gis and their usability

GIS are hard to use for mainly two reasons.
Let’s start with the obvious. Geographic Information (GI) is everywhere! Take a look around; you can use your cell phone to find the closest restaurant, you can use GPS to tract your lawn-mower, you can look at your neighbourhood by using satellite imagery provided in the Internet, you use route planners to detect the shortest route for you to get to friends. Therefore, we need applications to help us understand and work with this information.
Geographic Information Systems (GIS) are by definition such applications; they allow capturing, analyzing, storing, and managing of geographic information. Thus, they allow tasks that could easily be split up for use in several applications, each of those rather complex itself. Thus, it is understandable that they are difficult to use. This climaxed in the 90s, when Thrall & Thrall (1999) mentioned in their chapter on desktop GIS software in the "GIS Bible" that ESRI itself recommended three years of training to become fully proficient in Arc/INFO.

The other main reason is the way they evolved over the years. In the early years, only very few people were interested in GIS and those were mainly people who were personally involved in their development. So, firstly, they were not interested in being usable for "outsiders" and secondly, they were competing with other vendors to define a new GIS standard and thus shook off other developers’ efforts. Thus, we ended up with having numerous proprietary GIS in the 90s, where each vendor claimed to be the best and the standard for work with GI. By ignoring the other GIS, they made work for the increasing number of users very difficult as

• data exchange was very difficult
• the technical jargon used was diverse, to put it mildly

 

In the 90s this was recognized as an increasingly imminent problem. Many new users came from non-technical sides, e.g. were researchers intending to use GIS to analyse and present their data in a more sophisticated way. But they needed to be able to exchange data with colleagues who used a different GIS, therefore the discontentment increased.

This was tackled by the foundation of the OpenGIS Consortium in 1994, which has later been renamed to Open Geospatial Consortium (OGC). Citing their homepage, "The Open Geospatial Consortium, Inc. ® (OGC) is a non-profit, international, voluntary consensus standards organization that is leading the development of standards for geospatial and location based services." They today aggregate all relevant GIS developers and users and aim at describing new standards to make the usage of GIS and GI easier.

Sitting all in the same boat helped to push their work tremendously, so today we see a commonly signed simple features specification and we are headed towards interoperable systems at high speed.

 

Despite these efforts, the overall usability of GIS is still very bad. They still offer

• unbelievably bad error messages
• rather intimidating technical jargon
• ignorance regarding consistency with other software suites

Consider error messages, for example. A good error message should help users by giving a non-intimidating, friendly, and concise message stating what went wrong and what to do to avoid this problem.

Instead, see the next error message for en example of what once happened when I worked with IDRISI. (Unfortunately, I lost my snapshot of the infamous "Segmentation Violation" error thrown every once in a while for apparently no reason at all. If you happen to have this error message pop up, please make a snapshot of your screen and – I’d be so happy to include it here.)