Michael Hüttermann

different place to go

2011-11-20T10:37:00.001+01:00

I think I'll continue here.

Agile ALM, at JavaOne 2011

2011-10-10T16:14:00.000+02:00

My taped session "Agile ALM" from JavaOne 2011. This session illustrated recipes for Java developers who want to integrate flexible agile practices and lightweight tools along software development phases. Besides strategies, this session explores state-of-the-art tool chains. The first part of the session discusses Agile ALM, its context, building blocks and history. In the second part, I've zoomed in to one major facet of an Agile ALM: releasing software. I've talked about different common and spread ways how to release Maven based projects, both concepts as well as tools such as Jenkins/Hudson, Artifactory and Sonar.

Agile ALM book available

2011-09-03T18:54:00.003+02:00

Happy Birthday!

Fragile Agile available

2010-08-28T19:50:00.007+02:00

My German book Fragile Agile is available. Have much fun. :-)

Agile ALM: release in November

2010-08-28T19:49:00.000+02:00

Dear enthusiasts, who are interested in my book Agile ALM, and are waiting for it (as I'm waiting for it myself). I've submitted the completed manuscript to Manning end of March. Due to several delays (sorry, I cannot influence that), the book will be available in November, finally, according to the recent schedule of the publisher. I hope to see it printed early enough for Devoxx conference, the second biggest Java conference.

Agile ALM: -update-

2010-07-02T20:14:00.004+02:00

Some news about Agile ALM.

Manning is reading through the manuscript now (again). Chapters are streamed to production now (that includes the steps: copy edit, technical proofreading, typeset of chapters). As you know, chapters 1 to 4 are in MEAP already, the preface and chapter one are available for free.

The timeline:

Forecast (according to the publisher)
10/01 available in stores
07/07 Chapter 5 will be put to MEAP (every 7 to 10 days, a new chapter will be added to MEAP)

Past
06/27 I've incorporated all reviews of Manning's last review batch (3/3) consisting of 8 individual reviews
06/23 last publisher review received
06/14 review phase finished, officially
05/31 Last chapter edited/polished by Manning
05/23 Milestone: Manning's editing was scheduled to be completed today
04/23 Manning put a polisher/editor on the project, timeline for him:
editing 2 chapters a week
04/08 I submitted the revised manuscript, after incorporating feedback
from Manning's reviewers
03/28 I submitted the completed manuscript

Agile ALM -update-

2010-06-09T22:51:00.001+02:00

"Agile ALM" is in Early Access now, finally: http://manning.com/huettermann
At the moment, chapters 1 to 4 or available in early access, the first one for free.

The timeline:

Forecast (according to the publisher)
10/01 available in stores
07/01 review comments incorporated (at the latest), GO for production
06/14 review phase finished

Past
05/31 Last chapter edited/polished by Manning
05/23 Milestone: Manning's editing was scheduled to be completed today
04/23 Manning put a polisher/editor on the project, timeline for him:
editing 2 chapters a week
04/08 I submitted the revised manuscript, after incorporating feedback
from Manning's reviewers
03/28 I submitted the completed manuscript

Agile ALM: -update-

2010-05-31T10:49:00.007+02:00

Now, my completed manuscript is edited by Manning's polisher, completely. (edited based on the already edited version, that I've provided) At the moment, Manning is running another review on all chapters. I have no details about when this external review will be finished. And there is still no concrete timeline available about entering Manning Early Access Program (MEAP), or production. I have the publisher's GO meanwhile, though, that the book can enter MEAP mode, and that they do that these days. I hear that since end of 2009, thus I want to see that in my browser. :-) The book will be promoted to production very soon (according to publisher).

05/31 Last chapter edited/polished by Manning
05/23 Milestone: Manning's editing was scheduled to be completed today
04/23 Manning put a polisher/editor on the project, timeline for him:
editing 2 chapters a week
04/08 I submitted the revised manuscript, after incorporating feedback
from Manning's reviewers
03/28 I submitted the completed manuscript

Agile ALM -update-

2010-05-25T17:27:00.005+02:00

Manning has polished/edited the preface and the first six chapters of my manuscript. At the moment, I'm waiting for the editing result of the last two chapters of my book, chapters seven and eight. Manning committed to complete their polishing process by May, 23th, on the completed manuscript, which was also edited by a commercial editing service before, on my behalf and at my cost. Still no update on entering MEAP mode, and no concrete update on the timeline for going to production.

05/23 Milestone: Manning's editing/polishing was scheduled to be completed today
04/23 Manning put a polisher/editor on the project, timeline for him: editing 2 chapters a week
04/08 I submitted the revised manuscript, after incorporating feedback from Manning's reviewers
03/28 I submitted the completed manuscript

Agile ALM -update-

2010-05-16T17:21:00.000+02:00

- My completed manusript is idling in Manning's content management system since end of March, and waits to be read by everyone who is interested in Agile Application Lifecycle Management -- a more detailed definition of the target audience is provided in the preface (hmm, maybe I could already publish the preface on my web site? that would help to give you an even deeper impression. Will think about that..)

- May, 14th, Manning renamed the guy they put on the project to be an "editor" (formely they called him "polisher"). The guy is not an employee of Manning, he is a guy busy in the Configuration Management domain . I'm not sure, why Manning has choosen him to do the editing job. Configuration Management is an aspect of my book for sure, but, perhaps it has a part of 1%, depending on the definition.

- As you already know, the editor (formely known as "polisher"), sent me the preface and the first chapter back, in an edited revision. Beside punctuation, some synonyms here and there, we had verbose daily communications about many things, e.g. citations. I've returned the resulting versions of the preface and the first chapter on the same business day. Maybe I will give more details on those daily debates later. Interestingly, many phrasings now look the same as they looked like before I've incorporated the feedback from Manning's reviewers and other guys and services

- Manning committed to finish the polishing process by May, 23th, on the completed manuscript. Looks like, this will not happen: they "polished" about 35 pages from 300+ in total, until now.

- Last Friday (May, 14th), the polisher/editor committed on delivering edited revisions of chapter 2, chapter 3 and chapter 4, by Monday (this is tomorrow, May, 17th). I've not received anything yet (now it is Sunday).

- Polishing two chapters a week (Manning's original committment) was very feasible in my opinion, especially, because of the fact that the manuscript was reviewed and edited by 10+ people and also by a commercial service (that I've hired), beside the reviewers who were recruited by Manning. But obviously, the "polisher/editor" overcommited on this editing task, maybe due to other work, unfortunately

- Still no update on entering MEAP mode

- And, finally, I will include a statement of Ted Neward, that he gives me in the context of "continuous integration", more on that later (also more details on all these great contributions, reviews and all that, will follow)

Agile ALM -update- / polishing by end of May

2010-05-06T13:34:00.000+02:00

Manning set the timeline that they polish my completed manuscript by end of May. Looks like, the book is on its final spurt, finally.

Agile ALM and Fragile Agile -update-

2010-05-04T16:07:00.002+02:00

As already written, I've completed my manuscript of Agile ALM end of March. Beside my writing and editing, I've also organized a lot of other people who edited, reviewed and contributed content. Now we are in a phase, where we "bring the book to production" (according to the publisher). In the past, I've already incorporated feedback from Manning's reviewers. Because Manning did not give me a list of any unclear phrasings or similar, I've hired a professional editing service for supporting me, additionally. This service gave me many very good hints. Beginning of last week, Manning put someone on the project to further "polish" the completed manuscript. Unfortunately, I still did not get an official timeline from the publisher yet, when this polishing is supposed to be completed, and when the book will be in production, finally. "The polisher" just goes through each chapter at the moment. Because you know Manning you probably also know that they run an early access program named MEAP. My book is still not available there yet, and I have no idea, when it will be available through MEAP, finally. You may remember that I've announced the book for MEAP already September 2009. I have no influence on that, actually. I'm very sorry about that.

With my German book Fragile Agile, progress is much more visible and transparent. We finished the "bulk text writing", and do editing now. Then, Hanser's editor will give us concrete feedback what sections we should further optimize. But this is something that is scheduled for the near future, and did not start yet. Meanwhile, my book (that I co-author) is listed on the publisher's web site: Fragile Agile@Hanser. The book is scheduled to be available September, 2nd. It will cost EURO 24,90. I think, it should be possible already to buy the book via Hanser. The public web site also contains the official book description. It was created by our lovely editor. She is working in an extremely professional way. It is really a pleasure to write with Hanser. I've attached the book description here: (sorry, German book, German summary)

Gehören Sie auch zu denen, die sich schon in einem agilen Projekt quälen mussten und enttäuscht waren, dass die Agilität nicht funktioniert hat? Und das, obwohl man doch immer wieder hört, mit agilen Methoden oder Vorgehensweisen sei der Erfolg quasi garantiert.
Woran liegt es, dass viele agile Projekte nicht erfolgreich sind? Der Agilen Softwareentwicklung liegen vier Wertepaare und zwölf Prinzipien zugrunde, die verinnerlichen sollte, wer Projekterfolg haben will. Aber wie setzt man zum Beispiel dieses Prinzip um: "Einfachheit - die Kunst der Maximierung der Arbeit, die nicht getan wird - ist wesentlich."
In diesem Buch zeigen Pavlo Baron und Michael Hüttermann mit Witz und Esprit, welche Fallstricke die Agilen Werte und Prinzipien bereithalten und wie sie in vielen Projekten falsch interpretiert werden. Natürlich zeigen sie auch, wie man es besser machen kann, wie man Agilität richtig versteht und im Projekt lebt. Sie greifen dabei auf ihre Erfahrungen aus zahlreichen Projekten zurück und erzählen eine Reihe von lehrreichen Anekdoten, die Sie zum Schmunzeln bringen werden.

Stay in touch ..

Agile ALM: Additional information, discussions, source code, ..

2010-04-24T13:18:00.014+02:00

I have created a project on Kenai to host accompanying sources for my book Agile ALM.

I added configuration and test scripts to demonstrate discussed concepts, where appropriate. Some of the scripts are contributed by the leading experts of the covered topics, others were created in close collaboration. In my book, one of the major tool backbones is Maven. As a result, most of the provided scripts are Maven scripts. Much more details will follow.

Additionally, on the Kenai site, you can use the forum to give feedback, suggestions, drop questions, and so on. Since Manning still does not list my book on their public site, this Kenai project is the public anchor where you have the chance to profit from discussions and code snippets.

Agile ALM, cover

2010-04-23T16:42:00.000+02:00

The cover of my Manning book "Agile ALM" is available as a draft:

It is always better to have a face! :)

Agile ALM – Application Lifecycle Management, with Java and Lightweight Tools

2010-04-21T18:47:00.010+02:00

My new English book left its long running development stealth mode and is now visible publicity. It is about Agile ALM, and will be published by Manning. What is Agile Application Lifecycle Management? ALM is the intelligent composition of build-, configuration-, change-, test-, quality-, requirements-, integration- and release management. It is a comprehensive approach spanning development phases and project roles, taking care of all artifact types.

Agile ALM
- Is the marriage of business management to software engineering
- Targets processes and tools working together seamlessly, without silos (silos are for farmers!)
- Spans software development life-cycle phases, project roles and artifact types
- Enriches ALM with Agile strategies
- Is based on Software Configuration Management and version control
- Is based on a set of lightweight tools, enabling a team to collaborate efficiently

This book has three main purposes. What is Agile ALM and how can Application Lifecycle Management be rolled out using Agile strategies and best of breed tools? Secondly, what isolated tesserae (toolboxes) can be created for these de facto standard tools for use in advanced, real world use cases? Thirdly, how can we choose the right tools, use them right, and integrate them while applying Agile strategies? These questions are the focus of this book. The book delivers all information you need to implement ALM in an Agile environment, in a single location. This book will enable you to understand and put into practice entire tool chains for automating the builds, tests, and continuous integration of your applications. This book helps you deliver quality applications without wasting time on repetitive tasks, or spending time hunting for a toolbox of strategies and tools.

My book is split into the following chapters:

Part 1 Introducing Agile and ALM
1 Moving to Agile ALM
2 ALM and Agile strategies

Part 2 Functional ALM
3 Implementing Scrum
4 Task-Based Development

Part 3 Integration Management and Releasing
5 Dependency management and releasing
6 Productive development environment
7 Advanced Continuous Integration: Tooling and Recipes

Part 4 Outside-in development and barrier-free development/testing
8 Collaborative development and barrier-free development/testing

I have already completed and submitted the final manuscript end of March. At the moment, Manning is inspecting the manuscript and collects issues they want me to adjust, finally.

Please let me say thank you here: I very appreciate the contributing, reviewing and editing work of about 20 leading experts on the market. Thank you, that you helped me. Much more about that later, too.

There is much more to say of course about this book, and I will do that from now on. I will start now throwing updates about the book (content) itself and the further publishing process, on my communication channels including my blog and Twitter, in short intervals. Please stay in touch!

Fragile Agile

2010-04-16T20:24:00.005+02:00

My German book "Fragile Agile" has a face now, i.e. a cover. The book is published by Hanser. This has the effect that we may win the nobel prize with this book 2011, as Hanser did this year with another book. In contrast to this other book, our book here is more for a mass market.

This book is co-authored by Pavlo Baron. We write it in a very close collaboration, you may say, in an Agile way.

As you may expect when you read the title, the book is a unique, very interesting discussion of Agile and ... but stop, let's wait until May for more details on this book. With May, Hanser will list the book officially. This also means you can order it in advance then, e.g. via Amazon, this lovely online book store. This is all for now, cheers!

JAX and Entwicklertag

2010-04-08T22:35:00.001+02:00

Some next talks: May, 3rd, I will be at JAX giving the talk "Das agile Alibi", together with Pavlo Baron, 3.30 pm - 4.15 pm. June, 23th, 2.30 pm - 3.15 pm, I will present the same talk at Entwicklertag Karlsruhe, again with Pavlo Baron. If you are around, let me know.

what a wonderful error message, aka: two favorites

2010-03-31T21:33:00.001+02:00

There are many things I extremely like. Let's illustrate two of my favorites with a descriptive example. Firstly, I like "meaningful" error messages. You know what I mean, nonsense messages adding no value, just being meaningless, no help for the poor user to solve the issue. Rather killing his last razor-thin rest of motivation to get it running.

And a second favorite what I really like is the Eclipse plugin hell. Disclaimer: I really like working with Eclipse! I'm in the JetBrains Academy, but this does not say anything about that. All tools are good, all tools have their pros and cons, I'm also independent in this tooling context. It is just the amount of plugins and cross-dependencies and the way of required issue resolving in case you have configured your Eclipse distribution (and actually yourself too) into a horrible blind alley. Frequently, in cases where you are really fucked up, the only solution is to completely re-start from scratch. It was becoming much better meanwhile in the Eclipse ecosystem though due to some bunch releasing of components.

Guess what! and now bet what is the top of imagination: a fuXXing error message while installing an Eclipse plugin:

Happy hacking!

Cloud Manifesto

2009-12-09T22:52:00.004+01:00

We are very proud to announce that the Cloud Manifesto is available. After a long night of discussing, the world leading authorities launched the Cloud Manifesto. More than 420 years of summarized experience in that domain defined the four value pairs and principles.

One more year ...

2009-12-08T17:34:00.011+01:00

Almost 2010, wow, time is running. And so much happened this year. I supported different projects in the context of Java/JEE, build-, config-, deploy-, release-management and Agile software development, gave a bunch of seminars (including my Java Tooling Bootcamp), wrote multiple articles, gave presentations at JavaOne and other happenings, was responsible stage producer for the tooling track of "Agile 2009", organized many great JUG Cologne events with awesome speakers, and much more .. thrilling!

What will happen next year? Well, who knows! ;-) Some things are already fix including new publications. First of all, I'm writing two books at the moment. The first book is in English. It will be visible soon, stay tuned. The other book is in German. That one, I write with a co-author. Also this book will be public soon enough. :-)

What else? I got a lot of invitations speaking at conferences. Thank you, .. but I don't know yet which ones I can join for speaking. The priority is to finish the books.

OK, thank you for reading, and sorry, that I do not update the blog frequently. Blogs are becoming more and more a legacy channel IMO, it is easier to post a message on Twitter. ;-) And if I write some more detailed things, this happens in form of books, and from time to time articles. Sorry, I'm more a traditional print medium guy. ;-)

Have a nice day!

Maven 3 talk in Cologne, by Jason van Zyl

2009-10-31T17:55:00.005+01:00

I'm very proud that Jason van Zyl, founder and driver of Maven, will give a talk about Maven 3 in Cologne for the Java User Group, on November, 16th. I'm also very proud that I managed to catch another international high-end speaker for a presentation, for coming to Germany. Again, other groups profit from my preparatory work, and invited him too, now, for making a small side trip there. I will never understand, why there are few to zero cooperations between Java user groups, and where the benefit is to provide exactly the same talk in a distance of few days a bunch of miles away. Details about the talk here: http://jugcologne.eu.

Hudson@tutego

2009-08-14T00:19:00.002+02:00

I'm proud to announce that I'm now in cooperation with tutego giving Hudson seminars.

Remaining conference performances 2009

2009-07-26T21:58:00.008+02:00

The remaining months of 2009 I will have a different activities focus than conference performances. Nevertheless, there are a few talks I will give and events you can see me 2009:

- Agile 2009, August 24-28, as being the responsible stage producer for the Tools for Agility track.
- Java User Group Munich, October, 12th, "Simplifying Development and Testing of GUIs with the Swing Application Framework (JSR 296) and FEST"
- Devoxx, November, 19th, Methodology talk, details will follow ..

This list is limited to conferences where I will appear. It does not include Java User Group Cologne events I organize myself, seminars/workshops and similar.

Völliger Stillstand in Unternehmen

2009-07-16T18:35:00.004+02:00

Jens Coldewey pointed me to this video of Peter Kruse, Honorarprofessor Organisationspsychologie. Peter set up eight rules to achieve the complete deadline in companies. 3 minutes 37 seconds which I really like. Thanks Jens. (the video is in German)

What a wonderful location ..

2009-07-14T22:55:00.003+02:00

Agile @ Microsoft, or: did Microsoft manage the turnaround?

2009-07-10T12:34:00.012+02:00

Do you want to know how to deliver software in a big (what I mean is really big) project in-time and aligned at customers' needs? Please have a look at the very interesting article about the development of Windows 7.

Some of my highlights out of this article (in German, sorry):

"Microsoft hat gelernt, den Fehler zu fürchten. Die Firma musste mitansehen, wie sich der Start des Betriebssystems Windows Vista [Windows 6] um Jahr und Tag verzögerte. Es war viel schwerer als gedacht, alle Fehler auszuräumen. Als Vista endlich auf den Markt kam, wirkte es sperrig, überfrachtet und mitunter quälend langsam."

"Windows 7 ist stabil, schnell und leicht zu bedienen. Es ist sogar elegant."

"Die Entwicklung [von Windows 6] war ingenieursgetrieben", sagt Microsoft-Sprecher Thomas Mickeleit. Von nun an, so viel stand fest, sollte der Kunde treiben. Nur: Wer war dieser Mensch?"

"Der Riese aus Redmond - 90.000 Mitarbeiter, 60 Milliarden Dollar Umsatz [...] gilt seit 34 Jahren als das Gegenteil von cool."

"Die Extremprogrammierer, wie sie sich nannten, zeigten die typische Strenge von Häretikern. Ihre Disziplin war die Antwort auf die bürokratische Verzettelung, die in der Software-Industrie damals um sich griff. Immer monströsere Programme wurden in Auftrag gegeben, mit Anforderungen oft hoffnungslos überfrachtet."

"Die kleinste Einheit vieler Entwicklergruppen bei Microsoft ist nun das Paar: je ein Programmierer und sein wachsamer Gegenpart. Die beiden hängen zusammen wie eine Zweierseilschaft, die durch eine Steilwand steigt: einer voraus, der andere sichert am Seil. Nur scheinbar ist das eine Vergeudung hochbezahlter Arbeitszeit. "Das Vieraugenprinzip hat die Fehlerzahl drastisch gesenkt", sagt Microsoft-Manager Fischer."

"Die gemeinsame Stunde der Wahrheit schlägt in Redmond täglich um 16 Uhr. Dann wird das gesamte Tagwerk der Programmierer, das "daily build", auf rund 5.000 Testrechner aufgespielt. Ein jeder hat seine Eigenarten: veraltete Festplatten, obskure Grafikprozessoren, schlecht programmierte Drittsoftware oder neueste Exotenprogramme. Hinzu kommt ein Simulator, der dem Betriebssystem etliche hundert Mäuse, Tastaturen und Scanner vorgaukelt."

"In jeder dieser elektronischen Lebenswelten muss sich das neue Windows bewähren. Nacht für Nacht durchläuft die Software hier ihre automatischen Testprozeduren. Die erste Vorabversion, die Microsoft zum öffentlichen Herunterladen freigab, trug die Nummer 7.100."

"Viele Projekte gerieten in Rückstand. Die Leiter schickten zusätzliche Programmierer ins Getümmel. Das machte alles noch schlimmer, denn mit steigender Kopfzahl wuchs auch der Abstimmungsbedarf."

"Spätestens seit der schier endlosen Geschichte von Windows Vista sind solche Schrecken auch den Entwicklern von Microsoft nicht ganz fremd. Windows 7 dagegen ist nun nach kaum zwei Jahren fast fertig. Und der Start wurde schon zweimal vorverlegt, zuletzt auf den 22. Oktober."

It looks like Microsoft managed the turnaround from a bumpy big company taking years for new error-prone releases, to a lean, focused, communication-driven, leading IT company with frequent, stable releases aligned at customers' needs. Back to the roots, congratulations!

I'm on Twitter

2009-07-08T22:00:00.005+02:00

Now I'm also on Twitter .. I did resist three years (or was it two?) .. where I saw all those nerds at JavaOne hacking those small messages into the machine. I wrote about Twitter in my JavaOne conference report for JavaSPEKTRUM .. and now, I have one channel more to manage -- wow, cool! The first task was installing the bridge between my Twitter account and my Facebook account.

Tooling + Team = Agile

2009-06-30T18:51:00.046+02:00

Pretty often I monitor anti patterns, suboptimal or even manual centric approaches and misuse of tools. As described a couple of times in my blog already or explained in my conference talks, workshops and seminars it is very important to create a setting of right tools which are used right -- aligned at the customers' individual requirements.

But tooling is only one recipe to work in an efficient and "Agile" way, the other is the team communication. A direct, open, constructive team communication is even more important than using the right tools right as you may remember from the first value pair of the Agile Manifesto. Yes, you may say "Natch! Those soft skills are widly spread". But also here I often monitor anti patterns which are of disadvantage for the customer.

If you want to learn more about Agile development with using the right tools (for your individual environment) in a cooperative atmosphere just to save money, increase motivation and finish your project successfully please read my book or join my bootcamp or my other seminars or wait for my second book :-). You may also want to drop me a line if you think you may have some room for improvement, or just need another unladen opinion.

Many of you ask me to write more and in more detail about tooling, teaming and best practices giving my opinion online .. many of my fellows do that, I know .. I'm sorry, but running already on 140% workload, .. I will point to performances here mainly, but .. ok, I will improve :-) Please monitor the performance area on my site to be informed about my articles and all this other stuff where I discuss Agile in more detail continuously ..

Addendum: Tooling + Team = Agile. Yes, you are right, it must say Tooling + Team + x = Agile. ;-)

Plumbers

2009-05-26T09:23:00.007+02:00

This morning we had a plumber in our house. Our shower was defect in we gave him a call. What did he do? He installed a new hardware. But it was not only a default one -the cheapest on the market- it also has less features than the defect one. No, we do not want to fill the bath tube manually! To make matters worse the plumber wanted to bill 2 1/2 working hours for that, a task which took him 15 minutes. And he wanted to repair something totally different too, a part not defect at all. Of course this will go into another iteration.

Why do I write this? I think about some externals I worked with recently: billing (too many hours) for some activities which do not fit to the customer's goal, delivering the cheapest worst solution without synchronizing with the customer at all. The provided solution do provoke intensive manual activities continuously. Wow, there are some massive parallels between (bad) plumbers and (bad) IT consultants.

JavaOne 2009

2009-04-27T20:02:00.005+02:00

TS-4421 "Simplifying Development and Testing of GUIs with the Swing Application Framework (JSR 296) and FEST"; June 05 2:50 PM - 3:50 PM, together with Alex Ruiz, Oracle

BOF-5394 "Improving the Java User Groups (JUGs)" June 04 7:30 PM - 8:20 PM, together with Dan Sline and John Yeary

Technical session at Jazoon

2009-03-30T23:48:00.001+02:00

I will give a technical long talk about "Simplifying Development and Testing of GUIs with the Swing Application Framework (JSR-296) and FEST" at Jazoon, June 22-25, 2009 in Zurich.

Devoxx 2009 conference report

2009-03-30T23:46:00.000+02:00

The current issue 2/2009 of JavaSPEKTRUM contains my conference report about Devoxx 2008.

JavaOne 2009 proposal accepted: SAF and FEST

2009-02-27T13:27:00.003+01:00

The technical submission Alex Ruiz and me submitted for JavaOne 2009 was accepted. So please stop by and enjoy the session "Simplifying Development and Testing of GUIs with the Swing Application Framework (JSR 296) and FEST". I will post further details on the exact location and time as soon as the program is finalized.

JUGC, GeeCON, SQS Conference, Javamagazin, Agile 2009 ...

2009-02-15T15:37:00.015+01:00

My last post was some days ago .. so here I catch up with a couple of news. The last Java User Group Cologne event on February, 9th was a big success. We had an advanced talk about AOP. Additionally I managed to get Corneliu Vasile Creanga from Adobe to Cologne to talk about Flex, Air and the integration to Java. Stay tuned for the next events and prepare yourself for some surprises. ;-)

GeeCON invited me to give a talk in Cracow in May. I will talk about "From Zero to Hero: Simplifying Development and Testing of GUIs with the Swing Application Framework (JSR-296) and FEST". Thank you, looking forwarding visiting you fantastic guys and that lovely city.

Also in May I will give a workshop about "Agile Testing of Java Rich Clients" during the SQS Software & Systems Quality Conference 2009. Beside agile concepts, functional/acceptance/component testing and general GUI testing strategies the workshop will cover a full-fledged infrastructure to test and inspect the design/code of your application. Tools discussed are Findbugs, Checkstyle, EMMA, TestNG, FEST, EasyMock, PMD, Fit, Ant, Maven, a continuous integration server and a couple of others. Beside other frameworks and solutions I will also discuss JavaFX. Looking forward giving that interactive power workshop.

The current isse 3.2009 of Javamagazin published my article about IntelliJ IDEA 8.0 and RubyMine. Check it out if you like to be informed about what is going on at the bleeding edge of IDE development.

March, 3rd the Agile 2009 will close its call for papers. I'm responsible for the Tools for Agility track. Many good submissions already went in .. expect a great program.

I'm in the program committee of JavaOne 2009. A bunch of talks were already accepted, many are still in review. It is hard to filter the best and most interesting talks out of those billions submissions. You can already register for the worlds best and biggest Java conference -- sorry, Stephan ;-) -- taking place in San Francisco beginning of June.

Session proposal "Refactoring from Visio to Powerpoint, and Prefactoring to Code"

2009-01-05T22:01:00.003+01:00

Recently the following proposal for Agile 2009 come to my mind:

“Refactoring from Visio to Powerpoint, and Prefactoring to Code” — A practical guide for an agile architect. The presented patterns and Web 2.0 tools show how architects can cope with recent challenges in designing complex n-tier, scalable systems and generating code seemlessly. The automated process transforms design artifacts from PowerPoint to Visio (extended pattern catalogue for roundtrip-engineering in progress) and prefactors the artifacts to working Assembler code. This process is fully SOA enabled.

Process/Mechanics
We will finish with a real world cloud computing demo.

Learning outcomes
* learn how to refactor from Visio to Powerpoint
* learn how to prefactor to working code

Level: Expert

If you also have a paper you want to propose, perhaps a more realistic one, go to the agile tooling track and submit your paper .. looking forward seeing your submissions! :-)

JavaFX article, letter to the editor

2009-01-04T12:55:00.005+01:00

Recently I sent a letter to the editor to the Javamagazin concerning the German JavaFX article "JavaFX 1.0 - Ein erster Eindruck" published in edition 2.09.

Liebe Redaktion,

der Artikel „JavaFX 1.0“ aus der Ausgabe 2.2009 wie auch schon sein Vorgänger aus 12.2008 haben mein Interesse geweckt. Gleich vorweg: Grundsätzlich teile ich die Begeisterung des Autors bzgl. JavaFX (JFX). Ich freue mich über das an JavaFX, was bis dato bereitgestellt wurde und freue mich auf die weitere Evolution. Dennoch sollte man das eine oder andere etwas kritischer betrachten. Der Artikel ist aber praktisch so wohlwollend geschrieben, er könnte auch von der Marketing-Abteilung von Sun Microsystems geschrieben worden sein. ;-)

JavaFX hat im Hause Sun Microsystems eine strategisch bedeutende Rolle eingenommen. Diese Feststellung des Autors ist vollkommen richtig. Ich vermisse allerdings ein Hinweise auf die doch weitreichenden Implikationen und Reibungspunkte, die damit verbunden sind. Der enorme Abzug von Ressourcen aus anderen Bereichen macht sich beim Fortschritt der jeweils anderen Aktivitäten bemerkbar. Was ist aus der Ankündigung geworden in kurzen Abständen Java-Hauptreleases bereitzustellen. Nach langer Unruhe und suboptimaler Kommunikation wurde nun erst kürzlich ein Mission Statement bzgl. Java 7 an die Öffentlichkeit getragen. Das allerdings erst schön sichtbar und publikumswirksam auf der Devoxx-Konferenz. Trotz der Stärkung des Clients sind in eben der vom Autor titulierten 500-Tage Phase Sun's Rich Client Urgesteine und führende Köpfe Hans Muller (auch Vater vom Swing Application Framework, JSR 296) und Chet Haase zu Adobe gewechselt. Wie passt das zusammen?

Ob Sun sich mit dem Release 1.0 einen großen Gefallen getan hat, so die These des Autors, sehe ich ambivalent. Sicherlich ein guter Marketing-Schachzug nun (endlich) mit etwas Greifbarem herauszukommen. Aber hat hier nicht auch das Marketing die Taktfrequenz bestimmt, genauso wie vor 18 Monaten bei der doch sehr überstürzten Ankündigung von JavaFX auf der JavaOne 2007? War es nicht jetzt das Marketing, dass auf eben dieser Devoxx 2008 die jedes Jahr praktizierten Sun-Keynotes publikumswirksam füllen wollte und ggf. deswegen JFX 1.0 kurz vorher bereitgestellt wurde? Rein technologisch hat sich Sun mit 1.0 für meine Begriffe keinen Gefallen getan. Zu viele Bugs, Ungereimtheiten und fehlende Features sind latent vorhanden, dazu im Laufe dieses Beitrags noch mehr. Andererseits: Insgesamt sehe ich doch Parallelen zwischen der Historie von Java selbst und nun JavaFX. Erst ab Versionen Java 1.1 und Java 1.2 wurde es tatsächlich produktiv nutzbar. Zwischen einem Preview- und einem finalen 1.0 Release unzählige Änderungen umsetzen ist eine Sache, die direkte Ankündigung, dass noch weitere schwergewichtige Änderungen folgen werden eine andere.

Laut Autor sei die Vermischung von Swing- und Java-2D mit JFX 1.0 deutlich besser gelungen.Das ist aber nur eine Seite. Sowohl Swing als auch Java-2D sind nicht neu, ganz im Gegenteil. Für mich ist vielmehr die entscheidende Frage, wie es mit der Vermischung JFX 1.0 mit der bereits existenten und genutzten Bestandstechnologie aussieht. OK, man kann Swing-Komponenten wrappen. Es bleibt aber weiterhin mehr als undurchsichtig, was JFX technologisch genau bezwecken will? Der Ankündigung JFX als Swing-Abstraktion anzubieten, sind kaum Taten gefolgt bzw.: sie sind inkonsistent. Vielmehr baut sich immer mehr eine gänzlich neue Client-Technologie auf. Schaut man mal genauer auf die von JFX generierten Objektgeflechte, wird es einem schwindelig. Swing-Komponenten und proprietäre FX-Komponenten, FXNodes, SGNodes und SGParents laufen einem da beispielsweise über den Weg – nur mit viel Wohlwollen kann ich hier ein schulbuchmäßiges Design ausmachen. Obwohl das Wrappen und der knotenbasierte Ansatz (doch nun wirklich nicht innovativ) in der Theorie gut klingen, wirkt in der Praxis das eine oder andere unter Zeitdruck “zusammengezimmert”. Die ganze Holprigkeit manifestiert sich beispielsweise in der Interoperabilität zwischen Bestandstechnologie und JavaFX. Der derzeit sauberste Weg eines Aufrufs einer FX-Anwendung über Java gelingt beispielsweise derzeit über Java's Scripting Engine. Das kann aber wohl doch nicht das Ende der Fahnenstange perfekter Interoperabilität sein?!

Bei der Erstellung der FX-Anwendung lassen sich insbesondere auch durch Wiederverwendung von existenten Komponenten in der Tat recht schnell brauchbare Ergebnisse liefern. Stimmt. Dies trifft aber grundsätzlich auf jedes Java-Projekt zu, in der hier diskutierten Zieldomäne beispielsweise auch auf das “Swing Application Framework”. Dem Binding-Feature von JavaFX ist dabei zurecht ein mindestens gleichgroßer Stellenwert einzuräumen. Schade, dass es JSR 295 nicht nach Java 7 schaffen wird. Ein Binding-Mechanismus für JavaFX ist auch dringend nötig: Eine munterer Sprachblumenstrauß innerhalb eines Artefakts wird einem “Designer”schwerlich zumutbar sein.

Die Feststellung die Sprache JavaFX sei nicht Java ist nicht überraschend, die Ableitung JavaFX sei eine domänenspezifische Sprache halte ich aber für irreführend und recht wohlwollend. Irgendwie ist dann doch plötzlich alles eine DSL. Nüchtern betrachtet: Nicht wenige sehen die Einführung einer gänzlich neuen Sprache sehr ambivalent. Warum muß alles neu sein? Hätte es nicht auch eine bereits bekannte, stabile, reife Sprache sein können?

Es stimmt schon, dass der Einstieg für kleinere Testanwendungen verhältnismäßig leicht ist. Wie weit das alles schon produktionsreif ist, bleibt fragwürdig. Sehr schnell stößt man auch auf interessante Erkenntnisse. Dies gilt beispielsweise für das funktionale, über das GUI angestoßene Testen solcher JFX-Anwendungen, was für mich als Commiter des FEST-Frameworks von besonderer Bedeutung ist. So sind die darunterliegenden Swing-Komponenten tatsächlich “versteckt und vergraben” und Oberflächen nur mit raffinierten Mechanismen funktional testbar. Auch hier wangt das Ziel der Interoperabilität ganz offensichtlich. Eine weitere Grenze ist das Tooling. Das eine Sprache, die noch „under construction“ ist, nicht sonderlich viele Unternehmen zwecks Bereitstellung von Werkzeugunterstützung anzieht, scheint offensichtlich. Der wesentliche Punkt dabei ist allerdings ein anderer: JFX ist bis dato eine proprietäre Angelegenheit von Sun Microsystems. Es fehlt ein JSR. Es wäre natürlich noch sauberer gewesen zunächst mal mit einer Spezifikation anzufangen oder nebenläufig eine zu schreiben (hervorragende Kommunikationsgrundlage!), insbesondere auch, um darin zu definieren, wohin die Reise eigentlich hingehen soll. Derzeit ist der JFX-Leistungsumfang kaum zu beurteilen, fehlt doch jedliche konkrete Definition der Zielsetzung. Von der (fehlenden) Anforderung zur “weich umrissenden” Zielgruppe: Traurigerweise sind konventionelle Java/Swing-Entwickler sicher nicht Bestandteil der Zielgruppe von Version 1.0, was man beispielsweise an den größeren Defiziten bei der Interoperabilität erkennt. Das ganze Drama spiegelt sich beispielsweise auch bei der aktuell einzigen, halbwegs soliden JFX-IDE wider: NetBeans 6.5. Hier wird fleißig unterschieden zwischen Java und JavaFX. Warum? Die ersten Fragen stellt sich ein Entwickler, wenn er vor der Entscheidung steht ein Java-Projekt zu erstellen, um dort FX-Artefakte zu hinterlegen. Oder soll er lieber den anderen Weg gehen: Ein JavaFX-Projekt aufmachen, und dann dort Java-Artefakte erstellen? Die Entscheidung, wie immer sie ausfällt, hat operative, nur halbwegs nachvollziehbare Konsequenzen, beispielsweise wie ein Artefakt ausgeführt werden kann. Für mich ein Schritt rückwärts auf der Reise der polyglotten Gemeinde.

Ob man den aktuellen Meilenstein 1.0 nennen muß oder nicht: Es ist noch viel Arbeit notwendig. Es mußte in den vom Autor angeführten 500 Tage sehr viel Grundlagenarbeit durchgeführt werden (hausgemacht. Nochmal die Frage: warum eine neue Sprache?), und der aktuelle Stand macht einen guten Eindruck. Natürlich füllt JavaFX eine Lücke bei reichhaltigen, auf Java basierenden Client-Anwendungen. Multimediale, mit Animationen versehene, deklarativ zu beschreibende Applikationen für Browser, Desktop und später auch für mobile Endgeräte, und allem voran das schöne Herausziehen der FX-Anwendungen aus dem Browser auf den Desktop, all das macht Lust auf mehr. Ob aber noch weitere 500 Tage reichen werden, um den enormen Vorsprung von Marktbegleitern wie Adobe aufzuholen, bleibt abzuwarten. Denn auch jetzt schon und auch ohne JavaFX lassen sich Frontend- und Backend-Technologien von Enterprise-Anwendungen sauber entkoppeln. Ich hoffe, die weitere Evolution von JavaFX gelingt und es wird keine Weiterentwicklung von anderen Technologien und Standards darunter leiden müssen.

Mit freundlichem Gruß
Michael Hüttermann

JavaFX - some first deeper impressions

2008-12-30T01:08:00.005+01:00

I'm still very thrilled about JavaFX -- don't get me wrong, but I'm and more and more disappointed too.

This begins with that I feel like an early adopter of an alpha-stage technology working with JavaFX 1.0. The more I look into that technology the more I'm wondering: is JavaFX supposed to encapsule Swing or not? I thought it will be an abstraction? Under the hood you use well-known components like JButton, but pretty nicely hidden and wrapped through FXNodes, SGNodes, SGParents and many other components. Other controls are native JavaFX components like TextBox/text. Some kind of missing consistency ...

The whole interoperability JavaFX/Swing is really, really bad. Working with the ScriptEngine isn't the optimum right? Being dependent on getResourceAsStream is so error-prone. Actually I'm wondering who was the target audience for JavaFX 1.0. I guess not the Java / Swing developers ... ?!?

The next is the tooling .. ok, a lack of tooling is not a big surprise as long as no stable spec is available. I tried JavaFX with NB 6.5 and it is pretty ok, but I hack my scripts inside the NB editor just to valid that they are correct and then copying them into my IntelliJ IDEA or Eclipse workspace. Why? I'm not sure ... but it starts with that NB distinguishs between Java and JavaFX artifacts and projects everywhere. What a pain.

The whole JavaFX stuff gets better .. starting from zero at JavaOne 2007. But I have the impression for JavaFX the whole product is still driven by (suboptimal) marketing, not by software engineering. And all that with a lack of communication ..

Agile 2009's Tools for Agility Stage: CfP is opened!

2008-12-17T19:38:00.004+01:00

The Agile 2009 conference opened its CfP. I'm the responsible producer of the agile tooling track this year .. looking forward seeing your submissions! :-)

Back from Devoxx ...

2008-12-13T14:49:00.006+01:00

Back from Devoxx! Meanwhile I returned from my journey to Devoxx, the biggest and best independent Java conference in the world hosted in Antwerp, Belgium. After four days (traditionally I skip the Friday half-day chill-out) of university sessions, tools in action talks, BOFs, conference talks and 293827 chats (estimation) I'm pretty blown up with Java - even more than I'm already normally.

Topics were well spread cross the whole stack more or less, the quality was again very good, I met a lot of old friends and got in touch with new interesting guys. Arriving on Sunday I finished preparation of my own talk I gave on Monday. I talked about the combination of TestNG, Fit and FEST for functional testing of Rich Clients and it was inspiring. Interesting a guy asking me afterwards if I see many projects testing their applications. :-) Actually, not many enough. With another guy I talked about running functional tests on a build server with the help of Xvfb.

The University session "Scrum in Practice" by Chery Sylvain and Yves Hanoulle was pretty good. Later I had some drinks with Aaron Houston, Antonio Goncalves and many others mainly French guys.

My highlight on Tuesday was Filthy Rich Clients: beyond Java by Romain Guy and Chet Haase. Later on I chatted with Mike Keith asking him if Oracle will by Sun. :-) Nice also was having a beer with Jason van Zyl, founder of Maven. He told me Sonatype will more focus on products than on services in the future. I asked Reggie from Sun for a special speaker for the JUG Cologne - secret who it is. ;-)

I also had long conversations with John Smart, author of the book "Java Power Tools". I managed to recruit him before to be assistant stage producer for the Agile 2009's Tooling track .. and he is such a nice guy!

Wednesday I had a nice "nice to see you again coffee" with Mike Wiesner after his session "Security Patterns revealed" and joined the session "Introduction to the SpringSource dm Server" by Sam Brannen. Actually the server is better than I thought the time they went live with it. A running gag was meeting Toni Epple and Sven Reimers from NetBeans DreamTeam doing some kind of IDE bashing. I really like NetBeans and tried NB 6.5 with the JavaFx bundle. To drag some basic visual types into my stage .. is this already the final version of functionality richness? Actually I told the two dream teams guys about the JetBrains MPS program which is in beta now. But Toni, I do not agree about the sizes of the companies and their team velocity. ;-) BTW, the JetBrains booth was one of the biggest this year (and I met a lot of JB and JB Academy guys), Eclipse had a small one niched. JUG Cologne speaker Frank Nimphius gave a talk about JDeveloper and JSF. In the night I was downtown with Rainer and Rene from the Struts PMC. Nice meeting you guys !!

Wednesday was also the JUG leaders BOF. About 20 up to 30 JUG leaders and some VIPs attended. We had some interesting topics especially a European JUG roadshow. But, honestly, here in Germany too many JUGs are too much isolated and only look on themselves. But there already is some kind of Roadshow approach here: If I manage to get world-class speakers to our JUG in Cologne, some other JUGs nearby ask those speakers to give a talk there too. That's of course unofficial and is not want the BOF wanted to achieve: a JUG Roadshow under a common branding where it is a giving and taking for all not only for some.

On Thursday I liked the "JavaFX: Building & Deploying JavaFX Applications" talk by Richard Bair and Joshua Marinacci very much. It was a nice contra talk to all those Flex and Android sessions. Afterwards the JavaPosse was some kind of disappointing. Is it only me or does the quality decreased over the months and years? Later on I drove back to Cologne .. did I mention that it is only two hours by car from Cologne to Antwerp? Very convenient!! Driving to Antwerp David, a guy from the JUG Cologne joined the journey ... we had some really nice discussions about what is going on in the Java ecosystem. Looking forward seeing you at the next JUGC regulars table.

A comprehensive report with details and valuations will be available soon in the JavaSPEKTRUM.

Being part of the FEST team ..

2008-12-12T15:03:00.006+01:00

Since a couple of days I'm team member of the project FEST - Fixtures for Easy Software Testing. That's really awesome .. FEST is very cool especially for efficient and agile testing of Swing clients. I used it myself for some time now and became a contributor and power user. Looking forward to the next tasks .. at the moment finalizing and going public with version 1.0 of the Swing testing framework (and the companion frameworks) is the most recent job on the roadmap. Thanks!

Tooling infrastructure for agile projects

2008-11-21T21:21:00.006+01:00

The current issue 6.2008 of JavaSPEKTRUM contains my article "Eine technologische Infrastruktur für agile Projekte" describing a tooling infrastructure for agile projects. Balanced at the building blocks SCM, testing, VCS, Continuous Integration and Standards/Audits I discuss the right use of right tools like Maven, TestNG, FEST, Fit, EMMA, TeamCity, Subversion, Trac. Actually you can apply the concepts, tools and patterns in agile and also in more traditional process models. The article is a compact overview, for more details and a more big picture you should read my book. :-)

Agile 2009 - Tools for Agility Stage Producer

2008-11-16T13:36:00.004+01:00

Next year the biggest and best dedicated Agile software development conference "Agile" will take place in Chicago, Aug 26-29 2009. It is a big honour that I'm the stage producer responsible for the tooling track. The call for submission is almost finished and will be published soon. Before I assemble a smart review team I can already announce that I will have John Smart in my team as the assistance stage producer. John is a well-known agile tooling guy. Thank you John for joining the team!

Agile Tooling talk at Devoxx

2008-10-19T13:00:00.014+02:00

December 8th I will give a talk at the Devoxx (formerly known as Javapolis) conference in Antwerp, Belgium, about Agile Testing of Java Rich Clients. In this session we show how to test Java Rich Client applications in a comprehensive and sophisticated way. Firstly we write component tests with TestNG. Then we write functional acceptance in HTML with Fit. We bind our Fit-tests to TestNG in order to bridge the technical barrier between unit and functional tests. Visual components are driven and tested by FEST in a sophisticated fluent interface notation. These facets are part of a full-fledged infrastructure based on Maven. Practical tips and and best-practices do complete this highly interactive session.

You will also meet me at the JUG leaders event December, 10th and I think there will be a Java Champions gathering too. If you are interested in having a coffee together at European's best Java conference drop me a line! My schedule is full with official and semi-official meetings and events, but we will definitely find a minute! :-)

Functional Testing of RIAs

2008-09-26T16:27:00.003+02:00

The recent JavaSpektrum issue 5.2008 contains my article about "Functional Testing of Rich Internet Applications". After discussing concepts und best-practices in the context of testing web applications I introduce the two leading and free testing tools WebTest and Selenium. After a deep comparison I include both frameworks (respectively example projects) into a continuous integration process based on TeamCity. If you are interested in reading the German article you may take a look here.

XPDays program online

2008-08-26T20:53:00.002+02:00

The program of this year's XPDays. the German confernce on agile topics, is online. The happening will take place end of November and consists of three different days: November 27th, there will be a one track program with different general talks. Day two on 28th is the main conference with three parallel tracks, day three a community day with open space formats. I was part of the program organisation this year and I'm happy with the program itself and the split to three different days including the Community Day.

JUG Cologne@Javamagazin

2008-08-21T19:32:00.002+02:00

The recent Javamagazin issue 9.2008 contains my article about the Java User Group Cologne. Have fun reading about one of the most active JUGs in Germany.

JUG Cologne - coming up ... !

2008-08-14T19:39:00.006+02:00

Last Monday the Java User Group Cologne hosted an event consisting of two talks: GWT, the Google Web Toolkit, with Patrick Chanezon from Google and "Scrum with JSR 264: Order Management API" with Andreas Ebbert-Karroum.

After that great success the next happening is already coming. On August, 25th WebTest will be presented by its main contributor Marc Guillemot. Afterwards beginning of September we will have a double-feature again .. yes, two conferences during one week .. "Spring Security Patterns", with Mike Wiesner from SpringSource and "Scala" from Ingo Maier, Ecole Polytechnique Federale de Lausanne EPFL.

The autumn is a Spring automn: October, 13th we'll have a Spring double-feature: "Dependency injection for business objects" with Nils Wloka and "Spring Source Advanced Pack for Oracle" with Stefan Glase und Stefan Scheidt. All three guys are well-known Java experts from Opitz-Consulting. Then the next highlight is on November, 3rd. There Stefan Tilkov will speak about JSR 311 / REST and Dalibor Topic about "OpenJDK".

And do not forget the regulars' table of JUGC every last Friday in the month! A lot is happening in Cologne in the Java community ecosystem.

JavaSpektrum: JavaOne 2008 report available

2008-07-26T20:44:00.002+02:00

My JavaOne 2008 report was published in the recent JavaSpektrum edition JS 04/2008. It contains the main news and developments, comments and trends of the Java world and its community. The PDF is available here.

Java Conference Munich: September, 12th

2008-07-19T20:28:00.013+02:00

For a pretty long time I planned to travel to Munich in summer or autumn to visit some old friends for a tasty wheat bear in some nice locations. Now we commited to September and I scheduled September, 11th to September, 13th for my journey. Actually in that week there is the Jahrestagung of the GI event including a Java Conference on September 12th. So very practical to give a talk during the day before jumping into Munich's night life. If you will be there in that time slot and want to chat a bit please let me know and we may get in touch .. ah ok, I forgot to mention that in my talk I will speak about sophisticated testing of rich clients.

Jazoon: 2008-06-25 "Agile Testing of Rich UI Swing Applications"

2008-06-22T19:35:00.010+02:00

During Jazoon I will give the Lab "Agile Testing of Rich UI Swing Applications"

Timeline
Wednesday, 2008-06-25, 11:00 - 12:50, Arena 7
11.00 start part 1
11.50 break
12.00 start part 2
12.50 end

This is a Lab. It will not be an ex-cathedra teaching. So you are engaged to play with your sources and ask questions. I will go around for discussions and answering questions about the topic and -if you like- about the football match in the afternoon. ;-)
You may want to:
- check the infrastructure provided
- run the application
- run the tests via TestNG Eclipse plugin
- run the tests via Ant
- check the TestNG results
- check the Fit result
- modify the TestNG test
- write own TestNG component tests
- modify the Fit test
- write own Fit tests
- write own Jemmy interactions
- integrate the sources into a Continous Integration process (TeamCity)
- configure TeamCity projects and build configurations
- start builds from your TeamCity server site
- start builds from Eclipse
- check build result, examin reports and artifacts

Abstract
We will build up and use a comprehensive technical infrastructure for testing Java Rich Client applications. Based on an example Swing application we explore unit-testing components with TestNG. We discover that TestNG usage benefits especially if you want to group your tests, introduce dependencies or test concurrency.

We explore functional tests by using Jemmy for driving interactions on our UIs and Fit for writing state of the art acceptance tests in HTML syntax. We integrate Jemmy, Fit and TestNG to specify and drive our functional features. We integrate component and functional tests to overcome the technical barrier which is in place normally.

Finally we integrate all our free available tools with Ant and TeamCity. Our unit tests and functional tests run in a row continuously. Our full-fledged continuous integration and test-driven development environment gives feedback about the application quality and sync the application specification with the current application behaviour. We can trigger, monitor and react on (hopefully not broken) builds while using the TeamCity Web 2.0 UI or work seamlessly and do the same, additionally start remote runs and delayed commits from inside our IDE like Eclipse or IntelliJ IDEA.

Preconditions:
Eclipse 3.3.2
JDK Java 6
TeamCity 3.1.1 "Professinal Edition"

Step 1: Install Subversion support "Eclipse Software update"
Subclipse 1.2.4 (only plugin itself)
Update site: "http://subclipse.tigris.org/update_1.2.x"
(although the Subversive plugin is also ok for checking out the sources you will not gain maximum benefit in the context of the TeamCity integration)

Step 2: Checkout project "RichClientTesting"
http://87.230.78.21:4711/svn/projects/trunk
(Fit, Jemmy, TestNG artifacts already included in the project)

Step 3: Customize build.xml: Path to your home folder where your workspace is located

Step 4: Install TestNG support "Eclipse Software update"
Update site: "http://beust.com/eclipse"

Step 5: Install TeamCity Server and TeamCity build agent (running both as a service is best)
http://www.jetbrains.com/teamcity/download/index.html
(default configuration with bundled Tomcat and HSQLDB is enough)

Step 6: Install Eclipse TeamCity plugin
Local TeamCity site: "My settings & Tools" (http://localhost:8080)

JavaOne 2008 - revisited

2008-06-21T13:08:00.017+02:00

There are some new pictures available from my JavaOne 2008 visit. The first picture shows four guys participating in the CommunityOne panel discussion including me the second one from right.

The second picture shows me standing beside Aaron Houston from the Technology Outreach Program of Sun Microsystems. What a great guy .. always an ear for the Java community. The picture was shot in the java.net Community Corner - once again the communication hub of JavaOne and the Java community.

What is a "Team" ?

2008-06-14T15:58:00.007+02:00

Working in teams is essential for running IT projects successfully. But what distinguishes a group from a team? First of all a team is a working group assembled for a longer time to accomplish a goal. The performance of a team is more than the sum of performances of the individual team members. Its members communicate face to face, have respect to each other, avoid any kind of team killer phrases and do not work in ivory towers. The working style of the team can be characterized as cooperative interaction and a collective spirit of responsiblity. In the daily work the boundaries of roles like "the specialist" or "the integrator" as well as disciplinal hierarchies or grades like "junior" or "senior" skilled members are resolved and melt. The team is driven by team spirit and a strong group cohesiveness. Teams often have rituals and renounce vanity fair and jealousy. Seeing that most failed projects fail because of communication problems and for social reasons and very, very seldom because of technical reasons it is more than necessary to form, establish and support the team sense. Unfortunately you are never sure if a working group will emerge to a real team. But if it forms to be a team .. the probability of finishing the project successfully increases dramatically. And: Working in teams makes just pure fun!

Javamagazin: Polyglot Programming

2008-05-25T20:36:00.003+02:00

The current issue 6.2008 of the Javamagazin contains my article about "Polyglot Programming". It deals with the evolution of the Java platform and its trisection in the language itself, the JVM and the APIs/libraries. I talk about standards and approaches unifying the efforts for providing more languages on the VM also dealing with tooling, Closures, Sun Microsystems and the JCP .. as concrete examples I discuss Groovy, JRuby and Scala with IntelliJ IDEA, NetBeans and Eclipse. My article is the leading article of that issue. Additionally I'm also "Author of the month" .. thank you.

JUGC: Terracotta, IntelliJ IDEA and TeamCity

2008-05-22T20:20:00.009+02:00

The Java User Group Cologne is very active in May. Actually we have two events in one week. On Monday, 26th Jonas Boner will give a talk about Terracotta. Jonas is a well-known guy and Terracotta provides a thrilling VM clustering. So it promises to be a nice show. It will take place in the University of Cologne beginning 6.30 pm.

Some days later, on Friday, 30th Vaclav Pech from JetBrains will give a talk at Cafe Teba beginning 6.30 pm. We will speak about IntelliJ IDEA and TeamCity in this very comfortable, intimate location. For the time after the talk I organised a belly dancer .. she will give us a nice performance. The event is open end .. so we will go further to other bars later but will stay in Cafe Teba until 10 or even 11 for sure (having tasty eat and drinks). Seating is limited to 35 persons. Expect some VIPs and much fun .. the regulars table in May is cancelled!

For more details please have a look on the JUGC site. There you can also register for the talks. BTW: recently I gave the JUG web site a little lifting .. now you can find all JUGC facets very easy (hopefully) and without any knickknack. .. and especially .. I did not like the wordpress blog format any more .. and I did activate some spam bots in the past which hacked wordpress adding funny comments. I skipped upgrading to a different version of wordpress, but installed Confluence enterprise wiki now.

JavaOne 2008 .... I'm coming !!!! :-)

2008-05-01T15:56:00.010+02:00

I'm just finishing my preparations for my journey to San Francisco where I will join the JavaOne next week. I'm excited meeting all those old friends there and having some new ones afterwards possibly. The schedule is full with sessions, dates, ..., and of course there are some required courses for the moments when Moscone Center closes its doors. Do not miss the JUG community panel session I'm part of and the Java Champions and JUG Leaders BOFs where I'm active too. If you want to meet me you have good additional chances at the java.net community booth, the JetBrains booth and local pubs. ;-)

For those of you interested in my experiences there and my opinions and an information aggregation of the JavaOne 2008 please stay tuned: I will write a report for the JavaSpektrum.

CU in SF

Jazoon '08 speaker

2008-04-23T22:04:00.009+02:00

I will give a lab at Jazoon '08 about "Agile Testing of Rich UI Swing Applications". Topics covered are Agile Development, types of tests i.e. unit tests and functional/acceptance tests, TestNG, EMMA, Jemmy, FEST, Fit, Ant and TeamCity. It will take place at Wednesday, 2008-06-25, 11:00 - 12:50 in Arena 7. Please expect some recent tech rocket science stuff there .. I will prepare hands-on material in advance that you can download to start directly in that ultra interactive session.

San Francisco: JUG community panel session

2008-04-18T15:16:00.006+02:00

In context of the JavaOne 2008 on May, 5th a JUG Community panel session will take place at Moscone Center during CommunityOne in San Francisco.

Lineup:

Moderator
Dick Wall: Java Posse member and Java Champion

Panelists
Michael "Van" Riper - Silicon Valley Web JUG Leader
Chris Richardson - East Bay Java SIG Leader and Java Champion
Petar Tahchiev - Bulgarian Student JUG Leader
Thor Henning Hetland - javaBin: The Norwegian JUG Leader and Java Champion
Michael Hüttermann - JUG Community Leader, JUG Cologne Leader and Java Champion
Reggie Hutcherson - Sun's Java Technology Evangelism Manager

For information see here and here.

JetBrains Development Academy launched

2008-04-15T22:27:00.001+02:00

Today the JetBrains Development Academy launched .. and I'm part of it. The JetBrains Development Academy fosters a community of experts and evangelists to champion best development practices and promote software innovation worldwide. The Academy serves as a connection point for developers who strive to adopt top methodologies and use JetBrains products to help them achieve that goal. For further details please look here.

IntelliJ IDEA 8 First EAP: JBoss Seam Support

2008-03-28T14:49:00.015+01:00

Since a couple of days the first EAP of IntelliJ IDEA 8 is available. It contains many nice new features. One nice feature is the Seam support. JBoss Seam is an application framework which integrates Asynchronous JavaScript and XML (AJAX), Java Server Faces (JSF), Enterprise Java Beans (EJB3), Java Portlets and Business Process Management (BPM). Seam drives an uniform component model which avoids the distinction between presentation tier components and business logic components. It extends the basic functionality of those standards .. in the area of JSF e.g. with conversations.

JBoss provides several examples for Seam .. e.g. the Booking example. This example is based on Seam 2.0, EJB 3.0, JSF 1.2, Hibernate 3 and Facelets. With this booking Web 2.0 application you can register, login, search hotels and book them.

Formats and meta-informations are annotated in the Entity Beans directly e.g. input length, required fields, pattern and are propagated to the UI seamlessly. The example uses some nice UI components like the date picker and the real-time search. The Web 2.0 and JSF friend has much fun with that.

IntelliJ IDEA provides a very good Seam support including Find Usages, Refactorings (Rename ..), Inspections and code completion (e.g. with EL value and method binding) across the artifacts. The navigation rules from pages.xml are visualized in a navigation graph.

I did deploy the appliation from inside IntelliJ IDEA directly to JBoss 4.2.2.GA. Check the EAP release notes for more information.

German print article: "Agile Testing of Rich UI Applications"

2008-03-16T21:04:00.011+01:00

The recent edition 02/2008 of the JavaSpektrum contains my german article about agile testing of rich Java UI applications named "Agiles Testen von Java-Rich-UI-Anwendungen" -- a subset of my book.

I explain a smooth infrastructure combining the frameworks Fit, TestNG and Jemmy to drive the development of your Java Swing applications drafted with a concrete use case. The use case, the test and specification artifacts, a build script and a small glue module are available for download here.

Beside the frameworks and the integration I show the run in Eclipse and discuss agile best-practices along the way including anchoring the setting into a CI-server like TeamCity. Also mocking and architectural considerations are kept in mind.

You may buy a copy or just the PDF if you like.

German abstract:

"Dieser Artikel möchte das Verständnis für das Testen von Rich-UI-Anwendungen schärfen. Er stellt die drei kostenlosen Werkzeuge TestNG, Fit und Jemmy vor, mit deren Hilfe Swing-Anwendungen umfassend getestet werden können. Eine Integration von Fit/Jemmy-Akzeptanztests und TestNG-Komponententests führt zu schnellen Rückkopplungen. Anhand eines Beispiels wird eine Integration dieser Werkzeuge vorgestellt, die hilft, die technische Barriere zwischen den komplementären Komponenten- und Akzeptanztests zu überwinden."

Dr. Neal Gafter @ JUGC and VIP

2008-03-12T22:23:00.010+01:00

Last Monday, March, 10th, Dr. Neal Gafter gave a talk for the Java User Group Cologne ... and it was an awesome one .. to be correct, it was more than one talk. But let's start at the beginning. I met Neal at noon at the central station in Cologne and we did some sightseeing through Cologne. We were in the Colgone Dom, its treasure room, went to the Rhine, through the old town, I did show him some remaining parts of the past city wall .. actually it was a still existing gate to the city.

Later on we met Ann Oreshnikova, marketing director from JetBrains. We had some further hours of fun together in the city. We went to different breweries, the Früh and the Päffgen, had some local beers and local food .. before going to the event location at about 6 pm. There everything was set up nicely again as always .. with a nice Kölsch buffet (buffet with food and drinks typical for Cologne).

At 7 pm the rooms were full .. about 90 persons attended Neal's first talk in Germany. Neal gave his 60-minutes talk about JDK 7 Language changes .. explaining what should be done in the JDK and why .. he pointed out that it is important to do some cleaning up in the JDK with the coming release e.g. in the area of Generics and Switch Statements. More long-term with a target release of JDK 8 he sees some more features in the language to evolve it naturally -- especially Closures. Closures are pretty easy to use, he said, are transparent for the developer and are not really rocket science .. many other languages have them already and for a long time.

Neal pointed out that "reading" is much more important than "writing" .. you write code once, but you read it very often. So the code must be easy to understand.

After about 60 minutes he was done and answered many questions ... the plan was to postpone all questions to later at the evening .. but there was so much hot reaction .. so we did a Q&A part.

After a 10 minutes break then Ann presented some Javascript puzzlers to the audience. Very, very interesting what is possible in this area and seeing some pitfalls! After that Neal asked the audience what they wanted to hear next. :-) He took his Closures talk out of his bag and gave it completely. Afterwards and in between he answered many, many questions and very interesting discussions arised about Java, Groovy and 10000 topics.

We then distributed five IntelliJ IDEA full-licenses to lucky winners and over a hand full of O'Reilly books while doing a raffle. It was clearly after midnight (!) we were done with this "official" part of the show. We then went down to a bar beside for some further beer.

The event itself lasted over five hours -- many thanks to Neal for coming to Cologne for this great show, your absolutely outstanding performance and your time. I personally learned a lot .. also some non-technical stuff e.g. about the working conditions at Google. I knew some facts before, what I did not know was: employee reviews and ratings are done only by peers!! The manager does not have any influence on that. Isn't that great!?! I'm a supporter of agile disciplines but I very often ask myself in projects .. "Why is this called agile here"? It is about selling an "agile" product very often without understanding the core facets of collaboration being in competition with peers. Walking Google's way you have to help your peers, be polite without any vanity fair, accelerating information exchange ... Many projects should do that .. in regular, short intervals.

Thanks again to Neal and to Ann for your time.

Seamless integration of Java and Groovy with IntelliJ IDEA

2008-03-12T20:27:00.002+01:00

Recently I wrote an article on the Javalobby respectively JetBrains DZone about Seamless integration of Java and Groovy with IntelliJ IDEA. Check it out!

LOP MPS TeamCity defect tracker

2008-03-11T20:13:00.003+01:00

As it is public now I can also mention it and add the following statement to my recent post on the JetBrains TeamCity tracker: The tracker is completely written in a language-oriented programming way based on the JetBrains Meta Programming System MPS framework. Awesome!!!

JetBrains Tracker

2008-03-08T20:26:00.000+01:00

Recently JetBrains migrated the issue tracking of the leading continuous integration and collaboration platform TeamCity from Jira to JetBrains' own and new ticketing system JetBrains Tracker. JetBrains eats its own dog food again here. JetBrains Tracker was developed some time ago already, is very feature-rich and candy for your eyes. Stay tuned to the further development of this awesome ticketing system.

Transition hints: Eclipse > IntelliJ IDEA

2008-02-19T19:02:00.007+01:00

Look at this article about move from Eclipse to IDEA for a nice list of best practices how to migrate from Eclipse to IntelliJ IDEA. Many people start their Java life on Eclipse .. and end there .. ;-) So these hints may simplify your transition even more ..

Weekend in Prague, or: "Meeting JetBrains" ..

2008-02-17T20:13:00.044+01:00

From Friday to Sunday I was on a business trip to Prague. It was a fantastic tour. After this and that I visited the JetBrains office in Prague. I met over half a dozen JetBrains guys and gals, did a walkabout through the location, the pool, the kitchen all this stuff you need for productive work. :-)

After some chatting and after I pilgrimaged three times around the magic desktops ;-) we moved to downtown eating something and having some drinks .. I was happy that the native ones had some very good hints what to eat and especially what to drink .. there are some parallels comparing Prague and Cologne .. also in Prague it is typical to drink very tasty beer brewed by small, comfy taverns .. a great happening and exchange of ideas.

Later on some of us moved further for meeting Roman and Geertjan from Sun Microsystems. Sun has a pretty big office in Prague. We went to further bars, chatting about this and that, exchanging very interesting things which are not eligible to be posted here. :-) But believe me: it was much, much fun and veeery interesting!
We also had some very intensive technical discussions about Java and IDEs of course including an informal IDE shootout (part II). Saturday I saw a bit of the city (with daylight) and worked a bit .. before I flew back to Cologne today.

It was a very nice weekend kicking off some new projects and activities and having much fun .. thrilling meeting Ann, Vaclav, Ilia, Dave, Britt, Jana, ...., and from the dark side ;-) Roman (NetBeans evangelist, who gave a talk last Monday in Cologne on NetBeans 6), Geertjan (NetBeans technical writer, co-author of the book "Rich Client Programming" and zone leader on DZone) and others ...

One conclusion could be: stay tuned for further development. ;-)

Some news ...

2008-02-07T22:25:00.000+01:00

Sun Microsystems invited me again to take part at the JavaOne 2008 in San Francisco in May. The biggest, most leading and thrilling Java conference in the world! Thank you very much, Sun! I'm also proud that I was in the group which reviewed the JavaOne submissions .. I'm thrilled about the program this year. Again fantastic! +++ Before, in March I will be guest at the JSFDays in Vienna meeting Ed Burns again and all these nice JSF guys. +++ I'm Zone Leader at JetBrains Zone at DZone. Join the zone and read leading reports and news about JetBrains and its products like IntelliJ IDEA, TeamCity ... +++ If you like to see me and learn more about agile development (what it really means, anti-patterns, infrastructure) you may want to join my 1-day seminar March 3rd (in German). +++ As you know I drive the Java User Group Cologne (JUGC) ... the next meeting is February, 11th. We have two talks then .. Johannes Link will speak about Groovy and Roman Strobl will visit us coming from Prague and speaks about NetBeans 6.0. Further details on jugcologne.org. Oh yes .. the JUGC changed its location for the regulars table .. it takes place every last Friday in the month in Bar Txangurro. Very nice there. And a last hint to the next JUGC talk .. I managed it to lure Dr. Neil Gafter (Google) to Cologne. He will speak on March 10th about a topic of your choice .. decide on the JUGC mailing list (enter it via JUGC site, right sidebar) if he will speak about Closures, Java Puzzlers, new language features in JDK 7 ... if you have an advanced question about Java, this is the chance to get it answered. You should come early this evening .. to get a seat for seeing and discussing with this JUGC rock star speaker !! +++ There is a small bunch of copies of my book available at Amazon again. Hurry up with ordering ;-) .. I guess also this re-order of Amazon will probably be sold out soon and O'Reilly has to fulfil the aimed reprint of copies ..

IntelliJ IDEA plugin contest 2007

2008-01-31T21:40:00.000+01:00

Today the winners of the JetBrains IntelliJ IDEA plugin contest were announced. The first prize goes to the XSLT-Debugger by Sascha Weinreuter followed by the Struts 2 and the database navigator plugins. The team leader's choice goes to IntelliJad. Congratulations!!

a batch of new copies

2008-01-13T17:44:00.000+01:00

On Friday I got the message from my wonderful editor that O'Reilly wants to start a reprint of my book in a couple of weeks, not a complete new reprint but a new batch of copies including the same content. Not bad for a book which was published only three months before. After publishing the eBook that are the next interesting news ...

My visit at the .NET User Group Cologne

2008-01-09T21:48:00.000+01:00

Yesterday I was guest at the January meeting of the .NET User Group Köln (dnugk). I arrived a bit late at about 7.30 pm after finishing my daily project work and missed the beginning of the first talk: Silverlight 1.1 alpha by Stefan Lieser. The talk was nice giving a good impression what is going on in this field. As Silverlight 1.1 is shipped with an own language runtime (CLR) it is possible to develop with languages like JavaScript, Python and Ruby (beside C# and VB.NET of course). BTW: Microsoft will broadcast the olympic summer games 2008 in Beijing on MSN with Silverlight. ... this will spread the usage definitely!

Comparing all this stuff I saw with JavaFX I question how much time is necessary to catch up the advance of Silverlight, Flash ... (Sun's speed to build up the JavaFX infrastructure is thrilling though). BTW: Interesting was to see the Visual Studio from today .. a lot of eye candy. As I got in touch with .NET some years ago all this looked more frumpy.

The second talk was given by Heiko Hatzfeld. He told a bit about Test-Driven Development. As an agile professional it was interesting to see the .NET view on this topic. Heiko gave some introductions and a demo. Nice to see NUnit in action. A discussion started when unit tests end and when integration tests start. In my opinion if you leave your close unit artifact view and access foreign (from the class point of view) ressources (like a database), you definitely have an integration test. Possibly it would have been even better to mention the role of mock objects more often during this talk phase. Good was that he mentioned JUnit's approach to create a new instance for all separate tests. He told us that NUnit does not do this. I would like to add that TestNG also goes this approach and is more feature-rich for complex tests than JUnit.

Overall a very nice experience. A good atmosphere with about ~30 members and the service at the location was outstanding. It was not the last time there for me definitely. Today I suggested doing a cross-over event where the JUGC meets the dnugk. There is a lot of potential for this.

PDF version of my book is out

2007-12-28T11:23:00.000+01:00

Meanwhile O'Reilly published my book as PDF too. Have a look here.

1. Auflage Dezember 2007
ISBN 978-3-89721-791-1
Seiten 432, PDF ca.13.7 MB
EUR36.00

As you remember the main site for the paper version is here. And further information and errata can be found here.

TeamCity 3.0

2007-12-24T02:49:00.000+01:00

A couple of days ago I upgraded a prototype environment from TeamCity 2.1 to TeamCity 3.0 and I want to share my feelings. TeamCity is free of charge in its professional edition. This edition is limited in some few areas but small to medium sized projects will hardly reach those limits:

* 3 Build Agents
* 20 User Accounts
* 20 Build Configurations

If this is not enough you can buy additional build agents or buy the full enterprise edition. Since the release of TeamCity 1.0 the product was fully commercial. Now you have the entry level edition free of charge. I think this new prizing model will help to spread the market usage of this nice CI-server and was a good choice.

So after installing TeamCity on my Kubuntu I removed .BuildServer in my home directory. I want to start from scratch and do not want to migrate user and project build data to my new version. Once started there are not so many differences to TeamCity 2.1 first glance (beside the new product model of course). Looking deeper into it there are these new build statistics and trends charts. This is powerful. In the past I worked in a semi-agile project which hosts its daily builds on CruiseControl.

Fun enough not all tests are started on the central server and the build results were something like hidden. If you run all your tests on the build server and want to show the test results in detail then these trends and statistics are most probably of benefit for you.

Beside that we now have some per-project access rights. It has been nice to set up user accounts and give them permissions in the past already but now this is even more powerful. Finally there are some new VCS features (new StarTeam support, automatic labeling of code) and further maintenance and incident management improvements (like having thread dumps of crashed builds, a very good idea).

Commonly a new build is started on demand, after check-ins (with respect to some defined sleeping time -- you all know this stuff from CruiseControl e.g. too but with the need to edit xml for that) or in time intervals. Unfortunately the most fine-grained interval is "daily". I would appreciate having a more frequent and regular interval. Edit: TeamCity 3.1 will enable to specify cron-expressions.

TeamCity improved even further and is an excellent tool ... and if you did not do it yet: try it out.

Amazon rating

2007-12-21T18:46:00.000+01:00

Do you have any idea how Amazon rates the books? Seems like they fix the sales rank in real time. The rank of my book oscillates a bit over time.

Reviewer for JavaOne 2008

2007-12-18T14:08:00.000+01:00

The JavaOne 2008 will take place May 6-9, 2008 in San Francisco, California. I'm happy to be part of the team reviewing the submissions. After being invited to JavaOne 2007 as a speaker another big honour for me. Thank you very much, Sun Microsystems! The amount of papers submitted is four-digit by the way.

Back from JavaPolis 2007 ...

2007-12-15T10:16:00.000+01:00

Yesterday I returned back from Javapolis 2007 ... after a long week of diversified power fueling.

I arrived on Monday in Antwerp. It took me about 2.30 hours by car driving from Cologne through the beautiful Netherlands to Antwerp. The most grave part of the small journey was the final meters in Antwerp city during the rush hour. There was not only the early evening traffic, it was also some kind of stress test for my navigation system. The street plan is .. unique? After checking in to the hotel "Express Holiday Inn" in Antwerp city I did write a techical article which will be published in a leading Java print magazine beginning 2008. Please keep interested. ;) Tuesday morning then I did a touch-and-go: after breakfast I drove to the Metropolis center. The line stops near the hotel and arrives at the congress center only about 10 minutes later.

So there was the university day 2 in front of me and I started with the three hours session "Be productive with JSF with Ed Burns and Yara Senger". The content was well presented but not really new. I expected some more details but it was nice to hear that JSF 2 will include some more convenient way to customize standard components or even write owns. Ed was the leading presenter and Yara did a charming and cushy co-speaking.

After the break I dropped into the sessions "Introduction to Java SE 5 and 6" with Sang Shin (just to see him again, he is a smart guy) and "Guidelines and Hints to EJB3 and JPA" with Linda Demichiel and Kenneth Saks. Nothing new here too and I left the room in order to explore this year's exhibition. Many companies were visible there including JetBrains. On their booth the JetBrains guys promoted IDEA and TeamCity (which was released in its version 3, free in the Professional Edition). JetBrains also hosted some sessions, beside Eclipse and Netbeans. Later on I joined the "Ivy" session by Xavier Hanin. I really like Ivy -- I mentioned it in my book as an alternative to Maven 2 for dependency management. Xavier pointed out that there are stilll struggles to use Ivy inside Maven 2 for the dependency management and that its usage is pretty straight forward. But: Maven 2 does a lot more than dependency managmenet so both tools have their value. The last session on Tuesday was "Easy GUI testing with FEST" with Alex Ruiz. With FEST you can write functinal tests embedded in JUnit and TestNG tests. FEST uses a fluent interface to find and control Swing components. I just wonder why these Oracle guys do develop a new framework and do not support Jemmy?!

During the day I met the first guys I wanted to meet ... and it was nice to see old friends again. At 8 pm the speakers' dinner started in a restaurant downtown. I drove there by taxi together with Michael Van Riper (since this week also called "TT") and after arriving there we met Dick Wall, one of the (Google) guys of the Javaposse and we had a drink before the others came. ... And then the party started: it was great having the opportunity to speak to these persons. Thrilling that also James Gosling (Java godfather) joined, I met James during an intimate round table dinner in Frankfurt during the Sun Tech Days one week earlier so my focus was more on the other who-is-who. I was in touch with Neal Gafter for some weeks concerning a talk for the Java User Group Cologne and I got his GO there for a Closures talk in March 2008. As the party was over I went to the hotel and surprisingly there was an after-party with about 15 people including some Sun Microsystems people, Ed Burns, Javapolis crew ... I have to mention I was a bit sick the next morning. The last beer in the lobby must have been bad.

... so I skipped the morning keynotes on Wednesday and started with the JavaFX session by Jim Weaver. I have spoken to him the night before and he is a very nice guy. Afterwards I joined the Google Web Toolkit sesson with Dick Wall. GWT is nice enabling writing Java code which is transformed to a web appliation (Javascript). So you have the chance to write unit tests on your original code for example. There was also the Clustering/Scaling happening with Guy Nirpaz from Gigaspaces (gave a talk for JUGC). I had the same taxi with Guy the day and he is a nice guy. Thanks for paying the trip. Later on I went to the EJB 3.1 event with Kenneth Saks. The content was not that new but the main reason I was there was to have a good seat for the next session: The future of Computing panel with James Gosling, Neal Gafter, Joshua Block and Martin Odersky (who also gave the Scala talk at Javapolis). The light this-and-that talk was amazing and moderated by Carl Quinn -- one of the highlights of this years Javapolis. Afterwards there was the Java Champions BOF. Rags Srinivas, CTO of Technology Evangelism at Sun Microsystems, and others dropped in there too. Klaasjan Tukker from the Netherlands raised some good points concerning the JavaFX debate. It was not perfect how the champions have been included into the information flow in this case. Afterwards the closures BOF with Neil Gafter took place and he explained us the content of his proposal. Neil wrote a prototype for his specification which is complete more or less. He also included function points and rewrite Doug Lea's jsr166y fork-join framework with closures (with and without function types). Neil asks for feedback so everyone should download the prototype and work with closures. After a small hello into the "Java DB and Apache Derby" BOF I returned to the hotel ..

Thursday was probably the most thrilling day. In the morning I missed the bus to the Metropolis twice -- you should know that it is better to raise your hand if you stand at the bus stop and want to take the coming bus. Thanks to Neil and Josh (both staying in the same hotel) I know now where/how to get the bus. The keynotes were another highlight of this year: Adobe Flex respectively the relaunch of Parleys.com. Parleys is the web portal hosted by the BeJUG to collect and publish Java conference talks. It is based on state-of-the-art frameworks and has a new front-end in v2. It is based on Flex/AIR and will be released beginning of 2008. Great work Stephan and Benjamin!!!

After some meet-and-greet and recovering I joined Josh's session. Originally planned as a "Effective Java Reloaded" session (Josh is working on the second edition of his book as you know) it was revised short-hand to be "the closures controversy" session. Wow!! Javapolis managed it to break the freeze between Josh Block and Neil Gafter, keep in mind both are working for Google. Josh discussed Neil's proposal and pointed out what James Gosling said ten years ago: Java is a consolidation language. It is not a science language but includes many "useful" facets from other language to be easy used and widely spread. Generics did had the problem to add much complexity to Java especially because its inclusion of wildcards (done a bit quick on the trigger). And this is the problem Josh sees in Neil's proposal. It is pretty complex and only very few guys will ever need the new functionality. So his suggestion is to include only as much new complexity and functionality into the programming language needed to solve the original problems: in his opinion one problem is "function methods" (you already have a solution with anonymous classes more or less), the second is ressource management.

Afterwards there was the Javaposse live -- beer was sponsored by Atlassian and the four speaker guys were split up: Dick and Carl were live on stage, the other two guys connected by Skype from US. It was fun -- especially as Josh and Neil participated. A very relaxed, nice chat evolved where Josh ended up with a beer in his right hand, a micro in his left and a scarf tied over his head. Another step forward for the Josh/Neil - Closures debate ... and much, much fun!!!

Later on I visited the Web Beans talk with Bob Lee and had a look into the Java Puzzlers (this time again Josh and Neil not like at JavaOne). Josh and Neil also presented a list of possible new language features in a BOF, more or less small sugar ones:
-Improved type inference
-Enum comparison
-String switch
-Chained invocations & Extension methods
-Improved catch clauses
-Array notation for Map, List
-Typedef
-Serialization annotations
-Self type
-Properties

Let's see which ones will get it into JSE 7.

Google and Atlassian sponsored free beers afterwards ... I was part of another delegation (JUG Leaders, Aaron Houston and Nichole Scott from Sun Microsystems, the new JCP chair Patrick Curran, Chet Haase, for several seconds Crazy Bob, ...) in the same bar which was gladly sponsored by Sun Microsystems.

Finally we went back to the hotel where we got some further drinks. Panos and Paris from Greece as well as Michael "TT" Riper joined the last evening happening more accidently. The last night was pretty long compared to the others and I went back to Cologne next day.

Thanks to Stephan Jansen and his BeJUG crew to make this real .. !

The week was thrilling like the year before. Javapolis is the most influencing, the biggest Java conference in Europe (sold out with 3200 people) with the best speakers. It itself sets benchmarks e.g. in the Closures debate and bringing Neil and Josh together to one table. Javapolis is also community driven with many big company sponsors. It is very cheap to get a ticket and the relation of price/quality is awesome. Beside that all JUG Leaders (and an additional JUG delegation) have free entry for the whole week. I do not want to mention the free drinks, the give-aways, the unique atmosphere (without any airs and graces) -- as you see many many points other conference can learn from, especially in Germany.

BTW: do you know what a teetotaler is?

role profile "canned laughter device"

2007-12-10T11:41:00.000+01:00

Some people are very funny, are laughing day and night although life or the project situation is not comic. I'm sure fun has to be a big part especially also for IT projects. Why acting like being on a burial. But don't forget to set your communication on a serious baseplate, being reliable and honest to yourself and your environment.

I know some of these "canned laughter device" candidates ... one of them I saw in front of the Kölner Dom some days ago.

After the game is for the game .... from Sun Tech Days to JavaPolis

2007-12-05T18:48:00.000+01:00

From Monday until today I was in Frankfurt joining the Sun Tech Days. ... and I was impressed about the effort Sun did afford. Almost 1000 attendees visited three days full of high quality technical sessions and hands-on labs in the Messe area of Frankfurt. There have been some booths, a java.net community corner like at J1 (I inspired) and a very good catering with soft drinks, wine, beer, glühwein, currywurst, cake ...

The man on duty in the opening speech was Reggie Hutcherson, the manager of the Sun Technology Evangelism group. He introduced into the conference. During his opener he borrowed a joke: he told about a guy in a bar the night before who asked an english Java evangelist in which group England is playing next year in the football European championships...the guy in the bar was me and the English evangelist was Simon Ritter. Not only Simon but about ten other evangelists, marketing and technical Sun employees I did meet in the bar drinking some beer and chatting about this and that. It was a nice evening ... but I think the last beer was stale: the day after I was a bit sick.

Nevertheless I visited many sessions and had many, many talks. One hotspot was a dinner meeting with James Gosling on Tuesday where several Java Champions and JUG leaders had the opportunity for an intimate talk with the Java godfather himself.

Tuesday evening about 20 persons (JUG leaders, Java enthusiasts, Java campus ambassadors) went out for some Bembel Apfelwein in a typical Frankfurt Apfelwein-Bar. Very nice ...

On Wednesday Dr. Heinz Kabutz had a talk about Java concurrency. Heinz wrote the prolog for my book as you remember. ... oh yes, and NetBeans 6 was released on the days.

The Tech Days were good and I thank Sun Microsystems for inviting me. From a community/developer view it would be even better do have some more attendees involvement. The Tech Days now were mainly driven by Sun and Sun speakers (do you know Sang Shin? he is cool). The involvement could have the form of an unconference where attendees more drive the content or at least some popular speakers enrich the programm.

So let's relax a bit in Cologne, do some IT work and refuel tanks ... next week is the next Java highlight: the JavaPolis 2007. JavaPolis is the biggest Java conference in Europe and in my opinion the best. I will be in Antwerp from Tuesday to Friday and hope to see you there. I will be e.g. at the JUG BOF, the Java Champions BOF ... the speakers dinner on Tuesday evening. Look for a big german hat to identify me. :-)

Sun Tech Days Frankfurt, 3.-5. December 2007

2007-11-29T21:53:00.001+01:00

Monday I will drive to Frankfurt for joining the Sun Tech Days, a nice opportunity to meet some old friends. I will also meet Dr. Heinz Kabutz there, the Java guru who wrote the prolog of my book. The conference has an estimated size of about 1000 attendees and is hosted by Sun Microsystems. I was the first who was informed about the Tech Days ... which were originally targeted for Berlin.

Beside some interesting sessions the community facet is even more important. Sun accepted my proposal to set up a "community corner". If you have been at J1 you know what I mean. I'm "the man on duty" at the community booth on Tuesday during two time slots. Try to find me. :-)

One highlight is also the meeting the James Gosling on Tuesday. Some guys like the Java Champions have the special opportunity to meet the godfather of Java for an intimate talk.

O'Reilly interview

2007-11-29T19:26:00.000+01:00

Today my publisher set an interview with me online on their main web site. Have a look on O'Reilly web site or directly on the interview site.

JUGC November, 30th: Sun Java Evangelists visit

2007-11-26T22:37:00.000+01:00

On November, 30th we once again have a Java Evangelist in Cologne ... no, we even have two Sun Microsystem Java Technology Evangelists who will talk in front of the Cologne community. The first talk is given by Carol McDonald about "Examining a sample Application built 3 different ways: 1) with JavaServer Faces, EJB 3.0 and the Java Persistence API, 2) Spring 2.0, JPA, and JSF 3) JBoss Seam". The second talk is given by Joey Shen about "Java SE trouble shooting tools in a nutshell". I look forward seeing you at the Ebertplatz from 6 pm on. Details here.

One night in Karlsruhe aka: Scrum at SAP, Siemens ... Guiness vs. JUG

2007-11-24T18:04:00.000+01:00

This morning I returned from my trip to Karlsruhe to the XPDays and the JUG-KA. I left Cologne yesterday at about 4.30 with my car and arrived at the Best Western Queens Hotel Karlsruhe at 7.30 am. I had breakfast and was waiting for Johannes Link's official XPDays Welcome.

My first session was "XP = legales (agiles) Doping für Projektleiter in Grossunternehmen" ("xp - legal doping for project manager"). It was very good showing some insight into an agile project at DBV-Winterthurversicherung. The most funny think was that they did admit implementing a performance bottleneck into their application consciously. The customer was pleased about the performance nevertheless ... and the development team removed the bottleneck in babysteps synchronously to adding new features.

The next session was "Why Agile Projects Fail" with Joseph Pelrine. Jiri Lundak, the second speaker, could not attend because of job reasons. You all know Joseph .. his performance is unique and the content brisk. He is so right that you must solve the hidden problems in teams, the casual challenges, not the symptoms. But: how seldom is this? Only the good guys, the good coaches know who to find the original reasons for conflicts. He also pointed out to try something out, to make a concrete (development) step, DO something instead of analysing. I like these lateral thinkers. There are so many agile people on the market who do not know what agile is, who read something and adopt this opinion withouth thinking ...

My second session was "Das Ganze sehen - Organisationsweite Optimierung bei Einführung von agiler Entwicklung" ("to see the whole - agile processes") at Siemens AG. Also this session was very good. It gave information about Siemens' strategy to introduce and work with agile processes (Scrum) in their complex, heterogenous company environment. They open stated that their timeboxing was not always timeboxing because of many necessary Siemens downstream activities.

I used the keynote time to take a bit sleep in my hotel room after my small three hours night. This hour was ok, but not enough.

Later I joined the session "Einfachheit in Softwareprojekten" ("simplicity in software projects"). Stefan Roock tried to be funny and it worked many times. The talk was more based on a show style performance and the funny slides, not on the content. One message was that it is not necessary to use a build server or a bug tracker system. "You must keep your project simple and these infrastructure componentes are overhead." I have to say, this is a nice thesis, but in practice ... ?!? Also the statement that the guys in one of his projects had daily problems day after day the same problem with working with Eclipse was questionable. He does not know if they manage the problem and reduce the daily waste time because he left the project.

My last session was "Was folgt nach Scrum bei SAP?" ("what happens after Scrum at SAP"). Christian Schmidkonz and Henrik Stotz explained that there have been many agile Scrum projects at SAP so far. For 2008 they plan to spread Scrum to more locations, to intensify and deepen the single Scrum projects and make agility to a company strategy. The also want to introduce TDD, accelerate the role of the "product owner" and retrospective ... but please: do the retrospectives in frequent intervals, and do them right, otherwise your team is dead (if it ever was a "team").

Overall the whole event was very good. The sessions were of high quality and gave a good status quo of the struggles and insights of consultants and companies.

After this first part some nice guys drove me downtown. I had the talk "Agile Developing of Rich UI Applications" for the JUG Karlsruhe in the university. Afterwards we went in several bars for some drinks finding the way between football fans. The Guiness in Karlsruhe is perfect, also this other local beer I drunk later ... I do not remember its name probably because of the Guiness drinks before. A nice journey and I'm happy that I had the chance had some nice talks to some very nice guys ...

November, 23th Karlsruhe: XP Days and "Java User Group Karlsruhe"

2007-11-16T21:50:00.000+01:00

On November, 23th I will make a trip to Karlsruhe. During the day I will join the XP Days 2007, the conference dedicated to agile development. My current plan is to join the following sessions:

"DSL am Beispiel Groovy" with Bernd Schiffer
"Agile Entwicklung einführen" with Jens Coldewey
"Das Ganze sehen" with Sabine Canditt and Jennifer Schiller
"Why Agile Projects fail" with Joseph Pelrine and Jiri Lundak
"Was folgt nach Scrum bei SAP?" with Christian Schmidkonz and Henrik Stotz

In my opinion the whole conference is nicely set up ... and I feel good about the output being in the program committee this year.

At 7.15 pm I will give the talk
"Developing Java Rich UI Applications"
for the Java User Group Karlsruhe at
Fachschaft Informatik an der Universität Karlsruhe (TH)
Gebäude 50.34
Multimediahörsal im UG
Raum -101
Am Fasanengarten 5
76131 Karlsruhe

The topic is similar to the talk I gave in Hamburg some days ago. BTW: the talk in Hamburg was pretty interesting. About 100 people were count there. Thanks to Björn, Dirk and Daniel for inviting me and organizing the happening. Not only the talk was fun with some nice questions, but also the time afterwards. We had some drinks downtown ...

Expect several copies of my book as give-aways in Karlsruhe.

November, 14th: Lehmanns book store, Hamburg

2007-11-04T22:45:00.000+01:00

On November, 14th I will follow an invitation and give a talk in the Lehmanns book store, Kurze Mühren 6, 20095 Hamburg. Host is Lehmanns and the Java User Group Hamburg. My talk will discuss an agile infrastructure to develop Java Rich UI Swing appliations. And beside that I present my book "Agile Java-Entwicklung in der Praxis".

"Individuals and interactions over processes and tools" is one value pair of the Agile Manifesto. Yes, communication is more important than tools, but this does not mean tools are unimportant. Far from it! Using the right tools right we will improve our quality significantly.

Starting with Subversion I will integrate TestNG, Jemmy and Fit to specify and drive the features. Complementary tools like Checkstyle for code metrics, EasyMock for test isolation and EMMA for code coverage optimize the process and the quality of our artifacts. We will use Eclipse as IDE and set up an continuous integration environment with Ant and CruiseControl.

If you will be there on Wednesday drop in, join the happening and say hello. I look forward talking to you.

Start is November, 14th at 8 pm.

Book in hands

2007-10-27T10:39:00.000+02:00

My book is out! I have a copy on my shelf now. Please expect your copy delivered the next days.

Further information, details and errata can be found here.

Question of the day: What is agile?

2007-10-21T14:52:00.000+02:00

What is "agile software development"?

(a) "Agile Development is a chaotic ad hoc development."
(b) "Agile Development is a product to sell. I'm a consultant and 'agile expert' and sell agile development although I do not understand the agile values and basic rules of successful together and team work. I'm the most important person in the project. I collect knowledge and do not share it as often as possible."
(c) "Agile Development means the whole team does 50 push-ups every morning before sitting at the computer."
(d) "Agile development is a set of rules and methods to increase the probability of successful finishing the IT project while having motivated engineers in real teams with respect for each other and helping peers, satisfied customers and relaxed project managers, managers and HR staff."
(e) "Agile what?"

unit tests: multiple instances vs single instance

2007-10-11T16:47:00.000+02:00

Often I shock participants of my testing seminars with the following question: "what is the result of the following tests running with JUnit?".


package com.huettermann;

import org.junit.*;
import static org.junit.Assert.*;

/**
 * @author Michael Hüttermann
 */
public class JUnitTest {

  private int count = 0;

  @Test
  public void test1() {
    count++;
    assertEquals(1, count);
  }

  @Test
  public void test2() {
    count++;
    assertEquals(1, count);
  }
}

The tests run through successfully! Why? For every test method a new instance of the test class will be created. count is 0 first place starting method test1 and test2.

Why is this shocking? Well, many people expect one test class runs a group of tests isolated. But with JUnit this isn't true. There one test (i.e. one test method in the test class) runs isolated. Interesting enough this approach is not unique for all test frameworks. Have a look on TestNG.


package com.huettermann;

import org.testng.annotations.Test;
import org.testng.Assert;

/**
 * @author Michael Hüttermann
 */
public class TestNGTest {

  private int count = 0;

  @Test
  public void test1() {
    count++;
    Assert.assertEquals(1, count);
  }

  @Test
  public void test2() {
    count++;
    Assert.assertEquals(1, count);
  }
}

TestNG creates one instance for the test class which all tests in this class share. So the test run outputs:

PASSED: test1
FAILED: test2
java.lang.AssertionError: expected:<2> but was:<1>
at com.huettermann.TestNGTest.test2(TestNGTest.java:22)
... Removed 27 stack frames

What do you think which approach is better? Why?

Agile, agile, everywhere ...

2007-10-10T20:17:00.000+02:00

My book is now listed by Amazon for being released October, 31th. The final version has 432 pages ... I think this is the best size for this topic -- not too thin like a newspaper, not too big like a blather tome.

The last days I gave some seminars. The seminars discussed tools and agile processes I have also written about in my book: firstly a seminar about agile testing of Java Swing applications using TestNG, JUnit, EMMA, EasyMock, CruiseControl, Ant and Jemmy. The seminar was good received and gave me some insight in interesting participant projects. Another seminar was Maven 2. In two days we set up a repository working with JUnit 3, JUnit 4, Checkstyle, JavaDocs, remote and local repositories ..., generated a customized site ... and had a lot of fun.

It is so important that you choose the right tools for your project. There is no "golden hammer", a set of tools that fits everywhere. Every project is different. In one project Ant may be best in another project Maven. One project may profit most from CruiseControl others from Continuum or TeamCity. There is not the one and only infrastructure ... but a tool set of very good de facto standard tools.

... and yes! "Individuals and interactions over processes and tools" is the first value pair of the Agile Manifesto. Humans are working in projects, different characteristics you have to bring together to build up a real team. A team where all respect and help each other, where is no room for vanity fair, for persons always wanting to be in the limelight. If you have a team and the right tools, you have a very big chance to finish your IT project very successfully.

book release

2007-09-30T13:23:00.000+02:00

My book "Agile Java-Entwicklung in der Praxis" will be released in October. The last weeks O'Reilly set the final print. I reviewed it, gave some last feedback and re-created several screenshots. I have to say the quality management of O'Reilly is absolutely excellent. Now the time is come where the book is 100% in the internal publisher process ...

JUGC: JavaFX talk October, 16th

2007-09-30T12:50:00.000+02:00

I managed to get Sun Microsystem's Java Technology Evangelist Simon Ritter for a talk about JavaFX for our Cologne Java community. On October, 16th 2007 Simon Ritter is visiting the Cologne JUG talking about JavaFX. JavaFX is a new Java Client technology announced by Sun on the JavaOne this year. Have a look on my JavaSpektrum JavaOne article. For more information on the Java technology outreach program have a look here and for the event have a look on http://jugcologne.org.

Sun Tech Days invitation, December

2007-09-19T20:53:00.000+02:00

I was invited by Sun Microsystems joining the Sun Tech Days in Frankfurt, Germany, December 3-5, 2007. Thanks for that ... I will definitely come and join the conference. I look forward meeting old friends and attending the nice sessions. See you in Frankfurt.

JUGC talk with Mike Keith, Oracle, August, 21th

2007-08-23T23:15:00.000+02:00

On August, 21th Oracle persistence expert Mike Keith gave a talk for the Java community in Cologne. About 65 persons attended the amazing presentation about EJB 3.0 and JPA. The EJB 3 co-spec lead devided the talk into two parts of 45 minutes: firstly he introduced EJB 3 and its benefits especially compared to EJB 2.x, the second part dealed with JPA. Many questions were asked and answered thoroughly. Mike's slides are here. After the talk I did a raffle distributing books, shirts, the IntelliJ IDEA license and some other stuff.

After the official part about 15 persons went to a local bar having fun and some more deep discussions. Many thanks to PROKODA for sponsoring the great location, finger-food and drinks including Kölsch on tap. Also thanks to Apress for sponsoring several copies of Mike's book, JetBrains again for providing a full license of IntelliJ IDEA and Mike Keith for his time. Mike is a very nice guy -- if you have the chance to talk to him at any conference, don't miss the opportunity. The next JUGC event will be in October. Have a look on the JUGC site for updates.

Closures for Java

2007-08-22T20:14:00.000+02:00

As you can also see here Closures for Java gains more and more supporters. Currently only Neal Gafter works on the Closures specification respectively prototype. Perhaps these three persons can help?

XPDays 2007: program online, aka: conference processes

2007-08-17T23:34:00.000+02:00

The program of the XPDays 2007 is arranged and can be found online here. The program is pretty thrilling and manifold. It contains e.g. talks about agile usage of Rails, Groovy, ABAP, talks about FDD, 37signals and a lot TDD stuff. I look forward hearing experience reports of SAP. My personal hightlight is "Einfluss der Programmiererfahrung auf den Prozess der testgetriebenen Entwicklung" -- an academic analysis of the dependency between programming experience and TDD. Very seldom that these kind of topics are analyzed funded.

I have to say I really enjoyed being part of the programm committee this year. On the other hand I still have shady feelings about the session election process of some conferences in general. Consider a talk which is only evaluated by some few guys. This is not really representative for the whole community. The censors have a big responsibility giving the rating.

Additionally, if there is a submission of a guy who is not very famous his submission may get a negative evaluation. If the submitter is well-known and knows guys in the committee he may get a good evaluation although his submission is pretty bad -- the censor may say "Oh, this session is proposed by Sponge Bob. I know him. He is funny. I don't understand his abstract, but it must be of value because Sponge had a nice session last year and he can drink beer like ten". Or censor Patrick says: "I do not understand Sponge's abstract! He writes he will present abc. But I do not understand the benefits of abc". OK, Patrick reads the abstract but does not anticipate the whole session -- only its topic and basic content. Is this a reason to vote against the session?

Ambivalent!

... I guess the program of this year's XPDays Germany is fair and exciting.

2007 Java Developer's Journal Readers' Choice Awards

2007-08-17T14:45:00.000+02:00

TeamCity (have a look on my current Javamagazin article) was recently voted to be the best "Best Team Development Tool" in conjunction with IntelliJ IDEA. Thrilling! IDEA itself was finalist in several other Java/JEE relevant categories, often better placed than Eclipse.

Beside that TeamCity was voted to be finalist at 2007 Java Developer's Journal Readers' Choice Awards in the category Best Java Project Build Tool. The number one build tool is Ant. Other finalists are Maven and JBuilder/TeamInsight. That is pretty interesting: Ant and Maven are basic technologies behind the scene which can also be used by TeamCity. Other CI servers like Luntbuild or CruiseControl are not mentioned. Be aware that the nomination is based on TeamCity 1.0 and meanwhile TeamCity 2.1 is available which is several dimensions more efficient, function-rich and usable than TeamCity 1.0.

Good work JetBrains! One year of work on TeamCity and now providing one of the most thrilling and most appreciated build servers! And IDEA is highly appreciated too. Look at the current milestone release of Version 7.

Agile projects vs. non-agile projects

2007-08-15T21:10:00.000+02:00

One point I mention in my book is that projects often call themselves "agile" but are everything but agile. In agile projects the whole team has respect to its members, and communicates in an advisable form -- no bubble! "Agile" is not a "product" to sell, agile is a basic attitude. All people do want to archieve the best for the projects, ask if something isn't clear, anti-patterns like a collected in my Wikipedia article, are seldom. Selfish behaviour is a no-no as well as people who always got to be in the limelight.

If you do want to learn more about the best approach for agile teams, when a so called "agile team" is not an agile team and if you want to read about the concrete advantages of agile processes ... please read my book. It is also of value if you want to learn about the infrastructure, the tools to use in order to support an agile process. Although agile projects do focus on basic values also tools are very, very important in order to accomplish the daily tasks (like Test-Driven Development, Continous Integration, testing of components, functional testing of Java Swing applications, functional testing of web applications, Software-Configuration Management, Standards and a lot more ...).

TeamCity article online

2007-08-15T21:02:00.000+02:00

My article about TeamCity (I have already mentioned it here) -- which is published in the current Javamagazin issue 9.2007 -- is now available as PDF.
Edit: in case of interest: my former IDEA 6 article is here.

JUGC: August, 21th - EJB3/JPA with Mike Keith

2007-08-10T18:32:00.000+02:00

August, 21th the next JUGC event will take place in Cologne -- the second happening in this month. I have managed to get Mike Keith from Oracle to Cologne, co-spec lead for EJB 3.0/JPA. He will speak about “Building Java Applications using EJB 3.0 and JPA”. Have a look on http://jugcologne.org for details.

Handover JCP EC chair in Munich, 08/17/07

2007-08-06T19:08:00.000+02:00

I was invited by Sun Microsystems to join a JCP EC event in Munich, August 17th. Thanks! There will be the official handover of the JCP Chair from Onno Kluyt (who led the community since July 2004) to Patrick Curran. Unfortunately I do not find the time for the trip to Munich, what a bummer. But: I will make it up and travel to Munich later in the year for visiting some friends and the Augustiner-Biergarten. Who lives in Munich and wants to meet, please drop me a line.

JUGC event, August 3rd: SOA and Ajax

2007-08-04T20:59:00.000+02:00

Yesterday the recent Java User Group Cologne event took place. It was a conference with three separate talks. Starting at about 5.30 pm I opened the event finding some warm words to the audience and the location sponsor: Deutsche Telekom.

The first talk was given by Dr. Arno Puder, Professor at the San Francisco State University, and his free framework core team consisting of Wolfgang Korn and Sascha Häberling. They presented their Ajax framework XML11. An impressive talk! XML11 allows you to write Ajax applications in Java, compiling them and migrating the Java bytecode to Javascript code. Many questions were asked to the presenters during their 60 minutes presentation.

After a small break for drinks, beer and pizza Owen Taylor from GigaSpaces was the next one talking about Gigaspaces' "spaces"-solution for a SOA-driven system architecture for linear scalable application distribution. Also this talk was escorted by many questions. Because of the thrilling content and these many questions I extended the talk duration from 60 minutes to 90 minutes situatively.

The last talk was given by Max Antoni about his EvaServer. Max is a Ajax geek and he gave excellent insight into the world of Ajax and his own Ajax server.

I awareded the two guys who asked so many excellent questions with a special price. Additionally I distributed some books while asking the audience some questions.

The conference finished at about 9.45 pm after over 4 hours 30 minutes. Wow. I was at the location at about 3.30 pm for supporting the Telekom preparing the seating, micro and this technical equipment. There was not so much work left, they already set up the location pretty well. Thanks! The location is very, very comfortable and suits perfect for such events. Although the JUGC was at the Telekom building last summer already for a talk of Dr. Heinz Kabutz, meanwhile I normally host the events at the University of Cologne.

One intention of the event was having fun again. And it was fun .. meeting the people and drinking some ice cold Kölsch beers. All my work organizing the event paid off again. GigaSpaces gave me a budget for organizing finger food and drinks. On the last day (!) before the event I found someone having a car who was willing to bring the drinks from a local shop. Currently I do not have a car ... thanks, Max! The community does also say thank you to Ulf for bringing the boxes and ice. It took him until the event day to commit on this ... but worked luckily and Max and Ulf, you got the money back directly. Ulf, thanks that you skipped your request for being paid for the fuel. If I would have asked the pizza service for drinks too, it would have been more expensive. And ice boxes are necessary: warm Kölsch is an Antipattern. After the event we went to a local bar having some more fun.

Hosting a JUGC event on a Friday was an experiment. Normally the happenings are during the week. I'm not sure about the result of the experiment. On the one hand about 35 attendants are a sign that it crashed (normally there are up to 100 people), on the other hand yesterday also a big city party was in Cologne, warm weather and still holiday time. The JUGC workshops last year which were also done on a Friday were pretty good received.

The next event will take place at August, 21st. Then Mike Keith, co-spec lead of EJB3 and working for Oracle, will give a talk about EJB3/JPA. Currently I'm not sure where I will arrange the evening. Please have a look on jugcologne.org for updates. Two events in one month -- meanwhile I feel like a full-time event manager.

XPDays 2007

2007-07-29T17:52:00.000+02:00

The XPDays Germany 2007 will take place November, 23th in Karlsruhe. It is the leading conference in Germany dealing with agile topics and the sister of other XPDays in other countries. This year I'm part of the program committee and currently evaluating the submissions.

BTW: have a look on my Wikipedia article about Extreme Programming. I wrote the article last summer with the help of a hand full censors. As I started the article it was something like a draft, some few sentences. After my editing it received the predicate "worth reading" and is a very verbose german essay on the topic -- I did not see any other article until now which is free as well and contains a similar complex discussion.

It was really helpful to get feedback by people who normally work in totally other disciplines. I remember a lot of nightly discussions with people who do not have any technical skills at all. Thanks Wikipedia.