Mails on cocoon-dev offer links:

• What's New for "Office 11" Developers?
• How open is the new Office.

Coming from IDEA I wouldn't settle for anything less than the 2.1 of eclipse, even it was still shaky at the time. Real men should not be afraid of any latest and greatest, only the last 2.1-RC1 broke my activity on the hibernate testing quite a bit...While everything worked in plain ant and in eclipse 2.1-M4, it would give me a quite odd 'ClassCastException' on the interpretation of a in the newest release.

After some needless looking into maybe incompatible jars and classpaths I went into stupid trial and error mode with some succes. I added my workarounds finally in their bugzilla and am now starting to wonder if a task like shouldn't always be inside a target. Stating that it just works in plain ant is maybe not the real argument? Have to check some ant code and doco's I guess. It shows in every case that the ant-build-file-parser inside the eclipse-ant-view is redoing (differently) stuff done already by the ant run-time...

The good news on the RC1: //TODO now adds directly the item into the Tasks View

One last thing, I promised Tom I would praise his eclipse consulting skills.

*update* snap eclipse developers have found the bug to be somewhere in their fast-parsing for targets that is in their ant-view thingy.... So trying to find a CCE reason I guess they just try to map all direct childnodes of to be of type or something?

We have the saying "goede wijn behoeft geen krans" -- something like "Quality doesn't need appraisal"

Yet, this feather will be worn with pride, I'm sure.

And shamelessly joining in the pride: This makes the percentage of apache-commiter-employees at outerthought jump from 66% to 100% again :-)

I guess this is a time where he would consider maintaining a blog :-)

more then words...

This is just fun: a new way to get to blue-underlined-words-I-can-click-upon! (gotcha ;-))

These got my immediate attention:

• findU operates some network of weather stations. This one is active in determing my mood.
• and some else is sharing interests (geo gadgets | calvin and hobbes | war opinions | XML | and a blog)

By the way, mozblog nuked one of my blogs again, have to take a look to this nucleus thing.

It only got better... read this and weep. Quoting only the yesyesyes-nodding-part:

A project requires:

• A Vision
• A need fulfilled (the itch)
• Users that will like "the way" how it works (for one reason or the other, be it simplicity of design, economy of resources, stability, cost, or even features ;-)
• + Management to keep it evolving and avoid it falling into pieces in the contrast between the needs of the users and the needs of the design
• Some "brand" that makes it easy for people to remember it exists
• Documentation
• A evolving developer team to fix bugs, document and re-structure code on a continuous base
• ... and a bit of code

The rest (I didn't quote) is downright helpfull and newbie supportive stuff: "way to go".I hereby declare 'Santiago Gala of hisitech fame' as my hero of the day.

I think therefore I do those activities somewhere... The statement can be verified officially, you just need to come over to

First tests coming through... Promosing, very promising. Well, current achievements (many-to-many link) isn't like real world-shocking news, only the way it is letting itself be handled by the developer is quite refreshing. For the first time I get the feeling that mapping the O to the R doesn't need to be, euh like hard work (CMP never gave me that feeling, but I wasn't using xdoclet then, maybe that is the secret)Had some work on the xdoclet side by the way... Lucky to have found (i.e. have Steven, the web-indexer, find me) someone that went ahead in the xdoclet-integration for Hibernate2. Since I wasn't coming from the 1.2.3 stuff there actually was one very relevant piece in there for me: (dig into the xdoclet-hibernate-module-jar for this)

Further updates to hibernate-properties.xdt to replace

paramName="role" to paramName="name",

also replaced paramName="readonly" with paramName="inverse".

Which exposed me to more xdoclet internals then I was looking for. But the surprise didn't make it less intriguing. Mmm, I wonder what Steven would say about the XMLism:

 

 
  paramName="name" default=""/>"

 

 table="" 

 

Someone sure doesn't like the -type constructions :-)

I saw this one coming...But it came out so much more beautiful then I could of have imagined.

I cannot "compute" finished + software together.

software is like a live creature, more so a project.

You are really starting, now that you are planning

to have "true" users and go out to "the real world" ;-)

I had to play all 40... :-( - thx to Mr. Bruggen, our favourite Novell Employee, for pointing out its existence.

And while we're confessing: I occasionaly do fancy a friendly game of Go. (And before the hardcore challenges come in: I'm at the level of knowing the rules, and enjoying the game)

Which actually opens up to my other pass-time: currently reading the "Alan Turing: the Enigma" (see here for some online parts) The book is utterly thought-provoking, and nice to read, so just do it. (The paperback costs just about the price of the paper and the ink in fact)

The advent of this little guy, made us all wake up a bit, and learn how to put things in perspective... (it even devaluated (not-that-)regular blogging)

Welcome Jef...

*note* for those only looking back for the score: the previously stated 5/10 is now becoming 6/12 (2 more happy couples acclaimed fruitful activity) - and for those that cannot relate the name Jef to any gender: more male dominance (5-1)

*note2* we got to see Saar in real life again this weekend, something tells me she can handle the 'load' and we should look for another word then 'dominance', uhuh.

*update* some congratulations in my mailbox confronted me with the fact that there were some mis-interpretations possible :-) Jef is the 2nd son of my wifes' brother, which makes it close enough to get a hold of the close-up-picture, and to be working on our emotional drive...

Read the hibernate doco's this weekend, will be doing some tests today.... Sure looks quite complete from where I'm standing. And much to my surprise, I got to notice interwovenness with loads of apache stuff I didn't know about. (Taking Sylvain's image: I urgently need a fourth eye or at least glasses for the others :-))

Some of this stuff could be usefull for xreporter as well:

• hibernate design did away with java.lang.reflect.Proxy in favour of using CGLIB
• they use the jakarta-commons-connection-pool

It was my first time. I got to meet him. Not just by accident, it was a carefully organized (public) setup.The speech was not on the 'free' thing, although he took the advantage of being on stage to introduce the concept (to an audience that was of course aware already.) Nice to see that there were quite a lot of university professors in the room as well, might be a sign of the principles getting into more of 'the established' layers of society.After the speech, the well-prepared student-organization suddenly focussed on the issue of planning 'what was to be after the speech?' Which basically meant that (as one of the few car-owners there) I got to drive rms to his following dinner appointment...

During the 15' drive we easily ended up stressing the differences between the hard-line free GNU vision, and the more pragmatic open Apache vision. Basically he would appeal to my 'if you have any feel for the issue of freedom' (which he knew I had, but it is more oriented to data-ownership then to algorithm-ownership) So we ended up not really communicating when I opposed 'isn't choosing no to do so also some fundamental expression of freedom?'The good news is that there is a lot more alignment in thinking then there is difference, so we concluded with the 'happy hacking' -goodbey anyway :-) And surely we are on the same side opposing the European pattent laws. (the topic of the speech)

All in all, the straight thinking he applies is something I do admire (even if it's not completely mine), and his fysical presence (motivationally glowing eyes included - he needs to be close up in your car before you see this) is one that carries natural authority. He has an (euh,) 'own' way of behaving on stage but the tone of voice, clear argumentation, didactical build-up and overall reasoning in answering questions make it a perfect easy-listening and thought-refreshing experience, grab the occasion if he stops by near you.

Talking about it again, but not alone, so I can also sit at the back and get some work done...

And I get to note how funny it is to have English spoken seminars here in this country (as the compromise between the French, Dutch and German official languages) I'm afraid I never hear my own pronounciation mistakes (or spot my own English typos), but my co-speakers on this event are making some funny ones:

• When 'jay' becomes 'djee' (as in G2EE), it has more of an orgasm-ring to it :-)
• When WSDL is colloquially refferd to with 'wistle' it's really not that far off from 'weasel' (upcoming Scott Adams drawings?)
• When META-data becomes MEAT-A-data, you get to understand better what is 'on the bone'. (He is persistent on this one!)

Look who's into python too :-)

I thought I recognised something about it.

Must say I'ld rather have Joe, his house-mouse and his dinner-plate be the subject of the social experiment :-)