Seek Nuance

EMC has released one update to Retrospect 8.0 since my review. They fixed a slew of bugs, but it’s still not yet ready for prime time. Don’t buy this product.

I haven’t kept records of the bad software releases of my life, but I think this has been the worst software release I’ve ever experienced. I’m inclined to fault the company’s management, and not the development team. Someone wanted this out the door before it was ready, for reasons doubtless having to do with company politics or a quarterly status report.

Rule #1 of any development: Do Not Shortchange QA. Unless you are far down testing’s typical negative exponential curve, $1 of missed QA will result in at least $10 of support costs.

Their support forum has a topic devoted to experiences with Retrospect 8.0 in production. I mildly lost it, and said this:

A company rushing to release something before it’s ready is a company trying to wish away the laws of physics. Every time I’ve seen QA time minimized by wishful thinking, I’ve seen a train wreck. Whatever the pressures were to release prematurely, someone now ought to make an accounting of the cost of the damage to Retrospect’s and EMC’s brand reputation, and the internal costs of the thrashing and running around that must be going on now to fix bugs “yesterday.” With the benefit of hindsight, was the decision the right one? And is anyone surprised at what happened?

I sympathize with your position. But the dividends paid by a quality release are numerous and diffuse, and last a long time. I’m surprised that in this day and age there are still some managers that haven’t woken up and smelled the coffee.

End of soapbox…