If you read this blog, you'll have noticed that I have tried a few new things. I used a little java scripting with Google Maps API, did current affairs and complete history projects on QGIS desktop then served both on giscloud.com, and tried a web service for small oil companies and agencies. I recently returned to la familia with Esri's Kuwaiti distributor, as I had been Esri's petroleum industry manager prior, after geoscience for GIS at Halliburton - in particular a Unix ArcView extension... though the only Avenue (its scripting language) I see today is the Mall of same name in Kuwait City!
Through my travels and travails, I found the single item that sets Esri apart is its Model Builder (MB), which allows to concatenate and package geo-processing tasks. My LinkedIn group discussion asking others' experiences echoes mine that it doesn't come easily, but it is powerful nonetheless, especially when wrapped into Python. Using MB has been a great training tool, the basis of my two most read Slideshare posts. I will revamp it to use the the public RMOTC dataset, complete with surface GIS and subsurface geological and geophysical data.
... don't touch that dial, stay tuned!