What WeatherSA could have done right

brad — Wed, 05 Nov 2008 08:31:29 +0200

Weather SA is promoing it's new and "contemporary" website. Twitter has been a-filled with comments about exactly how badly the new website has been executed (maimed, murder, or vomitted would be equally applicable). A friend mailed them with some very terse comments, and of course they got defensive, believing the lie they had created for themselves that "We hv jst creat'd and AWESOME website!!!11!!!1".

It is hard not to be terse. It is hard to be constructive in your criticism. It is like when someone shows you their 'best' work, and all you can think about is how badly suited they are for the job they are doing. You don't think they are an idiot, you just think they should stick to what they know.

So, lets start with that: Weather SA, you know WEATHER not Web Development|Design. You have done something wrong when acknowledged experts (I am not claiming to be one, but have a look through the twitter search and you will notice some well known names) in the field cannot hide their contempt and cannot be constructive in their criticism because, well, you have given them NOTHING to be positive about. Equipped with the technology and tools of today you have created a website that people in the 90s could only dream about. Where did they go wrong: Everywhere.

How would I have done it, if I had been in charge?

Focus on how to improve the current site layout:
- Use focus groups and use cases to improve the 90% of the site that people use on a daily basis
- Simplify and improve navigation, minimise the breadth and depth of the navigation, and develop focus areas
- Work hard on a version that is simple and easy to use on a mobile phone
- Hire a KICK ASS web designer to work on a clean, beautiful, functional layout that degrades gracefully for blind, mobile and low-res users
Weather SA has lots of relevant, interesting data, so focus on getting it out there. I would:
- Build solid, customizable, RSS, JSON, XML, etc. feeds of well structured, usable, relevant data
- Build a RESTful API for interacting with the data and services of WeatherSA, using a key based access mechanism to limit access based on license agreements and limit load on servers
- Investigate data licensing (Like, I dunno, the CC's BY-NC-SA) to encourage the tinkerers and interested parties to dissect and innovate with the data, without comprimising revenue streams from commercial outfits like SABC, 24, etc.

I think that is about it. The rest would fall into place because the guiding philosphy is sound. It is based on thinking forward and acting in the communities interest. It would hopefully lead to limited whizz-bang (the new site is all about the perceived whizz-bang). The beta site is a degradation from what was a mediocre/average website. The new site limits and/or prevents access to mobile, visually impared, browsers on low resolution and high resolution screens. It angers me that a resourced entity like Weather SA has ignored the expertise in the country and decided that going in-house on this one was the best approach. There are a couple of easily discovered/consulted eminent web thinkers in SA who could have catapulted the web interface into the future.

Instead we have yet another "retro" site in the SA web space that hurts the eyes of on-lookers, and probably violates access to information rights provisioned in the constitution. I wish people could be arguing about how they approached the problem of X or Y, instead of impairing people from being able to suggest improvements, because the previous iteration of the site is an improvement. I would love to hear your thoughts! Am I being overly critical, and a doos? Lets start a discussion with Weather SA, and see if we can get them to change direction!

progress

What WeatherSA could have done right