Schrödinger's Vet

Binding back

eneylon — Sun, 27 Mar 2011 07:54:01 +0000

The last article discussed how to convert XML content to bind rendered content to its source – by carrying references back to the originating document using an XSLT transformation. To make use of that data in the client (generally being a web browser), we need an interaction mechanism. HTML is already self-aware of its structure through the DOM, and so a user interaction can be easily captured in relation to some rendered content.

The following example demonstrates how simple it is to get JavaScript to make HTML aware of its own existence (existential HTML):

Click on this text to see that a different context can be recognised even when nested ^inside another context.

Clicking in the different areas of the text in the document results in different messages being displayed based on the context of the mouse in relation to the ids used in the document. By extension if every html element contained an XPath back to source (or a hash key for an XPath stored on the server for efficiency’s sake), then every presentational structure can be tied back to the semantic structure of the originating document. It’s up to the transformation writer to ensure that those structures that need relating back to the original are passed through.

Allows us to point to particular structure in the source document

Since XPaths can be arbitarily long, it might be more efficient to use a hash on the XPath and do a lookup on the server side if content is sent back (for example for adding annotations to a particular node in the XML document structure). This approach has been used to allow users to comment on data conversion quality by having jQuery popup a menu when the user right clicks some content.

Annotation Correlation

eneylon — Fri, 18 Jun 2010 21:06:23 +0000

When structured content is converted for presentation, the relationship between source and rendition is often lost.

Increasingly documents are being made available not just for reading, but also for writing. Wikis allow editing of content from a raw state, but the bulk of annotation (for example in consultation exercises) still needs moderation or processing before affecting the source document. So the publication of transformed documents for annotation is a legitimate model for soliciting input to those documents.

The problem comes when the comments need to be tied back to the source. Lossy transformations are common when documents are converted into HTML. Reversing a transformation is not often a design consideration and the semantics of source tags are often lost when rendering content.

What is needed is a means of commenting on a presentation form that allows annotation at the precision of the source document. The approach described here assumes the source is an XML document that is transformed for display to the reader.

In order to be able to tie items in the HTML to their corresponding elements in the source document, each element in the source must be uniquely represented in the rendition. The approach advocated is to insert the XPath for each node in the source in a attribute of the rendered content. It is proposed that this attribute be named noid.

Of course there are other ways of achieving the same result: such as creating a lookup table of XPaths and giving each node a corresponding guid. However the direct approach of inserting the XPath in the attribute has the advantage of simplicity, not needing another data structure, and transparency. One disadvantage is the increase in file size becomes a function of content structure and element naming rather than just the number of nodes in the source.

The solution uses XSLT to transform the source. This allows for easy extensibility and placement in a transformation pipeline. The code below enhances the identity transform by adding a new attribute to every element output. That attribute contains the XPath of the element that it corresponds to:
/ [ ]

The additional code performs a path trace which takes the current node and calculates the route that is needed to access that particular node in the source.

The result of documents transformed with this is an identical document which can be used to render HTML and provides a means back to the source for every element. Any subsequent processing can choose to make use of those links. This would typically by using the noid attribute to populate the id on a div or span element in HTML.

This post has shown that an XML document can be transformed to provide a route back to the source document in a subsequent rendition. In the next post in this series I will cover how to make use of that path from the rendered document using JavaScript events and the HTML document object model.

Event Horizon

eneylon — Sun, 04 Apr 2010 15:59:02 +0000

Going to, and looking forward to, events has always been important to me. On those occasions that I glimpsed the possibilities of groups drawn together by a common interest, something ignited inside.

The first regular events, that I elected to participate in, were sports-related. Living close to Highbury Stadium, before the spate of incidents that forced football stadia to control access, provided ample opportunity to indulge in the fix of fixtures. Back then, we [friends] could get in free to the stadium at half time. Jostling towards any confrontations that were brewing hinted at other possibilities from these gatherings, and provided relief from the boring football on offer.

Days going to Twickenham were memorable. The adventure was in travelling to strange parts of London. We could slip the gate a few pounds to get into the (equally boring) rugby internationals of the early-eighties. Even venturing to White Hart Lane (to watch more interesting football) was just a precursor for the passion that was to follow.

The epiphany happened one summer in Ireland. Our family went every year, and the summer of 1983 looked set to be no different to any other. Until the day an older cousin asked if I wanted to go along to a festival in Dublin. I was 15 and bored at being stuck in the countryside: and so jumped at the opportunity. Off I went to my first gig – Anvil, Motorhead, Twister Sister and Black Sabbath at Dalymount Stadium.

That gig changed my life. Back in London, I dragged my best friend to a virtuoso performance by Ritchie Blackmore’s Rainbow. We lost half a stone in sweat each that night – and he wound up working for a record label. It wasn’t just me that was caught up in the possibilities of the event experience.

The next ten years were focussed on getting to gigs. Being much more focussed on urban gigs than the festival scene, I developed many strategies around getting access to live music. It wasn’t necessarily expensive: getting to know the bouncers at the university venue was rewarding in many dimensions.

With age, it became less urgent to be at the right gigs. And then it started feeling akward – being at he front of the crowd was no longer important, and the sense that something important would be missed was plainly mistaken. The wilderness years occurred when my daughter was born. Going out was less important. Being there for her is what is important. But time passes, and we all move on ….

The passion was rekindled with the organization of a few tech events – things I’d like to see happen and couldn’t wait for someone else to organise (such as XMLOpen in 2004). Then I took my daughter to her first gig (Yassou N’Dour in Lucca, Italy). Suddenly I was reminded of what it was that had enthralled me for so long.

Then unconferences became commonplace. Suddenly I could indulge my interests without having to feel indebited to an employer. Perhaps not as focussed as some of the paid conferences I’ve been fortunate to attend; they nonetheless provide opportunity to discover what is really important to you. I’ve learned more about myself at these events than at weeks focussing on obscure technologies in distant countries.

So now I spend much more time thinking about how I’m spending time. Events are as important to me as ever, but I’m no longer looking to recreate the last buzz. What’s important now are the opportunities afforded: to learn new approaches, meet new people, and to think new thoughts. The type of event is just a constraint – and one that can be deconstructed by those attending.

I’ve just booked tickets for Womad in Charlton Park this July and would love to hear from anyone interested in doing something a bit different while we are there.

Hacking Sunrise

eneylon — Mon, 15 Mar 2010 21:51:10 +0000

Sunrise is my hack from wherecamp.eu.

It’s a calculator of sunrise and sunset times that can be used to build location-based applications. Potential users of such applications are anyone needing to know when surise or sunset will occur at their location – or another specified location. This could be photographers wanting to capture landscapes bathed in colour, farmers who wake a specified time before sunrise (if farmers still do that), religious observers with dietary or worship practices based on sunrise or sunset, etc.

Timings are calculated using an algorithm from the Almanac for Computers, 1990, published by Nautical Almanac Office, United States Naval Observatory. It makes sense to use an approach driven from a need for precision informtion so this seemed a good starting point. Implementation in code was trickier than originally thought because the trignomatric functions need converting between degrees and radians - fortunately there is a worked example available which also led to the discovery of an an error in the algorithm. Additionally there is the need to account for step changes in time zones when factoring in summer time.

The hack took this data and cycled through the days of the year. For each returned value a line was drawn on an SVG canvas – showing the variation across the year. Near the equator this variation is minimal (the days tend to be around 12 hours), but once we deviate into seasonally-affected areas the utility of knowing each days twilight tmes becomes more apparent.

The calculator source code is posted on github and I welcome implementations in other languages, extensions and improvements. If you would like to get involved ping me on twitter where my username is eneylon. Next steps on this are to figure how could the latitude and longitude can be retrieved: they could be provided by the application, entered into a web form, supplied from gps access, or accessed with a javascript library. There is also more work needed on calculating summer time change dates.

Unfortunately I’m not patient enough to be a graphic designer – so to move this forward it would be good to work with someone skilled in design to collaborate on rendition of an attractive interface to the data.

More WhereCamp.eu

eneylon — Sun, 14 Mar 2010 08:45:05 +0000

On the second day of WhereCamp.eu we moved to the luxurious Guardian offices in Kings Cross. There were fewer attendees in a much larger space, giving the event a distinctly different feeling.

At the start of the day I met up with a former colleague who was going to the open source in geographic information systems session. So I followed him. The session was about the organizational support provided by the Open Source Geospatial Foundation. A theme that came across, and echoed from yesterday’s sessions, was the problems with open source tools being focussed on the needs of developers, rather than the needs of users – although there is optimism that the Quantum GIS project may be addressing this issue.

Then there was a session about using historical data. It was led by Peter Miller from ITO – whose colleague impressed enough the previous day to make his affiliation on the wall enough to convince me to attend. The pitch was to initiate a project to make current and future data better quality. Using the example of high speed rail planning. We were shown currently available consultation data and invited to think about the social consequences of poor quality data locked in PDF documents. The discussion focussed around the idea of time-travel and there was some serious scope creep from the initial concept. Peter proposed turning historical and current images into data – extrapolating backwards in order to improve the tools to plan ahead. As he said: “getting people who are excited about the past makes the tooling much better for planning the future”. The session concluded with a group of highly motivated people agreeeing to work on a project to create a 1921 version of the rail network on Open Street Map – a good result for a well-prepared pitch.

The last session before lunch was about accessing location based services on the iphone. Tom Melamed took us through various ways to simplify writing an application using wrapping technologies. Contending that the best way to develop an iphone application is to write a web page, Tom first presented the W3C geolocation specification and demoed how a mobile accessed webpage can ask a browser to provide location. He went on to show PhoneGap – a wrapper around a webpage delivered as an application (needs a mac to use). Another solution to this is titanium (and there are several others providing similar Objective-C avoiding solutions). Another useful tool mentioned was jQTouch – a javascript user interface library that makes web pages look and feel like iphone apps. It can be combined with phonegap to build apps – phonegap apps are allowed on the app store (and it also works on android).

After lunch was spent watching the rugby international; and chatting to others who were torn over their priorities. As the closing session and hack review overlapped the last 20 minutes of the match (Ireland-Wales, in case you are wondering), I took my computer (and the audience) into the session with the rugby streaming over the wifi. As soon as the game was over, it was time to present my hack for the weekend. The hack is a sunrise/sunset calculator that will be posted here tomorrow. As there were only three hacks presented, it stood some chance of claiming the prize. It didn’t win, but the voted result was closer than expected. Congratulations to Michael Dales for his winning hack.

Wherecamp was a great experience. Both days were memorable for the quality of the sessions and the generous spirit in which everyone shared their thoughts and ideas.The organisers stressed that anyone can organise a WhereCamp (or indeed any other camp) and encouraged us all to thing about doing so.

Ups and downs at WhereCamp

eneylon — Sat, 13 Mar 2010 10:54:18 +0000

The first day started slowly for me. By lunch time doubts had crept in about the event. But the post-lunch sessions were excellent, and more than made up for a slow start.

With six tracks it was hard to know where to go. So, I stayed put in the main room and had the sunrise/sunset SVG rendition working by lunchtime. Lunch itself was healthy – a pleasant change from the norm at geek events. The only ting I remember about teh morning session was Gary Gale’s “it is essential to be able to lie about your location” – which should ring true for all of us (clandestine activities or not).

Post lunch, in the main hall where there was a Q&A session on Open Street Map. It uses wiki technology with extensions – an approach which reinforces my feelings about wikis being a base infrastructure for data collection. The team showed some visualizations from the Open Street Map data collections and highlighted “shards of wrong” where errors are made glaringly obvious by presenting them visually. It was apparent that the value of visualizations may be in finding anomalies, rather than in confirming expectations.

Next up was a session of “info porn” – gorgeous visualizations of transport data. Hal Bertram of ITO showed a series of time-lapse animations of traffic data. At the kick-off we had been shown an animation demonstrating how collaborative mapping initiatives have grown exponentially. But this was more interesting, because of the deductions that could be made from watching the data move. While we were still salivating at the high definition animations, Hal finished by saying “at its best visualization produces stuff that you want to look at, and learn from”.

Open Street Map 2008 Map Edits from Peter Dunn on Vimeo.

After that I found myself learning about lenticular printing . In the discussion Sarah Kate Norman suggested that it might be an idea to use lenticular printing for sign information in different symbols/languages when the Olympics come to London. The fine map print that was shown at the session was produced by a company called Riot Of Colour.

Then tried to get into the YQL session but the small room was heaving. There are many materials available online anyway - so that’s one for self-study.

The final session was a debate (not sure it was intended as such) about privacy. I stayed quiet but thought there was youthful optimism that the problem would fix itself after some high-profile incidents – it hasn’t worked for patents!

On the way home the train hit something – we were delayed by over two hours, but I used the time to hack. So it really was a day of highs and lows. Now to see what today has in store.

Looking forward to WhereCamp.eu

eneylon — Fri, 12 Mar 2010 08:21:12 +0000

En route to WhereCamp.eu (12-13 March 2010) ….

Having forgone the opportunity to be in the Czech Republic for XMLPrague, I decided to do some preparation for two days initiation into location-based computing at WhereCamp.eu.

So I started two projects before arriving: a sunrise/sunset calculator and a gps hardware hack. The calculator working and the £20 GPS USB dongle (from maplin) is emitting data – although it took a while to figure how to scan the com ports to detect which one was emitting the data. The plan before the event is to create graphs of annual sunrise/sunset data for given locations in SVG using the input from the GPS. But the point of these events is to be receptive to new deas - so who knows what will emerge.

So long as I can get to see the rugby tomorrow, follow xmlprague and sxsw, and find somewhere to crash tonight, there should be something interesting to demonstrate soon.

Small and random

eneylon — Thu, 20 Aug 2009 21:26:30 +0000

Such is the mindshare of twitter that the need to write structured prose recedes and becomes a distant echo. So why change that now (dear reader)? Well in part to let you know that the heart still beats – in fact better than for a long time. Hope you like the new steampunky theme.

Back to basics

eneylon — Fri, 10 Oct 2008 10:03:05 +0000

Tim Bray rewrote his keynote for FOWA yesterday based on how he’s feeling: scared. The economic downturn has him worried as he doesn’t know what is going to happen (how arrogant to assume he did know what was going to happen beforehand). I find the motivation for this type of talk to be lacking in humility – but he is onto something in the body of his talk.

He’s talking about how capital investment is not going to be approved anymore in organizations as times turn hard. But alongside that is the opportunity for those who can build things to satisfy core needs. We are used to luxury discretionary services. What we need to do now is to build tools that deal with the essentials and the revenue will be more based on small payments rather with tiny margins.

Tim told a ‘story’ that I heard several times in the halls yesterday (a meme as it were): The only user who you will totally please, and delight, is yourself. And the fact that other people may have the same needs means there may be a market for that same need satisfaction. So aim to satisfy your self and you may end up creating a mainstream application. He also cautioned would-be entrepreneurs to stay away from the venture capitalists, saying they did not add much in the good times so they will be worse in bad times.

Tim advises anyone who finds themselves with time on their hands to skill-up, learn new things, contribute to open source projects, blog, twitter, network and make a name for yourself. “If you don’t care enough about the web to make it better; they why would I want to hire you?”.

Suppose I’m wrong – he says. Well nothing he’s said is still not true … so back to basics for us all then. Funny though, Tim went back to form and assuming he was the only keynote speaker told the audience we could now go and get coffee.

Less of me

eneylon — Thu, 09 Oct 2008 10:05:09 +0000

Oh there you are. Where have you been. Me? Oh I’ve been doing stuff. Lots of work things that I can’t particularly talk about and some personal stuff that I can.

There’s now less of me that there was none weeks ago – three and a half stone to be precise – thanks to LighterLife which despite the branding does cover men as well. I think I’ve done the easy bit – losing the weight; the hard bit (maintenance) is still to come. But I feel much better and its reflecting in many aspects of life.

I’m at the Future Of Web Applications Conference at the atmosphere is electric. I’ll blog any interesting insights that come during the next two days, but for now I’m just enjoying being at an event that I choose to be at rather than one that I am required to be at.

Earlier this year I held a hack session at home during the Olympics – only three were able to make it (I chose a particularly bad weekend when many people were leaving for holiday), but we had a great time and learned a lot. The tangible results are at Inigo Surguy’s website and we are inspired to do more similar events.

Next month there’s a SWIG UK event at HP Labs in Bristol which I’ve helped organise. Sign up now if you are interested as there is a finite number that can be accommodated.

That’s all for now except to say that I haven’t disapperaed off the web between blog posts – I’m also twittering which for the uninitiated is a micro-blogging system that has some interesting social artefacts. See you there, perhaps?