Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!samsung!uunet!brunix!cs.brown.edu!pew From: pew@cs.brown.edu (Peter E. Wagner) Newsgroups: comp.databases Subject: Re: Market share for DOS database products Message-ID: <74272@brunix.UUCP> Date: 2 May 91 17:00:56 GMT References: <71280@brunix.UUCP> <1991Apr18.230012.12838@dbase.A-T.COM> <73732@brunix.UUCP> <1991Apr30.210621.12574@dbase.a-t.com> Sender: news@brunix.UUCP Organization: Brown Computer Science Dept. Lines: 237 In article <1991Apr30.210621.12574@dbase.a-t.com>, awd@dbase.a-t.com (Alastair Dallas) writes: |> Peter, let's make a deal. I'll rebut your mail then you rebut mine and |> we'll call this off. You can have the last word. Ok, I'm really done anyway. I've smashed you enough and you have responded with good arguments. I'll just make a couple of brief comments. Don't hold back if you need to say anything. |> The press does not write unwarranted attacks; the stories were based on |> reports of problems with 1.0, which had more problems than most released |> software. However, the problems were few enough that the vast majority |> of users might never have found them. Still, you shouldn't ignore those people. People don't want to think that they might be the one that actually hits the bug. As a developer, stretching a piece of software to its limits, I know that I would hit these bugs eventually. Hitting a bug in your software is one of THE most frustrating things in life. I can't blame the press for being critical. |> Let me guess: you don't sell your software in the marketplace. I |> can tell because a) you seem to think that shipping upgrades to |> thousands of users is a trivial expense and b) you appear to believe |> that other commercial software companies behave differently. I know |> for a fact that at least one of our competitors adopts this policy |> when it suits them. As for A-T, we are trying to be more |> responsive. We shipped something like 300,000 free upgrades to 1.1-- |> ignoring our cost of goods, that's probably half a million dollars |> in postage. In addition, our new president, David Proctor, announced |> at 1.1's release that we will be shipping regular upgrades on a |> timely basis (1.2 is due for beta soon and 1.3 is in early planning). |> Isn't that "endearing," or can we do nothing right enough at this point? No I don't sell software the way you do. But I know it's a big expense to ship upgrades. However, I'm sure it's worth it to you in the long run. Tell your accountants to charge it to Good Will. From what you say above it's clear that AT's come around to this realization. This is a good policy, and I'm glad you've adopted it. If AT continues to display such responsiveness, you may be able to win back cynics such as myself. |> |> > ...I subsequently became disenchanted with dBase, |> > because there was no comparison. Foxbase was simply a far, far |> > superior product. |> |> I could not get away with arguing like this. I have a high opinion |> of Microsoft C, for example, over our current compiler. But if I |> went to my boss and told him I had become disenchanted and wanted |> him to buy a "far, far superior product," he would ask for facts. |> I couldn't get away with, "boss, there's no comparison." You're right, I'm not giving facts. I'm not going to dig up trade magazines and stuff just to back me up. Let me add IMHO to the above. Also, I work with plenty of developers and they all agree, so it just seems like general knowledge. |> |> I deleted your comment to the effect that not only is Fox faster, |> but it's more reliable, too, "as it should be." Have you done _any_ |> analysis to prove this reliability? Did you know that when dBASE |> appends a blank record, we write the record and update the number |> of records in the header. Fox postpones updating the header in |> order to run faster on benchmarks. However, if a power failure |> interrupts your session, your appended records will be lost with Fox. |> In general, I would think that greater reliability would tend |> to be slower, not faster, so I don't think your unsupported |> findings are "as they should be" at all. Okay, got me here. I was talking reliability in terms of bugs causing problems. I have no idea about each program's mechanisms for maintaining data integrity. |> |> > Does Fox Software ship products with bugs? Yes, all software |> > companies do. |> |> Ah, good, reality checks in. |> |> > But Fox always sends free upgrades to its customers. |> > They don't try to sweep things under the carpet to save a few bucks. |> > They realize that they will earn even more money by retaining the |> > respect of their customers. |> |> You don't know what things they have successfully swept under the carpet, |> and you don't know what upgrades they haven't sent. In general, |> however, Fox has excellent relations with its customers because |> they have acted sensitively and professionally. Ashton-Tate, on |> the other hand, has earned much of the abuse we have gotten. Fox ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ This is the reality that I see. Now you guys have to work extra hard to win us back. Your products and service has to be A+ now, not just A. |> has concentrated on winning over developers and their customers |> are very loyal. We have a broader constituency and it is harder |> to make everyone happy, but I admit we've made mistakes and we've |> been insensitive. I hope you'll give us a chance to improve your |> opinion of us. You have done much in your responses to improve my opinion of you. However, I am a developer, I like being catered to, and it will be very difficult to win me back with software that also must cater to a broader constituency. I think dBase has gotten too monstrous, 6 zillion disks in the box. I would like to see a lean, mean, robust developer's version and a soft, juicy non-developer's version. I think Foxbase/Pro and Alpha 4 work well together in this way (though FoxPro itself has gotten a lot friendlier for the everyday user). |> I'm curious about your view of the software industry. My view is that |> companies are pretty much the same because the same players move from |> place to place. Many Ashton-Tate people now work at Borland; we just |> hired an ex-Borland employee a few weeks ago. You seem to think that |> Fox walks on water and that the developers at Borland "know exactly |> what they have to provide" but that at Ashton-Tate we're just waiting |> for the life-support to be switched off. No, but at AT you are trying to service everyone with one huge product (see above). I just don't think it should be done that way. You must be at cross purposes at every development meeting. |> |> > Do the software reviewers give Paradox |> > consistently high marks, consistently much higher than dBase... |> |> Data Based Advisor, May, 1991, our ad, page 4-5: Software Digest |> Independent Comparative Ratings Report for Selecting IBM PC Business |> Software. dBASE IV 7.0 overall, no category under 5.0. No other |> product can say that. Paradox got a particularly low mark in the |> Performance category. Software Digest is like Consumer Reports; |> magazine reviewers are like Family Circle, in my opinion. |> I wouldn't buy a refrigerator based on a |> review in Family Circle. But I digress: What specific reviews are |> _you_ quoting, Peter? I must admit, I've been in school for the past year and haven't followed things that closely. Congratulations on the high marks. I'm going to have to go look at those tests... |> That's my point; we agree. Fox is great enough that they have |> established a loyal following, and they have made this industry much |> more competitive in the last few years (of course we helped by |> releasing 1.0 :-). Good is _not_ enough. But wait until FoxPro 2.0 |> or 3.0 ships with too many bugs (e.g. Clipper 5.0). Wait until |> Fox sues someone over its proposed Rushmore patent. Wait until |> Ashton-Tate delivers what we're developing now. The situation |> can change. Absolutely. |> |> > Superior products speak for themselves to those who are qualified to |> > judge the differences. |> |> What's this? All animals are created equal except for the pigs who |> are more equal than the others? Are you really saying that stupid |> people buy dBASE but smart ones don't and therefore dBASE is no good? Nope. But if I go into Egghead and ask for a powerful database, the sales person is going to point me to dBase. He probably doesn't know much about it, but it's the known product. He can feel fine recommending it, and since I, the customer, will probably recognize the name, the sale is that much easier. If I'm not a database expert, dBase IV will probably do what I want it to, and I'll be happy enough. |> |> > The developer community is a good barometer. |> > I know lots of people who have switched from dBase to Fox, Clipper, |> > and Paradox. I don't know of anyone that has switched from one of |> > these products to dBase, IV or otherwise. |> |> Yes, the developer community is a good barometer, but they're not |> the only barometer. The reason the previous management didn't |> coddle developers is that they represent such a small fraction of |> our market, like it or not. Even though they are influential, we |> still sell a lot of dBASEs to people who don't know any serious |> dBASE developers. Yes, you do sell to these people, but I would bet that they would be much happier with a product tailored for the non-developer, like Q & A or Alpha 4. Much of your sales (most?) are due to the fact that you are the market leader. Why are you the market leader? Because you were the first with a reasonably powerful database in the PC market. However, today is a different story. I think there are better products available for the everyday user, and I think there are better products available for the serious developer. I will give you cross-platform compatibility (see below), but that's a really small niche! |> As for your last assertion, you are ignoring |> mail in this newsgroup from people who are using dBASE IV because |> of its cross-platform compatibility, it's standardization and |> it's bureaucratic acceptability. I know people who've switched |> back, but I also know a few with stars in their eyes, too. |> I hate bureaucratic acceptability! It sells a lot of inferior products. Buying dBase and IBM is a no brainer for most bureaucrats. Nobody can fault Purchasing for making these decisions, buying the market leaders. You get a lot of humdrum products this way, however. It sounds like you guys *are* turning things around at AT. That's great, it can only make for better software for all of us. I am cynical about your products, there's no doubt about it. You cater to your constituency, which mostly isn't me, and then say that your product is right for me. That probably won't ever be the case until you develop a product that is targeted for me, not one that just keeps me in mind. I also think that the majority of your constituency would be better off with another product. I`ve seen plenty of people who are using the dot prompt but not doing any development. I think that's tragic! This sort of person should have Q & A or other basic database program, not dBase III (which management bought). They could move up to dBase IV and get menus, but they don't need all the other stuff packed in IV which is geared for developers. I'm sure this is not news to you, it's a problem you've been working with for years. You're first real crack at it, dBase IV 1.0 didn't go so well. I can't say much about 1.1, but I don't think the strategy is any different. Let's see what AT can do in the future. I'm sure you'll still be around. Okay, I'm done. Peter -- ---------------------------------------------------------------- Peter E. Wagner (401)863-7685 pew@cs.brown.edu Department Computer Science Box 1910 pew@BROWNCS.BITNET Brown University, Providence, RI 02912 uunet!brunix!pew Woody Allen when asked if he thought sex was dirty; `If you do it right.' ----------------------------------------------------------------