On Tuesday 26 December 2006 08:04, Joachim Schrod wrote: > Randall R Schulz wrote: > > ... > > > Shame on us! (By which I mean us programmers and software > > developers!) > > I don't agree with that sentiment.
I'm really only criticizing the job that the software profession (of which I am and have long been a member) has done overall in producing reliable, predicatable, usable and useful programs. > How about: If someone buys a computer and/or needs a setup to use > that "Internet thingy", he or she must either spend effort to learn > it or pay money to buy the setup service. Thus, if you don't know > how to install a Linux computer -- or a Windows computer, for that > matter, go out to your friendly neighbourhood computer shop and ask > them to do it for you, against money. Most of them are eager to sell > you services, strange as it may sound -- it's more revenues for them > than simply selling computer boxes. Once you buy software, you should not be expected to also buy training in the use of that software. Not, at least, for consumer, end-user software and hardware. Buyers would rightly balk at such a scam. Doing so may make sense for certain classes of sophisticated software systems designed for specialists, but not for the ubiquitous computing needs that consumer systems satisfy. Basically, inscrutable software and hardware does not call for a coaching service for end users, it calls for redesign of the products to make them more usable. Suggesting that vendors should sell service plans to compensate for poor quality software and hardware is reminiscent of what caused the crisis in the U.S. auto industry in the eighties. At the time, Japanese cars were so much more reliable than their American counterparts that they took a big (and lasting) chunk out of the Bit Three auto makers' market share. Detroit got the message and improved (though they're still not up to the level of quality and reliability of the Japanese auto makers), but there was a lot of ill will generated before they did. > Thus the tip for those with setup and install trouble is: Buy > service, just like you need to do with most of the other > technologies that Randall cited above. Computing is not different to > that. Why should it be? I think you misunderstand my criticism. I'm not talking about service or maintenance or repair. I'm talking about simply using the software for its intended purpose. After all, software doesn't really break. The closest it comes is to become corrupted. Computing systems should detect such corruption themselves, not force a diagnostic procedure (or purchase) upon the user. And from the service perspective, buying it might be good advice if what you could buy was really useful, but for one thing, telephone support is and probably always will be a joke. A certain class of problems could probably be remotely diagnosed directly (via Internet connection), without requiring the owner in need of help having to be heavily involved, but some problems aren't amenable to that approach. If you could purchase real support, where the problem gets fixed, period, that would be great. And consider, too, that this would essentially be an insurance policy, and would only be affordable if the product quality was good enough to begin with. If problems were common, then the cost of the service policy would be prohibitive, and that's probably where we are now. It is because software today is intrinsically so failure-prone that vendors routinely disclaim any sort of warrantee for their products. [ As an aside, I'll mention that I was really shocked when my dad told me that when he switched from AOL (a six-month subscription was included with the computer by the person who gave it to him as a gift) to a local ISP and had some trouble, they sent someone to his home to investigate the problem! ] > Joachim Randall Schulz -- ( R a n d a l l S c h u l z @ a c m . o r g ) -- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
