Visicalc at Commodore (Score 1)

In the early 1980s we used Visicalc (on CBM-IIs) extensively at Commodore Business Machines. Spreadsheets could do what calculators and accounting paper could not: provide a fast summary of quarterly sales in the U.S. I used a spreadsheet to tell if we were on target to make EPS (earnings/share) projections. The Commodore product line was very small, so a 200 line spreadsheet could handle the job. All the financial guys (back then just one woman) knew Visicalc's results should be compared to IT reports. "Did it make sense?" if not, Jack Tramiel, CEO, would show you the door. One time a computer report dropped a leading "1" as Commodore sales passed the $100M mark. We found the error by comparing results on a Visicalc spreadsheet. Care in finance and accounting goes in every direction. The guys from Arthur Anderson had Jack Tramiel's ear, once he realized how stock prices were tied to year-end results verified by AA auditors. It takes one auditor to know one. It didn't take a spreadsheet program for Jack to know he needed those finance guys around him. Yes, spreadsheets made financial modeling much easier, but one shouldn't blame wrong results entirely on the lowly, ubiquitous spreadsheet. By the late '80s the IBM PC had become powerful enough to run newly developed sophisticated econometric and finance risk-based models for financial researchers and the newly minted MBAs bound for Wall Street. If the details of these models are based on poor assumptions, they were hidden in the software and there mistakenly forgotten. New technology always runs ahead of our wise understanding of its limitations and proper use. But it is surprising that it has taken 25 years and a world-wide financial meltdown for us to realize how great the risks were of a surfeit of poorly understood financial models.