Truth. I came into software just as the industry was shifting away from a seemingly-exclusive reliance on very specialised COBOL-based skills and pursuing the then-“innovative” business strategy of simply disposing of the “excess” staff rather than retraining them. Watching guys who’d put in 20 or 30 years realise they’d probably never work again at the level or for the pay they’d built themselves up to left a searing impression. I resolved to be as wide-ranging a generalist as I could. This worked rather well until the mid-’90s; by 2004 or so, it became impractical-to-pointless. Impractical because there are now so many ways to specialise that each require so much constant learning and growth; pointless because employers would seemingly rather wait for one of the half-dozen people in the history of the planet meeting all their “requirements” to walk in their door, rather than training up some bright young spark. That, in practice, rarely works well.