Een huis ziet er toch beter uit met ramen, mijn gedacht ;-)

Zo ziet ons huisje er tegenwoordig uit: met garagepoort !

Nog een beetje wachten, en dan komen onze ramen – eindelijk …

‘k hoop dat de plakker dezelfde mening is toegedaan is, maar het ziet er alvast goed uit ;-)

Zo ziet onze living/eetkamer er uit (althans voorzover op de foto’s past):

Zoals je ziet is er al flink geslepen en gekapt.  De potjes van de elektriciteit zitten er ook allemaal in.

De zolder zag er iets lichter uit zonder dak, maar dit is ook niet mis ;-)

De balken op de grond worden gebruikt voor de vloer opbouw – hier komt de isolatie tussen.

In de keuken tieren de stopcontacten welig …

My first collector ;-)

Deze zijn voor de CV boven – er hoort nog een ontluchter op de onderste collector, maar die plaatsen we pas als de alupex hangt (kwestie van wat handig te werken).

De verbinding rechts is de bypass.

Of niet?

Om mijn stelling te staven: enkele foto’s :

Nog een laatste inspanning en de metselwerken van onze woning zijn voltooid.

Hierbij een foto langs straat- en tuinzijde:

Intussen ben ik zelf gestart met het inslijpen van de leidingen voor sanitair en elektriciteit – wel een stoffige bezigheid, maar het gaat goed vooruit ;-)

Volgens planning komt de dakwerker volgende week langs om de timmerwerken voor de schuine daken uit te voeren. Het onderdak wordt dan ook direkt geplaatst, zodat we wind- en regendicht zitten (ik zal blij zijn als de daktippen niet meer zo los staan)

De aannemer heeft goede progressie gemaakt deze week: de welfsels van het gelijkvloers liggen er op en buitenmuren van de 1e verdieping zijn gestart:

Zicht van de straatkant – het linkergedeelte zit wel goed weggestopt achter de kraan:

Vanuit de tuin;

Zijmuur met de raampjes garage, raampje berging, deur berging en het raam in de keuken.

A little story:
Sunday evening, just finished Katrijn’s birthday party.   I open up my laptop and want to publish a few photos of our house and of the girls.

Question is : where?
I have a Windows live space, a LinkedIn account, a Plaxo account and a Facebook account, etc.
You can’t publish everywhere, right?

I’ve never had the ambition to be part of any of these networks, except for this space.  The only reasons I have an account on all the other networks is because someone invited me to look at their pictures.  Since you need an account to view other account, you need to create your own, etc.
Once you’ve created your profile, you’ll get invitations from all other people you may know on that network and the ball starts rolling. 

Funny thing: your connections are for a large part the same people that are also in your network with the other networks !

Sound familiar?

My opinion : it’s about time to get some interoperability between social networks.  I want to be able to publish with my tool of preference on my site of preference and still be able to connect with others that have another preference.
I don’t want to do this x times again!

Small chance this will happen in the near future : money makes the world turn round, and the benefits of a proprietary network seem bigger for the companies than having the most user friendly, open networking systems.
Perhaps Microsoft should simply buy all competition?? (that might be the fastest way to solve this ;-)

Conclusion of my personal dilemma: I went for my blog – at least that’s the network that I choose to be part of myself !

Na een korte ;-( vertraging is onze aannemer in gang geschoten met het metselen van de snelbouw. Na een dagje of 4, krijgt de benedenverdieping alvast vorm.

De voorgevel:

De achtergevel:

Keuken en linkerzijgevel:

Eetkamer, gezien vanuit het salon:

I"ve been ill for most of the past week, and that gives a man time to think…

About my blog for example ;-)

And the fact that I haven’t been publishing anything whatsoever.

Not that there’s no content to share:

• Plenty of content from the kids and the undertaking of building a house --- but there are other and better media to inform the people we know and love.
• Lots of interesting stuff in the corporate area too: but we have other means in the company to share knowledge.

So I come to the conclusion that I’m just not the kind of person that shares his last and every thought with all other people on the net.  At least that’s more honest than seeking excuses like “no time”, etc.

But still, the advantages of a global space are obvious: in my private and working live I’m happy to rely on the net for information, buying stuff, contacting people, etc

So, I guess I shouldn’t give up on my blog yet and therefore this post.

Having similar experiences? Don’t hesitate to react.

What: an interoperability stack for Java - .NET3.0/3.5

What's in it:

• Web services with reliabiltiy (guaranteed delivery/order of delivery)
• Transacted services
• Security (meaning more than SSL)

It's built for Sun's GlassFish appserver, but runs in Tomcat too.

Microsoft already released .NET 3.5 with the latest version of the spec.  Sun is expecting to release somewhere in June.

Good work, but little chance I'll be using it soon...

More info on:

https://metro.dev.java.net/

A presentation with a few answers to real world problems: nice!

I'm not going to copy the documentation of unitils (http://www.unitils.org/) here, but do would like to point out one recommendation:

If you want to have efficient datadriven tests, then you should disable your foreign keys and not null constraints...

Not a stupid idea - I'm going to try this out.

These 2 talks explained practical experiences with a project using Scrum.

Tip for (at least one of) the speakers: don't read your text from a piece of paper.

Having had some Scrum experiences myself, I'll stick to the things that were worthwhile writing down here:

• Achievement is not the same as activity (personally I prefer "doing your best is not the same as having success", but I guess that's a matter of personal flavor
• Sprint backlog with 4 cols: todo, in progress, funct test, done (we didn't use the functional test stage)
• use a burndown graph to maeasure progress
• determine your velocity to  estimate/ follow progress
• organize sprint retrospective meeting (no more than 90 min)
• do the scrum meeting standing (some people aren't gonna like this one ;-)
• Play planning poker to estimate, using the technique of story points and only fibonacci numbers (are a perhaps a variation?)
• If you can't estimate (SPIKE requirements): organize a poc to be able to estimate afterwards
• try to automate functional tests (Selenium framework for web apps - I'm definitily going to try this)
• non-dedicated resources decreases efficiency (I've this happening too often...)
• scrum master does the fire fighting, especially in the first sprints (well, what should I say - if it's only a few sprints, that's GOOD)
• Putthe team together (I agree, this really works)
• SCRUM does not solve your problems, it only makes them visible (but you must be willing to see)
• Scrum doesn't mean you don't document, it means documenting in a smart way
• Empower the team and take a distance (beware: the team must be ready for this)

This guy knew what he was talking about.

Why the hell do something about dependency mgmt?

• be able to replace components with minimal effort
• better quality (dependencies result in bugs)
• enable faster change  of the product

Tools can help:

• textual/graphical analysis tools don't really help (not usable)
• what does work: DSM (Dependency structure matrix).
• The DSM tool in IntelliJ IDEA looked impressive : scales well, looks good

How to organize DM:

-> Define the rules and make sure everybody sticks with them !

What can you do about them once you know you have them:

• move code
• extract interfaces/superclasses
• copy (may be ugly)
• pull members up
• introduce parameters
• replace constructors with factory methods
• Service locator design pattern
• Use inversion of control
• Dependency injection

I have my concerns about these last 2: you still have the dependencies, but instead of having them at compile time (in code) you have them at runtime (eg in xml when using Spring)

Not a bad talk. this is what I took away from it:

• SOA is technology agnostic
• SOA is about loose coupling, not about distributed RPC
• SOA infrastructure should support/encourage SOA principles
• The ESB glues other technology together, allowing an incremental approach to SOA. Don't try to do a big bang!

It was a first view of JBossESB for me.

Apparently it's a combination of several tools (not uncommon with ESB's): Brools, JBoosWS, JBossTS, JBossMessaging en JBPM

It's nice that it can run in the JBoss appserver, but also in other appservers.

All in all, my impression (not based on detailed research) was that the product is not very mature yet. Illustrating this:

• failover/clustering technology is still under construction
• whole modules are scheduled to be rewritten/drastically improved next year.

Compared to another ESB that I know a bit better (Sonic ESB), this product seems to be not of the same standard (yet)

Presentation by Scott Ambler

I've been interested in/using agile techniques for several years now, so it was fun to hear somebody with an open view commenting on the current state of things.  This is what I took away:

• Agile seems to be getting mainstream.
• Co-location improves chances of a succesfull project, and offshoring decreases those chances.  This probably doesn't mean you shouldn't offshore, but that there's work to be done.
• Agile allows to change what determines 'success': it doesn't have to be "on time - on budget',  but could be eg best quality - best ROI - etc.  You should ask your stakeholders this question before you start
• Agile doesn't really say anything about the lifecycle outside the development part, but you have to do these things! (setting up the project, deploying, migrating, running production, etc).  We should think about this!
• Scaling Test driven dev: only does the details, but not the global picture. What about modeling, business tests, usability tests, integration...?
• What about agile documentation?
• How to handle the data in agile? How to apply the agile techniques to the database ! www.agiledata.org.  He really has a point here: developers should know what a db can do and use those features when applicable!
• lean development governance - www.ibm.com/developerworks.  Enable motivation of people by making things simple. There's many more on the site - boy, this guy knows what he's talking about....
  "the site might give you some ammo to steer things in the right direction' I want to have that document and forward it to my boss ;-)

This one took a bad start : It took 20 minutes before the first speaker finished to get past the blabla explaining what his company is doing.

The second part about 'location technologies' was a quite interesting.  Some stuff to remember:

• If you want to get a location inside a building, GPS doesn't cut it. You can use info from the mobile phone network to get a location based on the network cells.
• JSR-179 : can be implemented for both mechanisms
• Microsoft has a slightly deviating mechanism in Windows Mobile
• OMA: location deterministic services standars
• OGC provides API's to use GIS geospatial web services (over XML/HTTP) - map rendering, geocoding, routing, POI searching for usage by developers
• AutoDesk has an open platform that implements these api's
• JRS-293: not finished yet, will provide a standard API

Pricing of the Autodesk services/Navteq data: free for dev/demo purposes!