approach for continuous code reviews where the programming and review phases are separate. This gives the programmer and reviewer respectively the chance to decide by themselves when, where and how fast they wish to work. In the development of the Open-Source programming tool ConQAT, we have been using these reviews for seven years in quality control of all code modifications. On a voluntary basis. We are convinced that this is the main reason why ConQAT is so easy on maintenance and so flexible. I will tell you about our experiences and will also report social challenges and best practices.

About the Speaker:

Dr. Elmar Jürgens is founder and associate of the CQSE GmbH. Elmar wrote his dissertational thesis at Munich Technical University about the discovery, consequences and handling of clones, for which he was awarded the software engineering prize of the Ernst Dehnert Foundation in 2011. As one of the founders of CQSE GmbH, he supports enterprises when they analyse and improve the quality of their software systems. He was among the five best speakers of the Software Quality Days 2013.

Moreover, Elmar is the co-chair of the Nineteenth International Workshop on Software Clones held in San Francisco this year. It is part of the “International Conference on Software Engineering”.

Elmar also initiated a very special community: together with other doctoral candidates of TUM and colleagues of CQSE GmbH, he organizes a regular “Tasting Group”. Whenever they meet, the “Tasting Group” tries out a new idea (in the form of an established scientific paper in the field of software quality) and a new taste (for instance in the form of testing wine or a special cuisine). To me, this sounds like a truly innovative concept!

RMD
(Translated by EG)

Roland Dürre
Wednesday January 16th, 2013

Knowledge and Humour for Computer Scientists

On January, 17th, 2013, the IF-Lab and IF Akademie at InterFace will start a new series of presentations (Vortragsreihe) at Unterhaching.

Thomas Baldus of IF Blueprint AG will be the first speaker at 6 p.m. (in the InterFace building, top floor, teaching area). He will speak about:

APPetite Whetted for Microsoft?!

How exactly does the new taste offensive look as it reaches our shores after its trans-Atlantic voyage?

For more information on the series of lectures, click here: IF-OPEN. Very important – this is not just about the lecture, but also very much about “meeting” during a presentation in the modern format with a “Happy Hour” after the event.

I look forward to a huge audience. Also, I am looking for speakers who would like to share their knowledge and fill the slots on other days. Here is our motto:

From practitioners for practitioners.
Or:
Informatics & Beer.

RMD

Sorry, this entry is only available in German.