The Propaganda Tiles are a number of exotic abstract graphics that can be seamlessly tiled to form backgrounds. I am particularly keen on using them for desktop backgrounds to avoid plain solid desktops (too boring) and photos of one's children (a but naff). Mind you, some of the Propaganda Tiles are somewhat psychedelic, so careful choice is required.

The tiles were originally created by Bowie J Poag and released under the GNU GPL. But finding them on the Web has been difficult of late.

But to my joy, I have discovered a mirror of the complete set here.

One off the odder things about them are the peculiarly evocative names of the tiles, such as plastic-dinner-plate-1.jpg and lowdown-popcorn-1.jpg.

Me? I've just chosen the-alias-line-1.jpg. Smart.

While browsing my favourite newsgroup, rec.music.classical.recordings, this morning I came across a link to a YouTube video of Arturo Benedetti Michelangeli playing Chopin's 1st Ballade.

The combination of my favourite Chopin composition and a pianist whose performances are always interesting proved irresistible, especially since the recording through which I've got to know the piece, Artur Rubinstein's on RCA, has – though fabulous – made me think that this is music that is, as Artur Schnabel put it, ‘better than it could be performed’.


Rubinstein's late 1950s Chopin studio recordings
of the Scherzos and Ballades


And the video was tantalising, with Michelangeli showing all the executive perfection, good taste and narrative sense that is expected of him. The one disappointment was the sound quality (obviously one doesn't expect high fidelity through YouTube) and so I was prompted to look to see if I could buy a CD with his performance in decent sound.

Some googling revealed that there was a modern recording in print from DGG, but on looking at the CD cover I was surprised to see this was a recording I already owned. Somehow I had overlooked the existence of the Ballade recording on this disc! (maybe because I bought it principally to hear Michelangeli's Mazurka performances).


Michelangeli's 1972 studio Chopin recital disc


The CD recording is of very high fidelity (though a little dry) and presents a similarly conceived – though a little less austere – performance in fuller tonal splendour. This is a first Ballade to rank in my affections alongside that of my Rubinstein disc.

Somehow though, I feel the perfect 1st Ballade recording is out there somewhere; though equally part of me knows searching for it is fruitless.

Hmmm, I wonder what Sviatoslav Richter's recordings of it are like …

On Tuesday I attended (and spoke at) an XML UK member presentation day in Reading.

Unfortunately things got off to a bad start when I was caught in the transport chaos caused by the closure of King's Cross station, meaning I arrived late and would have to leave early.

Nevertheless I did arrive in time to catch most of Antony Scott and Bilal Baig talking about their "Information Modelling Toolkit". The heart of this was software that allowed round tripping between XML and a form-based UI that would allow it to be edited (rendered via Microsoft Excel). This is an interesting area and one of the questions afterward pinpointed what I think is still the biggest question here: can good looking (i.e. more usable) forms be produced automatically for editing XML? Often there is a heavy design component on good forms, just as there is for any user interface onto data.

Next up was Stephen Green, representing a new Bristol-based company SystML. It is always good to see new ventures coming into the XML UK scene! Stephen has been working closely with the development of UBL (Universal Business Language) and his talk focused mainly on the formal definition of UBL using W3C XML Schema. Whether intentional or not, Stephen's talk turned into a damning indictment of XML Schema, as he chronically how ambition after ambition were thwarted by problems with the language, its implementations or both.

This talk reinforced my conviction that W3C XML Schema is best avoided (Since my first thoughts on this topic I have, in my commercial life, been recommending against the use of W3C Schemas in most circumstances). In many cases, good old DTDs + Schematron give an excellent modelling solution. But sometimes the will-o-the-wisp lure of W3C Schema's supposed advanced features (polymorphism, etc.) just proves irresitible to customers -- especially when developers who are new to XML (but used to OO programming) have a strong say.

During the coffee break it was great to catch up with old friends and customers and swap notes on projects and technologies. Among the publishers there seemed to be little stomach for schema-based solutions to projects; yet there is noticeably now more talk about using XML for semantic modelling of information using triples, RDF, OWL, Topic Maps, or the like. Maybe these technologies' time is finally coming ...

After the coffee break Francis Cave spoke about DSDL in general and in particular about Part 9 of the standard, of which he is editor. This introduces XML Namespaces and proper data typing into DTDs by extending the DTD syntax (though in a backwards-compatible way). Having observed the gestation of this standard I can vouch for the fact that it's been surprising hard to get right. However, in its current form I feel the big technical problems have been overcome and this is ready to roll.

The biggest issue is perhaps implementation. There really needs to be some implementation of these "enhanced DTDs" before they'll enjoy any take-up, and the necessary validating parser surgery will be non-trivial. If done though, datatype- and Namespace- aware DTDs would be a useful means of extending the life of all those 'legacy' DTDs out there.

I was up next to speak about DSDL Part 5 (blogged elsewhere). My talk went OK, I think, but some comments afterwards made me think I needed to be more ambitious in showing what DTLL can do. If I gave this presentation again I might be tempted to have less technical detail and talk more about how it addresses data validation problems that currently can't be solved (checksum testing during validation would make a good example).


Eamonn Neylon holding forth

After lunch, my colleague Eamonn Neylon spoke about "Creating an XML Data Firewall". This talk was based on our commercial experience putting validation technologies into customer workflows, but in order to comply with XML UK's ethic it had been abstracted so as to avoid becoming a product pitch.

Eamonn focused on how validation was not just a one-off process, but something that must be continually applied throughout a workflow. Validation (he said) was not just a technical issue, but something that actually helps to specify and document an organisation's business processes.

Judging by the number of questions this provoked, it seems these themes resonated with the audience.

Following this, I had to be rude and slip away for a tortuous return journey through London's broken transport system.

Slides from the event will be made available at XML UK's web site in due course.

At Deoksugung Palace


I am back in the UK (though my body clock isn't) after the SC34 meetings in Seoul.

Overall, and not withstanding the Office politics, the trip was very worthwhile, though the packed meetings schedule and jet lag combination led one UK colleague to describe the week as 'brutal'.

Technically, I felt very good progress was made on the evolving DSDL multi-part standard. The part which I am editing (DTLL) has been well received and I have been instructed to prepare a FCD (Final Committee Draft) before the next SC34 meetings in Montréal in August, just before the Extreme Markup 2006 Conference.

The iconoclastic DSRL (Document Schema Renaming Language) also took some bigs steps forward. Rick Jelliffe has blogged this if you want to know more.

Seoul itself proved to be an excellent backdrop to the meetings, and the admirable Hotel Lotte ran very smoothly, though with some eye-watering prices: a portion fish and chips in the lobby restaurant was 40,000 Korean Won – equivalent to US $40!

Our Korean hosts took excellent care of us, with lunch (and sometimes supper) provided. The high point was a Korean banquet of many courses, many of which centred around vegetables completely unlike anything I've eaten before, and all of which were delicious.

Much in Seoul was novel, nothing more than the streetside hotel bar's way of serving pitchers of beer. These transparent vessels featured both electric lighting and an internal compartment into which cubes of dry ice and water had been placed. The result was a glowing vessel, bubbling violently and pumping out vapour. In my opinion the exciting visual effect more than made up for the slightly bland nature of the beer, something Lars Marius Garshol – something of a beer connoisseur – would beg to differ from me on.


Graham Moore pouring bubbly beer


It was following an encounter with one of these that Lars Marius, myself, Graham Moore and Martin Bryan headed out into town in search of some authentic Korean cuisine (this was before our Korean hosts had spared us this effort, by providing it for us!) We wandered along and came to a place I had noted earlier because of the delicious smells drifting from it. A quick inspection revealed a sign saying "good restaurant" and so we decided that was recommendation enough, and went in.

Char-grilling. In the middle of each table was a hole, and above each table a lowerable metal pipe fitted with an extractor fan. We chose some beef (or 'tasty pieces of cow abdomen' as the menu had it) and watched as an urn of glowing coals was fitted into the table and a waitress proceeded to cook pieces of beef for us. The other (Korean) diners were cooking their own pieces, but we had obviously been singled-out for special assistance after I had committed a series of dining faux-pas, such as pouring my sake into my rice bowl, and dropping my chopsticks, a few times.

Nevertheless, the food was excellent with the char-grilled beef nicely complemented by a range of piquant pickled vegetables and unusual pastes and relishes. Our knees did get rather hot from the radiant heat of the coals, though ...


Plaza at dusk

One of the more tedious aspects of this multi-day SC34 meeting has been the politics surrounding the supposed office format wars, arising from the increasing likelihood that SC34 will participate in the creation of an ISO Standard for Microsoft Office documents. Signs that something was up came in the form of several conspiratorial phone calls and email exchanges beforehand, and the presence of unfamiliar OASIS, IBM, Microsoft, and Ecma people at the opening plenary.

ODF has already passed most of the ISO/IEC standardisation process, though OpenXML (or “Office Open XML” – OOXML – as Ecma are now calling it) has not yet begun its trek. I guess Microsoft/Ecma were checking out SC34 to determine what sort of reception their document would get, since clearly it would be stupid of them to put OOXML through a standards process that was predisposed to reject it.

So what will they have found?

If Microsoft HQ staff had asked their national standards body (ANSI) about SC34, they might have gathered we are not the sort of people one should associate with. As an illustration: while during these current meetings the hospitality and support of our Korean hosts has been generous and friendly, during last year's meeting in Atlanta the only message sent from ANSI was a snub: they wished to emphasise they were not hosting the meeting and did not want to be associated with it. (Hence our many American members work for SC34 in a personal capacity rather than with national backing.)

And from reading some of the web press one might have thought SC34/ISO would be a forum whose rules might be exploited to desired ends by one or other corporation.

The reality is of course different.

The members of SC34 can be generally characterised as level-headed, thoughtful technical experts, specialising in the technology of “document description and processing languages”. Okay, there are arguments – but they generally concern the intellectual and technical merit of ideas and positions. The mundane politics of commerce, and the sloppiness of thinking inherent in polarised Pro- or anti-Microsoft debate, is not the stuff of SC34 meetings. Indeed, I have heard the adjective “slashdottish” being used in SC34 as a rebuke.

No, SC34 doesn't do hype or crave the spotlight. Which is probably why nobody has heard of us — even though (to risk being described as slashdottish) many current and upcoming Web technologies are rooted in standards SC34 originated (SGML and HyTime in particular).

SC34 members are also volunteers, drawn mostly from small and medium-sized companies. Many of us know the reality is, for our customers, that Microsoft tools are often the tools of choice; and many of us judge that having a locked-down standard version of the Microsoft Office formats could being huge benefits when it comes to working with Office documents.

However, in the grand scheme of the proposed standardisation of OOXML, SC34, and its members' judgements, are only part of the picture – there are other considerations in this complex matter ...

First, SC34 members themselves will have no vote if and when OOXML is assigned to us for standardisation. It is standards bureaucrats in our national bodies (ISO member countries) who will vote. Sure, these people will often take advice from their country's SC34 members, but ultimately it is their votes, not ours, which count.

Secondly, the criteria for judging a standard are not whether it is good for competition or the software industry at large; so long as the technology is of use to the world, the criteria are more concerned with whether the standard is technically correct. This is often as much a judgement on the quality of the standards document, than on the attributes of the technology it describes.

Thirdly, there are ultimately inviolable rules which govern the justification for creating of ISO document, laid down in the rules for ISO/IEC.

It is this last point which has provoked most discussion in the bars and along the committee corridor this week. Crucially, the rules forbid the creation of “contradictory” standards – and the word “contradictory” is sufficiently ill-defined at present, for some to feel that OOXML “contradicts” ODF, in spite of the evidence that ISO itself is full of Standards which are ... well, overlapping and complementary at least.

There is also the question of time and resources. The kind of fast-tracking procedures used for ODF and mooted for OOXML give very little time for a small part-time volunteer group to give such bulky documents adequate scrutiny. There is no doubt ODF is, right now, less good than it would have been if subjected to the full rigours of ISO standardisation. With OOXML predicted to weigh in as a behemoth 7,000 page standard the danger that OOXML will be inadequately scrutinised is greater still.

It is this unseemly haste that most concerned me and my fellow UK delegates during this meeting. Although arguably the trend has already been set by some other standards groups, it is ultimately in nobody's interest if the standardisation process becomes devalued to the point where it is delivering technology standards which are done quicky, but which don't provide a solid and useful basis for implementers and users.

But, despite all the above, I expect to continue to hear the standardisation attempts of Microsoft being characterised into a dumbed-down narrative of “ODF vs OpenXML”.