Outer Web Thought Log
July 31, 2003
Picture gallery

If you like to put faces on names and can't cope with tiny thumbnails, here's some nice pictures of java.net involved gals and guys. Justyna's picture is really, really nice.

July 30, 2003
Cocoon GetTogether CFP
I just sent out the call for participation for the 2003 Cocoon GetTogether, to be held on October 7th, Ghent (Belgium):
Dear all,

with only two months left before the 7th of October, it's about time we
start planning and setting up the next edition of the Cocoon GetTogether.

For those of you who are new to this, please have a look at the previous
event website at http://outerthought.org/cocoon/gettogether/

As being announced before, the 2003 GT will be held in Ghent (Belgium)
again, in a different location though, smack in the middle of the city.
We were quite ambitious while prospecting a new venue, which means we
have a large room to fill. This year's event will be organized by Orixo
(http://www.orixo.com), but most and foremost will depend on your
participation and presence. Stefano has also agreed to be involved
content-wise, so again this will be a great occasion to meet some of the
core Cocoon developers in real life.

We won't be changing too much from the previous year success formula (we
had 100+ registrations at that time), although we hope to learn from the
experience and tweak things, hopefully also depending on your input.

This mail serves as a heads-up for both potential participants, speakers
and sponsors.

What are we planning to do?

* Hackathon

The day before, October 6th, Outerthought will be hosting a hackathon in
its offices. They/we have a nice meeting room with (Wifi) Internet
access that should be able to sit somewhere around 15 people. We would
like to see a number of 'guided' design sessions, hopefully focussing on
some big challenges ahead, prepared by active developers who want to
have some high bandwidth collaboration around a whiteboard.

The idea is that somebody prepares a brief intro presentation to focus
thoughts and ideas, and that we put some serious time in tackling the
issues at hand. Subjects like 'Real Blocks', 'Caching and Performance',
'WebDAV' or 'Flow' surely warant some quality time together. Anyone who
feels like up to preparing and shepherding such a session (which might
last for a good couple of hours), please speak up.

Outerthought will be sponsoring lunch and snacks during the Hackathon to
keep brains running and blood sugar level high.

* GetTogether

On the 7th of October, we'll be gathering in 'het Pand', where we have
rented a large conference hall of the Ghent university, nearby to a
closed car park and in easy reach of public transport. Also close to a
couple of 'Ibis' and 'Sofitel'-class hotels, and in walking distance
(15') of the Ghent Youth hostel for those running on tight budgets.

The GetTogether itself consists of a number of presentations, with
plenty of time (breaks) in-between to mix and intermingle with fellow
Cocoon developers and users. The presentations will be deliberately kept
short, so that we are able to provide you with an interesting mix of
topics, and plenty of time to discuss and interact.

Some surprise 'acts' are also being thought of, so we'll try and make
this an lively and not run-of-the-mill day for everybody.

People who want to present during the GetTogether are kindly requested
to send in a presentation proposal. Mind you that presentation topics
will be screened for cluefulness, in the sense that we don't want any
commercial presentations. Introductory, tutorial-like stuff into new or
advanced topics are much appreciated though. Please send in your
presentation proposal to gt2003-speaking[at]orixo.com - see below for a
template to help you in filling out a proposal.

* Social events

As with the previous edition, there will be hosted social events on the
night of the 6th and the 7th. If you need to travel to come to the GT,
please consider also to extend your stay in Ghent for a couple of days,
since it is quite a nice city for tourism as well.

* Financials and sponsoring

The previous edition learned us that simply breaking even on sponsors'
budgets only proves to be near to impossible (unless some of you know
where the money bags are at Sun or IBM ;-)

Therefore, with much reluctancy and many considerations, we decided to
ask for a small fee covering the cost of this event. While the exact
amount hasn't been settled yet, it will be in the range between 50 and
100 EUR. The GT organizers are very much committed to make this year's
edition to prove much bangs for these bucks.

Sponsoring the event: In the past year we were approached by several
large companies that use Cocoon about sponsoring the GetTogether. We
welcome sponsors and feel that this is a chance for companies to give
something back to the Cocoon community. Due to the nature of the event
however, sponsors will only have the possibility of placing a logo or
similar on the introductory presentation and hanging a banner or similar
in the location. Also, the social events, handouts and surprise acts can
only get better with some sponsoring. Of course, sponsors can also
remain anonymous. Please contact info[at]orixo.com for sponsoring the

That's all for now. We are working on the website (help is still welcome
though) and registration procedures, so expect another heads-up soonish.
We, Orixo, would be delighted in seeing you show up on the 7th of October.



Do not include proprietary or confidential material in your proposals.
We will assume that you do not consider any material included in the
proposals to be confidential. All conference presentations are
vendor-agnostic and should focus on technology. Your proposal should
contain the following information:

1. proposed title for the talk/session/debate.
2. brief outline or abstract of the talk/session/debate
2.a Please be as concise as possible.
2.b Please write in third-person context.
2.c Please list any prerequisite sessions or knowledge you feel is
2.d Indicate, if applicable, if this is a beginning or advanced talk.
3. For all proposed speakers, include: speaker name, title, company,
address, email, phone number, fax and website.
4. biography for the proposed speaker(s) showing relevant experience and
qualification to speak on the proposed subject matter (not to exceed 250
5. Primary contact/founder name, title, company, address, email and
phone number.

DEADLINE for proposal : September 7th, 2003
Proposals should be sent to: gt2003-speaking[at]orixo.com.

All speakers are required to submit copies of their presentations at
least 2 weeks before the conference. These presentations will be
included in the conference notes. All presentations will be made
available after the conference in electronic (PDF) format.



</Steven> - GT 2003 co-organizing puppet
Steven Noels http://outerthought.org/
Outerthought - Open Source, Java & XML Competence Support Center
Read my weblog at http://blogs.cocoondev.org/stevenn/
stevenn at outerthought.org stevenn at apache.org
July 27, 2003
On temporary holiday leave
3000+ new mails, several hundred weblog entries, all neatly stashed away in my collection of Mozilla mail folders and SharpReader awaiting some idle time to surf the news I missed during the past few weeks. Nothing very unusual, so it seems, besides AOL finally pulling the funding plug of Mozilla/Netscape - maybe a good time for this organisation to try and stand on it own feet and finally become a real open source project. The Mozilla Foundation seems like a good start.
Guido left Zope to join a security software-related startup (which might considered to be a good case for Python, since Zope Inc. and all its various interrelationships with the various Python open source efforts always seemed quite murky to me - too much control originating from a single commercial entity and too many people on the same payroll). The CFP for ApacheCon in November is open (and I'm still considering what I will send in as a presentation proposal), and the flow & forms discussion in Cocoon-world will sure get plenty of attention in my catch-up mail reading schedule during the next few days. Some very cool community efforts are happening in Cocoon, and I am sure the fall and winter of this year will bring Cocoon several steps closer to an ideal platform for serious webapp development. It's good to see that my repeated bugging and nudging of my dear colleagues to get working on a serious form handling alternative is starting to pay off, both in the sense of community adoption and also business-wise. If you are reading the Cocoon developers list, I'm pretty sure the Woody-related discussions have come to your attention, and I feel rather proud of this, if only because of motivating Bruno to come up with a working prototype, finding an early adopting customer for funding continued development of it, and coming up with Woody's name. My contribution to kickstarting all this was largely peripherical, but nevertheless I feel a bit like grandfathering, and I'm quite used to this grandfathering thing anyhow. Anecdote ahead. ;-)
It took me almost two years of lurking and posting stupid remarks on the Cocoon list, while working for my previous employer, starting somewhere in 2000 IIRC, to finally convince Marc, who was also working for that same company, to give Cocoon a try for a real customer project. It got him, and later on Bruno hooked on it, and the rest is (slowly becoming) history. With the same patience, I've been hinting Bruno since the end of last year to start working on Woody - let's hope that will become part of history (and Cocoon's core stuff) as well. I was especially proud seeing Marc being nominated as a Cocoon committer as well yesterday. Some people are making fun of this, since Outerthought is only a three-men company, but it feels kinda good to state you are a Cocoon competence center, and being able able to back that by the fact that all three of us are effectively committers on the project.
I'm on a temporary holiday leave, which means I'm working this week, but will be back on holiday (but not abroad) next week. Plenty of stuff on my plate however: the Cocoon GetTogether 2003 needs to be put on track, the Cocoon project needs its own, long-overdue project guidelines (basically rules and regulations), and then there's some customer's business to take care of, and I should be starting on a larger venture which will keep me quite occupied for the remainder of this year and possibly some time after that.
July 17, 2003
On the move

Just a quick hi to feedreaders from the Radme Telecom Museum somewhere in Bretany.

July 10, 2003

I am leaving soon for holidays to Plouaret (no real website found), Ctes d'Armor,
Bretany, France. No Net connection, not even a laptop: I'll switch to the 'medium quality' setting of my digicam so that one 256 MB CF card will last for a couple of weeks. Cheers!


Sam has noticed Adam taking a stand. Too bad Sam doesn't live over here, since he also would have been amused when watching a talkshow on local television yesterday night. Guests amongst others were Adam and his wife Patricia. When Adam isn't the supposedly-to-be A-list blogger, he's actually just a showbizz guy. Fair enough, my wife nearly jumped out of the sofa when seeing him on television, since he is that kind of guy she was collecting photographs of when she was like ... 15. Mind you, she's 33 by now. At that time, Adam was presenting a very popular Dutch TV music show called CountDown primarily aimed at teenagers. Adam faded away from (our) public consciousness when he moved across the big pond to mtv.com, so we were surprised to hear last night he's actually coming to live in Belgium, quite assumably because of the interesting tax status for Dutch people living in Belgium.
Adam has its own reality TV show in genuine Osbournes-style, and here in the Benelux is currently most talked about for some enquiries from the Dutch state tax departement on the legality of some of his business ventures. The couple appears in glossy magazines and all that. It's not quite Victoria and David, but they are really really trying hard.
Basically, it's sad to see Dave supporting people trying to cultivate a cultus around their so-called 'personality' by buying default subscriptions in Radio. But I'm sure it happens all the time. What would be really sad, is that the people involved in (n)Echo/Pie would take Curry's stand serious. I'm pretty sure they would just laugh and continue their work when somebody like Britney or Jennifer or Kelly would make a similar stand. As Kelly would say: Adam is screaming for attention, so come dig him out. He must be sunken deep in the Silicon Polder.
Don't bother with such 'stands'. It's about time we're seeing the end of the dark RSS tunnel. And if we can't break with the past, why would we ever want to live in the future?

July 09, 2003
Conference Presentation Judo

This was new to me and I had a good laugh. If you regularly present in front of other people, go and read it.

Business values and open source

While driving home this evening, I was still thinking of some mail I sent earlier during the day, to someone who is looking into establishing an Open Source Center to be used as a political lobby organ and a means to dissipate information about Open Source and business to Flemish companies. My mail was very close to a rant, stating that we do regard ourselves as being quite unique in our little corner of the world, since we do not only use open source software in an effort to create more cost-effective solutions for our customers, but we also build, preferably with the customer, open source software.
Of course, we are lucky to have clueful customers, who know IT application frameworks most often are not at the very core of the data that runs their business, or will only enable, but not capture the domain knowledge and business processes which differentiate them from their competitors. Building a database reporting framework is child's play compared with amassing all customer's data to fill that large datawarehouse so that you can actually run those reports. It's the data and business flow/logic that counts, not the infrastructural technology.
This kind of reasoning is luckily quite orthogonal to our line of thinking. Either the customer cares about open source or not, but in the end he just wants a framework that enables his business processes. We, and many other concullegas just as well, are able to provide him with such a framework, either by building from scratch, by downloading an open source framework, or by building and releasing. In the end, it might only be the reduction we give to customers which allow us to release the results of our collaboration, that eventually convinces them to "go open source". Technically, releasing the source won't help them much in reaching their goals sooner or at a lower cost.
So, what is the value behind building open source, then? As usual, in the end, it all boils down to the people, and the environment they operate in. If you are used to work inside an open source project, you are used at working with a non-strict environmental context. Interfaces can and will change, there will be discussions about design, and you better make sure your code is designed for maintainability and extensibility, or someone else will go in and refactor the hell out of it. Also, you have to stand up and defend your ideas, and you're aware of the fact that people are looking into each and every code change you are committing to the source repository.
Quite a few of our customers, who were lured by us into using CVS as the shared workspace, were surprised to see us going mad when their initial interactions with CVS were resemblant to the use of "a shared network drive with some versioning". We immediately tried to explain them that one is supposed to commit atomic (enough) changes, that log messages should be sensible, and that one must not forget about local environmental context changes in somebody else's sandbox. Not everybody has its CVS modules checked-out in e:\projects\blablabla. Not everybody uses Eclipse with indent set to (godforbid) 3, with tabs instead of spaces.
As you see, we have an attitude towards development practices. We learned these practices out in the open, by running into walls while others were watching over our shoulder. If we hadn't been involved into some open source projects, we might never have learned and experienced how different one must explain, code, document and interact, when working in a volatile, collaborative context.
So, what does our customers buy with this we-build-open-source thing? Flexibility and strong interfacing in design, resilience in the build setup, and the ability to cope with changing requirements and project plans. Oh, and that attitude as well, of course. ;-)

July 08, 2003
Ejaculatio praecox

... or, the two-minute-men from BEA marcom.
Looks like the little stone that we started rolling early May is prematurely being changed into a meteor hitting the mud puddle. Oh well - it must be hard to pull the brakes of that marketing train. Let's just ignore the fuzz and see how the code looks like instead. Trigger-happy marcom people... sigh.

July 07, 2003
HTMLArea bliss

When researching possible solutions for inline editing for a customer project, I stumbled onto HTMLArea, which is a WYSIWYG <textarea> replacement released under a BSD-ish license by the cool folks of InteractiveTools.com. The stable 2.03 version runs under IE5.5+ on Windoze only, but the new version, developed by Mihai Bazon (or Mishoo), funded by IT.com, runs under Mozilla 1.3, too. He claims it to be running under all OSes, but from a quick test by Bruno, it doesn't run under Linux.
Either case, I'm very happy with this inline WYSIWYG thingie, and I immediately ventured into integrating it with MovableType. Despite my severe Javascript brain damage, it was actually quite easy.

<script src="<TMPL_VAR NAME=STATIC_URI>htmlarea/htmlarea.js" language="JavaScript1.2"></script>
<script src="<TMPL_VAR NAME=STATIC_URI>htmlarea/htmlarea-lang-en.js" language="JavaScript1.2"></script>
<script src="<TMPL_VAR NAME=STATIC_URI>htmlarea/dialog.js" language="JavaScript1.2"></script>
<script type="text/javascript">
var editor = null;
function initEditor() {
editor = new HTMLArea("taha");
var cfg = editor.config;
cfg.editorURL = "<TMPL_VAR NAME=STATIC_URI>htmlarea/";
<body onload="initEditor()">
<td colspan="2" width="100%" valign="top"><textarea<TMPL_IF NAME=AGENT_MOZILLA> cols=""</TMPL_IF> 
class="width500" id="taha" name="text" rows="<TMPL_IF NAME=DISP_PREFS_SHOW_EXTENDED>10<TMPL_ELSE>30
</TMPL_IF>" wrap="virtual"><TMPL_VAR NAME=TEXT></textarea>

The HTML output is reasonably clean, if not XHTML of course. Caveat: Tom already reports the editor doesn't work under Moz 1.4b. Since his blog is running on the same MT instance as mine, he's an involuntary test guinea pig. But he can cope with that. ;-)

Update: the obligatory screenshot:

I owe Simon an explanation

Simon is wondering why I don't like his blog. Mind you, I'm faithfully subscribed, as I already was to his little email announcement list quite some time ago, and I remember posting him a note in favour of his musical preferences (it used to be Vertical Horizon at that time). Being the professional traveler that he is, I was surprised to receive a personal answer from him shortly after that. So I figured Simon to be a cool guy, even though he's a perpetual keynote puppy. Which is partly my fault since I also invited him over to Belgium early 2000 for such a thing.
I still carry a (credited) slide from him in my presentation collection, which I affectionally call "Phipp's Onion" - so it really took me some time to become slightly annoyed with his personal outlet. It began shortly after his eagerness to be on the self-invented A-list, and became gradualy worse when he was worried about Java.net not being sufficiently linked by that same A-list. All-in-all, Simon's outings are assumably more directed and tuned towards an audience of CTOs and CIOs, so I get bored quite fast when he comes with the once-in-a-while "Am I sure I haven't missed anything cool happening"-blogs, heavily cross-linked since he was explained this is considered to be bon ton in Bloggerati Land. But I wouldn't mind all this if there was some genuine content and thoughts to be discovered. Even though Sean's blog carries the laughable pompous title of CTO, Propylon, Sean regularly blogs some cool, techy URIs. Simon too often recycles stuff I read already in some other places. Except for the SocioPolitical stuff he is blogging about, I think the GoogleAds alongside his blog explain the problem: too much blogging about blogging. Too little about Simon, the clueful Sun Keynote Puppy. I want to read about the human being, entangled in a company which walks the extremely thin line between selling pizza boxes, bad-mouthing M$ and FUDing Open Source, but not offering a real alternative themselves. And I know Simon can do better than this.

July 06, 2003
Tracking Ted Back

Ted asks for a test of his trackback service. Here you go, Ted!

July 03, 2003
Java vs .Net developers

This has been something spinning around in my head for quite some time now. I used to be working for a mid-size company with a nice mixture of Java and Mickeysoft peeps, good and bad ones on either side of the virtual fence. I happen to know also some M$ and Sun folks (at least local ones here in Belgium), and I've been wondering about the essential difference in well-being of both groups, with their choice of software architecture and products.
Mickeys in general are very happy with the latest new tools thrown at them from Redmond, and very generalized also look slightly happier. Java developers on the other hand have a slightly more weary smile, and especially the open source-addicted ones like to make a lot of fuzz about Sun not doing justice to the great platform that Java is.
Technically, the differences in both software platforms are getting smaller every day. I wonder when the Java developers will be as happy as the Mickeys.

BEA's XMLBeans going open source?
It's funny to see how opinions can change over time - mine in the first place. Earlier this year, I was apparently still believing that the JCP might serve Java users well, but after some more months of hearing about the Sun Open Source FUD, and seeing how good, honest proposals are purposedly being held back and some stupid ones are being approved, I decided the JCP and effectively all Sun-involved community efforts are essentially doomed. Even Simon is making a clown of himself, hopelessly trying to be one of the pack, with all his A-list blogger hyping, and blogging about blogging or the new passing fad in town social software. Let's be serious: software can't be social, how could a computer program be considered to be anything near-human? Even a bunch of bonobos expose more group behaviour than a cluster of bytes on a hard disk.
Anyway: during XML Europe, I had the pleasure of talking to Cliff Schmidt. He was new to BEA, coming from M$, and apparently on target to bring some open source and community clue into BEA. And even though the first announcements of said XMLBeans weren't exactly heartily welcomed in my very own little blog, Cliff proved to be really serious, and actually caring about BEA doing Open Source good. We exchanged quite a few lengthy emails over the past two months, and now he went public with his XMLBeans proposal on the Apache lists. I won't interfere with the debate, since that's how the game should be played, but Cliff is standing up pretty well so far.
It would be cool to see BEA being added to one of the large software companies working on Apache projects - diversity is good!