A couple of weeks back, I outlined our interest in Electric Mill to public data projects , and an initial example of how A&E attendance figures could be presented. This is a good example of the data being available, but delivered in such a way as to make it kind of hard to get at.

So cue Side Project #2. If you live in Ireland, chances are you are either interested/intrigued/repulsed by house prices. This national obsession with home ownership is well recognised.

For reasons to profound to delve into here - planning regulations for residential or commmercial development - are also keenly followed by the media as well as by the public.

While individual county and city councils provide access on their websites to this information - and the information is also provided in regional and local newspapers - it is inevitably hard to get at.

Electric Mill Side Project #2 is Planning Buddy - a website that easily allows a user to create alerts based on particular named locations of interest to them.

For example, I live in Strandhill, Co Sligo. My account on planningbuddy.com is filtered to trigger emails based on matches of “strandhill” in Sligo. I can also receive these alerts as an RSS feed if I so wish.

In my “spare” time - I also maintain a community website for the area . For that site, I have configured a widget from within my Planning Buddy account - and placed this small piece of HTML code on the community site; Suddenly, anyone who visits there, has an instant view of the latest planning applications in the area. The widget is fully configurable for things like colours and so on, so that it fits nicely with what’s there already.

At the moment, Planning Buddy correlates just over half of the councils in the country including Carlow, Cavan, Clare, Cork, Donegal, Kerry, Leitrim, Limerick, Louth, Monaghan, Offaly, Roscommon, Sligo, Tipperary North and Waterford.
In order to get an understanding of what counties to include next, we’re asking for visitors to the site to suggest what counties might be completed next ; we will then look to facilitate this by building bespoke scraping rountines.

In the background to all of this, we’re currently working on expanding the reporting side - for example comparing applications per county per month, downloading the data, etc.

We’ve already had some queries from people that want to advertise on the site and that’s nice - we’ll be looking at this type of scenario and others over the coming weeks.

In the meantime, we would be delighted if you could send any comments or suggestions as to how we might improve what’s currently there.

We are becoming pretty familiar with bloggers every now and then ripping into businesses that are providing poor customer service and poor customer experience.

I’m talking here about someone that writes a less than complimentary blog post on a product or service; I am not talking about conversations and threads on an online forum, etc.

Most businesses will have little experience about how to engage properly with negative commentary from bloggers, who by definition will be pretty savvy communicators.

The jump from email, phone and forum dialogue through to discussion on the bloggers site - an away fixture if you will - is a lonely place to be slugging it out in defence of a product or service.

The thing that’s missing for me though in these encounters is the success rate or closure rate for the blogger vs. how it could have been achieved through “traditional” avenues?

Does the public dialogue and exposure help solve the problem?

Does the service provider often just dig their heels in and say “feck you” to the negative criticism of the blogger - thinking that it will go away.

The latter is the worst case scenario for the blogger - probably more than made up for by increased traffic and just a little more profile.

The worst case scenario for the service provider is in how their brand name will probably now be submerged in negativity in their first page of SERP’s.

From that perspective alone, they should be swallowing their pride, desisting from cease and desist and such foolishness, and attempt to make the customer happy - regardless of the feelings of indignation, anger and loathing that they feel.

Listened to a podcast featuring Chris Anderson of Wired Magazine on the IT
Conversations network last night while out walking.

His talk centred around his motivation and proposed proposed model for giving away his next book for free.

This was a pretty painless 30 minutes for a quick whip around of the notions of free that we are all familiar with, but have not maybe considered for our own businesses.

Yet everyone is familiar with the low cost printer where you pay premium for the replacement ink; or the cheap flight where you pay for onboard entertainment and refreshment; or most annoying of all, the cheap “razor system” with the expensive blades.

Most of the models that Anderson discusses not surprisingly, were based around publishing and books, with the ideas and statistics being particularly interesting. For example, making selected chapters available in PDF format, giving away audio versions, etc;

An example was given from Make Magazine where downloadable instructions were getting something like 5.7% click-thru rates on embedded advertisements. Impressive.

So what is the motivation for giving stuff away for free? Of course it has to be based on selling the customer something else. Free is the way to maximise the audience - you give it away in the expectation that you will be able to sell other products or services. Anderson’s own expectation is very simple - with free books, he believes he will sell more speaking events.

The link to the show notes and podcast are here - i really like his take on what makes books indispensable -

“excellent battery life, fanstasic sceen resolution, portable, looks good on your shelf and easy to flip through”

Public data and public information interests us a lot in Electric Mill. A good reason for this is that you will often find that government agencies, departments and other public bodies - while they may publish information - will normally do so in formats which while fulfilling duty, are not exactly easy to read, interpret or compare.

Think Microsoft Word or PDF files and you are pretty much along the right tracks. The reason for this of course is very simple. All the types of bodies that I have mentioned use a website that serves up pages through a Content Management System (CMS).

The main purpose of the CMS is to serve up this content; bread and butter functionality within these systems do not generally include specific capability or integration of data from other applications - at least in my experience. Likewise, the possibility of the provider providing bespoke services development around data integration is unlikely - it is just not their core business.

From a technical point of view, this would be the main reason why we don’t see easily accessible, drillable data; It would be unfair to suggest to think these are the only ones of course - you know what I mean….

In the UK, campaigns such as Free Our Data sprung up in reaction to these limitations and lack of availability. Closer to home, Conor at web2ireland outlined his thoughts - ambitious ones - for an API for Government which garnered a lot of comments.

That’s the background, what about an example?

At the beginning of 2006, Electric Mill were doing a fair bit of healthcare consulting, and it seemed like a good idea to think about clever ways to present information that people were interested in.

At the time, Hospital A&E attendance figures were hotly discussed, disputed and most importantly compared. We’re still listening to these claims and counter claims 2 years later.

So the obvious example then was to use these up to date and well maintained figures - and present them in a way easy to report and compare. Both the Health Service Executive and Irish Nurses Organisation maintain daily figures. Who is correct? - we’re not going to call that - but we did understand that the method of collection and reporting varied between the two.

So with a little bit of midnight coding, we came up with an easy way to compare A&E attendance figures in Ireland - you can read more about the background to the thought processes and the midnight coding here.

While we gathered and inputted both sets of figures from the HSE and INO, in the end we turned off the latter results; but even just with the one set of figures you can easily compare, graph and export to Excel, figures from each hospital on monthly intervals.

There were two downsides to this process. The first was obvious - we manually entered the figures for a 12 month period. This was pretty laborious. The second was a tin hat of sorts. In late May 2006 - the HSE changed their method of reporting. This meant that direct comparison of figures with the INO was no longer possible, and also the data input method we had originally developed needed modification. Hands up - we just never got around to it, which was a bit of a shame.

In a future post, we will introduce another Electric Mill public data project that we think Irish people will find useful. More anon.

A small piece of development work that we completed for a client in December 2007 threw up an unexpected problem.

We developed a small PHP based application to enable staff to easily import and validate data from a Lotus Notes system through an administration console, while at the same time enabling the public to easily search and view the same data.

Things were slightly complicated by the fact that the application was to seamlessly integrate with an existing ASP.Net CMS application - particularly with regard to the overall look and feel. the illusion to be created was of one system.

It would not be the first time that PHP has been asked to co-exist on an IIS server. We did our research, developed the system, did test installations on a QA box - and were happy with the upgrade approach that should be taken on the live implementation.

The upgrade of the live system went smoothly as planned - but then something cropped up which we didn’t expect. The style sheets which we had copied from the main CMS system (in order that the new application would appear to be pretty seamless) was refusing to load properly in all browsers - Firefox, IE and Safari.

Loading the CSS file directly into the address bar of the browser, and then looking at the page info of the file, we soon found what the problem was. The stylesheet was being incorrectly identified as text/html rather than text/css.

Searching for a fix to the issue (which seems to be an IIS one) didn’t turn up anything which we could used. We tried a few things, took on board some suggestions, but nothing worked - we still had the same problem.

In the end though, the fix was quite simple. We added the following line to be the first line in the offending style sheet.

<?php header(’Content-Type: text/css’); ?>

problem solved.

This paragraph, deep down in a Fast Company article questioning the fundamental reasonings behind The Tipping Point summed up the argument nicely :

Perhaps the problem with viral marketing is that the disease metaphor is misleading. Watts thinks trends are more like forest fires: There are thousands a year, but only a few become roaring monsters. That’s because in those rare situations, the landscape was ripe: sparse rain, dry woods, badly equipped fire departments. If these conditions exist, any old match will do. “And nobody,” Watts says wryly, “will go around talking about the exceptional properties of the spark that started the fire.”

So successful campaigns are just pot luck then? Explain that to the customer you’ve just sold your ideas on viral internet marketing campaigns.

Read the Full Article.