Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!tut.cis.ohio-state.edu!giza.cis.ohio-state.edu!francis From: francis@giza.cis.ohio-state.edu (RD Francis) Newsgroups: comp.sys.mac.system Subject: Re: It's off until first half of '91 Message-ID: <84193@tut.cis.ohio-state.edu> Date: 28 Sep 90 21:10:19 GMT References: <0093D408.262D5B40@uclapp.physics.ucla.edu> <1027@mdavcr.UUCP> Sender: news@tut.cis.ohio-state.edu Organization: Ohio State Computer Science Lines: 73 I'm going to have to take the side of Apple as well, here. Why? 1) Lots of people have talked about the shipping dates. In my opinion, you either announce a shipping date, or you don't. Announcing about when you'd expect that shipping date to be *isn't* announcing a shipping date; announcing that the product will ship on Mar 1, 1990 is announcing a shipping date. The latter is a promise (which, as has been mentioned, experience has taught as will probably be broken in the Mac world); the former is an informal guess to the obvious question "When will we get this marvelous thing?" I don't think Apple has ever promised a date on this thing. We can bitch all we want about how long it's taking, but I don't think it's reasonable to criticize them for not living up to a promise they never made. 2) Hard as it may be to remember, they did do this to help their developers (and it was announced at a developers' conference, not just in some relatively generic press release). How many people remember System 6? Not just 6.0, but 6 in general? As I recall, some of the bugs seen when System 6 came out were attributable to people not getting their programs into line with Apple's rules in time rather than 6.0. 7.0 is going to introduce some major changes; by being up front as soon as possible, Apple lets developers plan ahead so their programs are more likely to be compatible. Some widely-used utilities will be useless, or next to it, in their present form (TOPS, Suitcase, On Cue, etc.); the developers of those utilities have had time to figure out what to do about that. Finally, some features will enhance the way programs can work; this has given developers time to figure out how to use those capabilities. As far as that last point goes, I'm sure that some development houses would have worked on items similar to what Apple's doing; if Apple had kept that a secret, a developer might look at 1000's of man-hours as being down the drain, as they would have been able to use Apple's built-in hooks if they'd known they were coming. Damned if you do, damned if you don't Yes, I'm as impatient as the next guy for this to come out. Yes, there are people to whom I recommended waiting for a feature of 7.0 rather than buying something to work in the meantime, and yes I regret that somewhat. I've worked on enough projects to know that, when you're doing something new and innovative (even if it's only new and innovative to you, and not to the world at large), that Murphy's Law type adage is true -- figure out how long you expect it to take, double the number, and move up to the next highest time unit. Well, OK, that's still hyperbole, but you can basically assume that even your wildest guess as to how long it will take to complete the project will be wrong. Also, think about what beta test means (well, should mean). Beta-quality software should mean software that seems like it's about ready to go out, but needs some people to try it and find the bugs that the programmers haven't. Look at all the software that runs on a Mac; if 7.0 is at a level that I would call beta-quality by now (posts have indicated this, possibly erroneously), it's been run under a lot of conditions, and seems to work. There's a lot to test, and there are thousands of pieces of software to be used to test it. Each little init brings up the possibility of a bug. Every time 7.0 alpha ran into a problem with an init, someone would have to dig around in the system and in the init, presumably, to try to figure out if the problem was in the init or in the system. I'd bet that for every system problem turned up, at least 2 or 3 init problems turned up. Not to mention that they've had to copy alterations made in the system for the fx, maybe the ci, and presumably the 3 new machines. Something like that could easily wind up killing anywhere from a few days to a month, depending. It's an incredible task. Let's try to be patient, and let them do the best job they can. Hey, don't you think all those Apple people want to have a system 7.0 to play with/make their lives easier at *least* as badly as you do? -- R David Francis francis@cis.ohio-state.edu