The vibe coded aspect of this article seems quite irrelevant to me. If I read this correctly, an exec team (or just CEO?) got sold a terrible piece of software without proper vetting. Very bad software has existed for a long time, and this is an age-old story. Vibe-coded garbage perhaps has increased the volume of terrible stuff to wade through, but is secondary to a familiar story of dumb leaders making dumb decisions.
I worked for a few years at a 100+ year old privately owned (same family) B2B supplier with insane profits. Website was outdated but highly practical, sales/CRM (if you can call it that) systems were mostly command line and hadn't changed fundamentally since the 1980s. These systems worked, and any proposal to change anything took months of meetings and debates and review of every cost/benefit possible. Proving that a change directly translated to a clear revenue metric was nearly impossible– for at least this niche, would more modern sales software actually translate to more orders? (answer: not really, a question reanalyzed every few years in depth). Would a nicer website get more conversions? (also no, something A/B tested to death every few years). Changing the position of one product grouping by a few pixels might be a 6 month job, lol.
By contrast, their fulfillment center was cutting edge, highly automated, and relatively experimental– if it improved the speed and cut costs, they jumped right on. These are much easier to measure as profitable.
>It's quite possibly the smoothest and best running GIS software available today
lol, the bar is not high. It can be both the smoothest and extremely janky at times. Let's be honest with ourselves here. (and I do agree, it's among the best running... but also janky).
+100. There is very little QGIS cannot do as well or better than ArcGIS. For any shortcomings, there are generally other specialized tools that can fill the gaps. It's really just a training issue more than technical one at this point imo.