my recent reads..

Registering namespaces with SOAP::Lite

About time to post about something ... too many OpEd pieces of late!

An interesting question came up on the soaplite mailing list concerning how to modify the namespaces registered in the SOAP envelope. Documentation is not especially clear on this point.

Of course, a quick hack is to attach a full namespace to an element directly, as in:
SOAP::Data->name('itemName')->attr({'xmlns:mns' => 'http://my.namespace'})

With version SOAP::Lite 0.65 and above, the register_ns serializer method helps to correctly construct the envelope, as shown in the following example:

#!/usr/bin/perl -w
# $Id$
#

use strict;

#SOAP::serializer->register_ns requires 0.65
use SOAP::Lite 0.65 +trace => 'debug';

my $soap = SOAP::Lite
->proxy( 'http://localhost/blah/DummyService' );

my $serializer = $soap->serializer();
$serializer->register_ns( 'http://my.namespace', 'mns' );

my $som = $soap->call(
SOAP::Data
->name('mns:test')
=> SOAP::Data->name('mns:description' => 'an item in my namespace')->type('mns:mytype')
);

This generates the following SOAP request:
<?xml version="1.0" encoding="UTF-8"?>
<soap:Envelope
xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
xmlns:soapenc="http://schemas.xmlsoap.org/soap/encoding/"
xmlns:mns="http://my.namespace"
xmlns:xsd="http://www.w3.org/2001/XMLSchema"
soap:encodingStyle="http://schemas.xmlsoap.org/soap/encoding/"
xmlns:soap="http://schemas.xmlsoap.org/soap/envelope/">
<soap:Body>
<mns:test>
<mns:description xsi:type="mns:mytype">an item in my namespace</mns:description>
</mns:test>
</soap:Body>
</soap:Envelope>

read more and comment..

A blog about nothing (was: a can of worms)


Jake is right .. this is getting a little too much like Seinfeld.

So many posts about bloggers at OpenWorld.

Seems to me a storm in a teacup (admittedly, roughly the same size as a can of worms, but probably shouldn't be confused - print picture on right for easy reference). Most of it driven by a comparison to what "other" companies do.

As an avid blog reader, I'm actually more interested in an intelligent post from someone who has taken some time to reflect and write about their usage of the software (be it from a project, production or just playing with an OTN download).

Don't get me wrong, I definitely see value in blogging from OpenWorld - as much for the community buzz as discussion of the latest and greatest OpenWorld news. But to put too much emphasis on this I think actually plays into the hands of the supposed PR and Marketing heavies (everything focused around a single, well orchestrated event).

Since I also represent a "global audience" (living in Singapore), there's also the geographic factor. Notwithstanding whether travel costs are reimbursed, I never expect an event centred around a single location in the US to fully represent the diversity of the blogging community I'm so comfortable with.

As such, I'd be even more impressed if we saw OTN podcasts evolve into a more interactive channel (e.g. using talkshoe a la net@nite) to give true, open access for the community to key execs (and then blog about it).

Then there is the question of disclosure. Mary Ann Davidson just posted a poignant discussion of disclosure. Although arrowed at a security audience, the timing is perfect for the opening of Jake's can of worms;-)

I just can't seem to get myself worked up over this either. Perhaps something I drank last night? Or perhaps I've just got a bit more faith in the sophistication of the audience that is now growing up with blogging et al. Two factors: influence (declared or not) stands out like a sore thumb (or else why is it so easy to recognise the executive blogs that are straight out of the marketing playbook). Second, thanks to the magic of RSS I'm not just listening to one person's voice.

So all in all, I reckon getting free registration is a pretty good first step (despite the fact that other companies may do more for you). It's not like the information won't be available for all to see and share after the show, so if you want the prestige of having the first blog post up on a particular subject maybe paying some of your own way ain't such a bad deal.

I would make one concession however: I do think it would be in Oracle and the community's best interest for Oracle to have some flexibility when it comes to the (very) few bloggers who have truely crossed the line and are in fact analyst/press and should thus be treated as such.

So much for my post about nothing;-)

Disclosure: since I work at Oracle, I never expected a blogger invite (and I wasn't able to wangle an employee seat), so perhaps that explains my disinterest in getting all het up about the issue!
read more and comment..

The Best Software Writing I


It's a pity that there was never a volume II for this classic collection of the best software writing, selected and introduced by Joel Spolsky.

I just picked it up to re-read some of my favourites, like
  • Bruce Eckel's Strong Typing vs. Strong Testing, which is a bit of a milestone article in the rise of the "scripting" languages like ruby and Python.
  • Adam Bosworth's ICSOC04 Talk which reminds us it is more important for computer science to be understandable by people than by computers.

There's also some great pointed humour, with
Definitely one of the all-time-must-have tech reads (and re-reads).


read more and comment..

War Dogs


Picked up Michael Farmer's War Dogs from the library the other day. Proved to be a great airport/trip read.

Its the third in a series of "alternate future-histories" set around the middle-east conflicts from Desert Shield onwards.



read more and comment..