Date: Wed, 3 Apr 2002 08:44:53 +0200 From: "Anthony Atkielski" <anthony@atkielski.com> To: "Mike Meyer" <mwm-dated-1018213068.da8dcb@mired.org> Cc: <chat@FreeBSD.ORG> Subject: Re: Anti-Unix Site Runs Unix Message-ID: <012601c1dadb$104d5100$0a00000a@atkielski.com> References: <20020402113404.A52321@lpt.ens.fr><3CA9854E.A4D86CC4@mindspring.com><20020402123254.H49279@lpt.ens.fr><009301c1da83$9fa73170$0a00000a@atkielski.com> <15530.6987.977637.574551@guru.mired.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Mike writes: > There's no money in that, so why should > anyone do it? Agreed. > In particular, in an industry dominated by > a company that gained that position by shipping > buggy products and making money on upgrades, > there's no incentive to do anything else. I suspected you wouldn't be able to simply stop at that point without saying something about your Great Satan, and I was right. Microsoft (the company I presume you had in mind) did not invent this paradigm, nor is it any more a practitioner of it than anyone else in the IT industry. It is a problem with EVERY IT company I know of, at least for software (hardware makers cannot afford quite the same luxury, since they can't have people download a new processor if the one they shipped is massively defective). > The end result is that the only people interested > in shipping a bug-free product - as opposed > to one that they can get people to buy - are > the people who aren't trying to sell the product. Customers have never demanded bug-free products, and so they don't get them. If making products bug-free would sell more of them, that's what software companies would do. But you cannot sell software successfully on that basis, and so feature bloat is used instead--customers seem willing to pay for additional features, even features that they don't need. Other things that fail to sell software include higher security and better performance. This is why you don't see higher security and better performance in new versions of software, as a general rule. If customers really demanded security, performance, and reliability, they'd get it, as it costs no more to produce these than it does to produce feature bloat. > No, what makes a product successful is selling > lots of copies. You can only sell lots of copies if you can keep your technical support costs down. That imposes an upper limit on the number of bugs you can ship. And that limit is considerably lower than you seem to believe. > Being crashfree doesn't do that *nearly* as well > as convincing people that there isn't an alternative. Most people who have moved to OSes like NT and XP have done so because of better resistance to things like crashes. I didn't particularly care for XP because I don't like the "activation" feature (a euphemism for making you pay forever for the same software), but I have to admit that it has been completely and totally crash-proof thus far ... nothing seems to bring it down. It seems to be as reliable as NT (no surprise, since it is _based_ on NT). > Last time I checked, Linux had more desktops > than the Mac. Linux had zip, last time I saw numbers. Servers don't count when you are considering desktops. Similarly, FreeBSD is far more present among servers than among desktops. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-chat" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?012601c1dadb$104d5100$0a00000a>