scott Ford wrote: >What about theft of software products ? We all know it goes on ... >How do you prevent it ?
When I see someone assert that "We all know it goes on", I don't see it as a meaningful statement. Spontaneous fission also occurs, but not enough for me to worry about it. There have been a few war stories shared here of folks who found a cheater; I'd submit that the scarcity of such stories (in mainframe-land, not talking about PCs!) proves the point: this isn't an issue that's big enough to waste cycles on. As Timothy has suggested, shoplifting happens, but stopping it isn't a store's highest priority, at the expense of alienating customers, because it isn't a big enough problem (with notable exceptions, like liquor stores in certain neighborhoods). Enterprise shops stealing mainframe software happens, but stopping it at the expense of alienating customers (and costing a lot to boot) doesn't make sense because it isn't a big enough problem. Now, Charles raises some more substantive issues, like never-ending trials. A related problem is with performance software, for which it is really dangerous to offer trials, because "clever" customers will use the trial time to solve their performance problem, and then say "No thanks". And it also depends on how critical the software is: if CorreLog stops working, everyone is irritated, but the system probably doesn't go down, production continues, etc. (if perhaps slightly degraded, and with the ability to head off problems seriously damaged). But it's not necessarily an all-hands-on-deck issue. If, say, your product was a TCP/IP stack, you'd be in a different situation: if it goes down, the world effectively stops. So you might find shops willing to skate a bit with CorreLog who would not with our hypothetical stack product. That doesn't mean CorreLog is less valuable/useful/necessary, but it does inform the decisions about risk that Charles' management has to make, and the steps they may need to take to mitigate those. I applaud them for not wanting to cripple the software for trials, but find myself thinking that in their shoes, I'd be far more willing to do so than for that stack product. But it still wouldn't convince me that keys were needed for production. Seriously-I've heard the assertion for over 30 years; show me some evidence. One or two anecdotes don't qualify. And I've spent more than 25 of those 30 years selling mainframe software without keys, so either we left a lot of money on the table or we'd've noticed the theft. (Anyone who has always used keys doesn't really have a basis for making such an assertion, BTW, now do they?) ...phsiii (NOT trying to start a pissing contest, just honestly baffled by the seemingly baseless belief) P.S. Who here doesn't believe that, given a reason, (s)he couldn't break CPUID checking in a few hours anyway? ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
