So, this is where I've been (gee, I
love this stuff :-)). Courtesy of David and
Andrew.
Try it yourself.
Archive for January, 2004
I've been through
this already, and for sure most of you have too, but I fell in the trap
again today, so I'll blog it to my perpetual abomination.
Here we go: "foo.xml" is
<foo xmlns="something">
<cinclude:include src="bar.xml" xmlns:cinclude="http://apache.org/cocoon/cinclude/1.0/>
</foo>
and "bar.xml" as
<bar/>
run this through:
<map:match pattern="foo">
<map:generate src="foo.xml"/>
<map:transform type="cinclude/>
<map:transform type="log"/>
<map:serialize type="xml"/>
</map:match>
Watch your console and notice how sweet is your parser in stating (correctly), that <bar/> doesn't belong to any namespace.
Watch your browser output:
<foo xmlns="something">
<bar/>
</foo>
Imagine such output serialized and reparsed at a later time.
Start cursing.
Make a note to self: stay away from non fully namespaced XML whenever possible and avoid messing up with documents in the default namespace like the plague.
While at it, also note how Linotype sucks in writing code-oriented blogs and find some time to solve that.
I just sent a proposal to the Catacomb mailing list to see if there is any interest from the community in joining Apache. My fellow readers might remember that not a long ago we contributed an addition to Catacomb in order to support ODBC as an alternative database layer to the MySQL only implementation: as you might see, I haven't lost interest in WebDAV: quite the opposite actually (you might end up soon reading a few repository - content management - Wiki rants here too).
Anyway, if the proposal gets accepted it looks like I'll be trying on my bones the dreaded Apache incubation process: wish me luck (oh, and if you know of an Apache member willing to champion this proposal, I'd be more than interested). And, Nicola, get ready for some serious bugging. 
I have been invited by a good friend to join Orkut, the new Google-affiliated social experiment. I actually joined and started filling in personal information, but after a while I started wondering what might happen if and when this service will be highly successful. Will really Orkut owners refrain to use all the material we're happily giving them for free? I mean, they will end up with an enormous database of personal habits (which boil down to how you spend your money), preferences, geographic locations and the like: in an overparanoid scenario, consequences might be quite ... well... scary.
I keep reminding to myself that the Google motto is "don't be evil", and I sure hope that they'll stick to that. OTOH, it will be interesting to see where all those FOAF-like experiments will bring us: somehow I share Matthew'sconcern (will all this be really useful apart from the obvious dating thing?), but there might be some interesting scenarios open. Right now I'm impressed to see that after a few hours I am (in theory) connected to more than 5.000 people: when will we meet for a beer?
So this will be the last post on my faithful cocoondev blog as well as the first (official) post on the new one. I need to thank Steven for being so king to host me on his machine: it has been a nice first experiment in blogging, but after 5 years I finally had the occasion to move to my old faithful domain, and I just had to grasp it before work starts pouring again (first showers are already there, BTW, but it looks like it will be a fun ride). Stay tuned for more news: now that I have no more excuses, the next challenge is becoming a full citizen of Blogland and not just a casual user.
So this will be the last post on my faithful cocoondev blog as well as the first (official) post on the new one. I need to thank Steven for being so king to host me on his machine: it has been a nice first experiment in blogging, but after 5 years I finally had the occasion to move to my old faithful domain, and I just had to grasp it before work starts pouring again (first showers are already there, BTW, but it looks like it will be a fun ride). Stay tuned for more news: now that I have no more excuses, the next challenge is becoming a full citizen of BlogLand and not just a casual user.
At last! After a couple of months of enormous struggling, I'm finally settling down somehow, slowly catching up with all I left behind.
Short version: we have a new house and a new, powerful ADSL line with a static IP (courtesy of my employer): it sure seemed a very good time to clear the dust off my faithful SparcStation 4 (ah, memories!) to see if I could get something useful out of it.
So, I'm back blogging or, better, betablogging since there are quite a few rough edges to sort out: will the poor 32MB, 60Mhz and 200MB free-disk Sun pizza box survive any load? Will this Linotype based experiment suit what I need from a weblog? Well, I guess time will tell.
For the curious in you: the Sparc4 machine, of course, isn't running Cocoon. The Cocoon application is running on my notebook, with blog contents being dumpedon the local disk courtesy of wget (I had some weird issues with the Cocoon CLI insisting to rename my .css into .css.html) and rsync'ed to the web server, which is then serving just a static site.
So, this is Apache 1.3.27 (latest binary available from Sunfreeware, did I mention that I have no compiler on the baby since gcc just doesn't work?) serving you pages generated by a very slightly modified (path adjustments mainly) version of Linotype: as soon as some cash flows in, the Sparcky will be replaced by something able to run Cocoon directly. The photograph on the left is an old picture of the Lanterna
located in Genoa, a small tribute to the place I come from.
Latest Comments
RSS