glenn@welch.jhu.edu (Glenn M. Mason) (05/02/91)
------------------------------------------------------------------- Posted as a courtesy ... email responses to: bitnet%"devries@nauvax" ------------------------------------------------------------------- I've been following with interest the debate with Allistair about A-T vs everybody else. Unfortunately, I can't send to comp.databases from my system. I thought you (and perhaps the list) might find this interesting: When I was at the Paradox User's Conference a couple of weeks ago, Borland was of course blowing it's own horn. One particular item for gloating over was market share. According to Borland, Paradox' market share grew last year from 9% to 39%. They were not specific, but I assume that they are talking about current units being sold, not units sold since IBM PC's came into existence (which is what I think Allistair means by 55%). For what it's worth, I started out my life in databases with dBase II on an Osborne 1. When I switched to the IBM world my work didn't include using a database until 1985, around the same time that Paradox came out. I found Paradox very easy to use and yet having all the power that I needed. I bought version 1.0 in November 1985 and have been a steady user since. As the product has grown, I've come to realize just how much power was missing from the original, but it's still been a good decision for me. I've since had the opportunity to use dBase III+, and I've had the chance to review dBase IV-1.1. I am not a developer, but an end-user, and I made my decision to stay with Paradox based upon how the products worked and how I related to the documentation. Paradox works the way I work and I find the documentation well organized and easy to understand. These things have not been true for me with any version of dBase. The issue of bugs is important (and I wouldn't use a buggy product not matter how well it fit my other criteria), but it was just one more reason to stay with Paradox, not the primary reason. Allistair is right, no company ships a bug free product, but I never ran into any bugs in Paradox until version 3.5 (and Borland has already fixed it and sent me the fix). I did not ever use dBase IV, but the bug reports did influence me not to use it because it was one more negative on top of an already negative perception. I am currently in the position of recommending databases and spreadsheets to other users. I recommend Paradox as a database and Quattro Pro as a spread- sheet. Part of my argument is the past unreliability of the major competitor's code (dBase IV) or delays and dumb design decisions (Lotus 1-2-3 2.2 vs 3.1). The focus of both of these companies has seemed to be on suing competitors to protect uncompetitive products instead of competing in the marketplace. My experience with Borland's practice is that they put their energies into getting the code as solid as possible and that is who I want to trust with my data. I'm not trying to add to the flames, just to provide some info on how at least one user interprets the negative press that Allistair has complained about and how the decision gets made by non-developers to choose a database. Personally, I'd like to see us get off talking about the good/bad of a product or company and get on with helping people make the best use of the products they have. It really has been an interesting debate, but hopefully we'll move on. +======================================================+ | /\ Ernie DeVries bitnet%"devries@nauvax" | | / \/\ Academic/Personal Computer Services | | /\/ \ \/\ | | Northern Arizona University "The Mountain Campus" | +======================================================+ -- Senior Research Programmer Analyst Laboratory for Applied Research in Academic Information William H. Welch Medical Library, The Johns Hopkins University School of Medicine