Re: Suggestions for improving the 7.x OnLine Engine installation and configuration

Hi guys! **Warning** I have no experience whatsoever with 7 engines, so I'm wandering in the dark, here. Our is a small shop as far as Informix goes (we're big clients in other areas, though) and moving to 7 is far too expensive for us compared to the benefits we would have (This was being discussed in another thread, wasn't it?) My impression is that Tim, Nils and David are partly right: - yes, something has to be done about the installation process and configuration tools - yes, vendors of unix based software have always seen nice, flashy (...and consistent) user interfaces as something professionals didn't need and didn't want; but, hey, look at Bill, push a flashy GUI and people think you have a good product. Consistent and comprehesible user interfaces have been neglected for too much time in the unix arena. - yes, OnLine has to handle different user requirements and on far too many etherogeneous platforms However, given that the guru has already configured partitions and rebuilt the kernel with appropriate (or at least approximate) semaphores and shared memory parameters, I don't see it as impossible to have an installation / (don't get me wrong! char based) configuration wizard (You know those dialog boxes with a nice 'next' button on it? Bill is not always wrong :-) that creates everithing the right way and puts everything in the right place based on user responses. After that, of course the guru could always resort to ONCONFIG, if only to feel a better professional than the next guy. As for the installation bit and the TEN business, I don't see it that difficult to have separate directories for T, E & N, and have each installation script to check the dates of each common file and install only those that are more recent than the ones already installed. In the unlikely event that the above is too difficult top set up, then, given Informix prices, the least I would expect from Informix is a *port specific* installation process. This would probably be easier to maintain than a huge do-it-all monster. The real problem (and being the one that asseses rem's new acquisitions, from copying machines to linacs, I see this every day) is that vendors don't produce and sell what the market wants, but what they have or what *they* think the market will be attracted to. More than that, most of them don't listen to customers once the good has been sold. In areas where rem is a big customer, at least, in Italy (e.g. radiotheray and nuclear medicine), the following examples come to mind - GE and Elscint have joined forces in a new NM venture. For far too long GE has tried to sell 10-years old technology without adapting to prospective customer needs. - Siemens and GE have both dropped their flashy RT planning systems, and bougth someone else's - After two years of customer complaints from all over the world, Siemens has not changed some of it's very ridicolous features (I won't go into details) of its NM software. (and I could go on forever) As for Informix, Cdi is full of feature requests pertaining to all of T E & N areas, which Informix has chosen to downright ignore (you don't want me to name them all, do you? :-). They instead have concentrated on pushing the IUS (which many of us will probably never see, given the raise in price...) and to a smaller extent, new era. On the other hand, the fact that Informix has moved a step back and decided to continue distributing char based tools together with the engines (so says Carlton Doe), is in fact a small step forward. Ciao, Marco PS: Ah, Billy, it's not you I was talking about. I wouldn't dream saying that you are not always wrong :-) ____________________________________________________________________________ rem radioterapia, which I immeritately manage, seldom agrees with what I say marco greco (Catania, Italy) Work: marcog@ctonline.it rem radioterapia 39 95 447828 fax 446558 (was mar.greco@agora.stm.it) Achea 39 95 503117