-----Original Message----- From: Gerhard Fiedler To: PICLIST@MITVMA.MIT.EDU Date: 09 March 1999 03:39 Subject: Re: Customer Service @ CCS (was re: Bad Disk) >At 20:22 03/08/99 -0500, Ian Cull wrote: >>Do you ever pay any money for maintenance on your car? Or does your initial >>purchase include all necessary future charges to keep it in operational order >>and up to date? > >we're not talking here about "keeping up to date," we're talking about >bugs, which commonly means that the program does not do what it should, not >that it is outdated. if i buy an ANSI-compliant compiler and the ANSI spec >changes, i of course don't expect them to provide me with a compiler >compliant to the new standard. but if i buy an allegedly ANSI-compliant >compiler and i find that it isn't -- what would you call that? > >back to the car: i don't pay when there's something that doesn't work as >advertised in a new car. with most "material" items this is pretty clearly >covered in the warranty (if i'm not mistaken, there is even a law that >requires a minimum warranty). most (if not all) decent software >manufacturer also provide bug fixes for free. how can you call bug fixes >"maintenance"? when you buy a new car and the left rear shock is bad -- >which doesn't mean the car won't run, it just reacts a little bit unusual >in some rare situations, and you can easily "work around" this "feature" by >driving slow --, would you accept this as having to be covered by an extra >"maintenance agreement"? i guess not; you probably would expect this as to >be covered by the product warranty and fixed for free. > >similarily, if i have a compiler and it creates obviously wrong code -- i'm >not talking about sub-optimal code here or changed standards, i'm talking >about code that's not in accordance to the compiler specifications (as >provided by the manufacturer) and advertised claims, eg. a table read >algorithm that jumps into nowhere --, why should i pay for them to fix >that? after all, i bought the compiler based on what they say it does. if >it doesn't do it, they shouldn't claim it. or is that covered by "freedom >of speech"? > >to call "bug fixing" "maintenance" is a marketing trick, and as one can >see, it does work with some. but this doesn't make it real. as i said, if i >use, say, a GNU compiler and hire a guy to fix bugs when they come up in >order not to have to rely on the user community to fix them, =that's= >maintenance, and that's ok to pay for. or if i customize a complex >accounting program and hire somebody (may be from the manufacturer) to >"maintain" my system, which in this case mostly means maintain my >customization and keeping it up to date with ever-changing fiscal >legislation, that's maintenance, too. but when i buy a product, software or >hardware or any ware, i expect it to work as advertised and/or specified, >and making it work so is =not= maintenance, that's bug fixing. > >ge > Software companies use us to test the software, first release (v1.00) full of bugs, second release (v1.01) all fixed, but oh dear its v1.01 so we have to buy it again as its a new version! They send it out faulty as they cannot find all bugs as they cannot create all possible working conditions, if they did it would be out of date b4 it was released (I don't agree with this mind!) Andy.