rca@cs.brown.edu (Ronald C.F. Antony) (12/20/89)
>> As soon as you print on a printer with it's own postscript engine, there .> is hardly any noticable slowdown, there is complete spooling as in any UNIX .> system. > >...and the printer most likely to be used... > >OK, OK, that's not fair. But it does illuminate a point: we're talking about >standard features. Otherwise, I could wail about how easy it is to install >EtherNet support on the Mac. What I consider as an important design choice is the consistent support of postscript. NeXT would be stupid to eat up it's resources to develop another postscript printer. There are enough on the marktet. The important question was, is there a way to provide users with a relatively cheap PERSONAL PRINTER. The cleanest way to do this, without going the way of the image writer, was to make a non-intelligent postscript printer. I do not think, that NeXT really believes, that their printer would be of any use as a network printer, unless you dedicate a whole cube to it. Printers are not standerd, neiter for the NeXT nor for the Mac. In neither case you are bound to the respective manufacturer. The only thing NeXT wanted to make sure is, that there should not emerge a dual standard for printing (postscript for the high end, and something else for the personal range). The only way to do this, was the solution they chose. (Well, they could have built upon the HP-deskjet, but this would have been even slower, and someone who pays in the order of 10k$ for the system, is likely not to look if the printer costs 1500 or 2000$) > I was trying to communicate was that Apple pretty >much puts developers in a straighjacket. And I don't mean that negatively, of >course. I mean it to be the best possible thing they could be doing. Mac would >have failed without the straightjacket. NeXT seems to have the Toolbox without >the conscience. Well, I think this can be answered only over time. In the NeXT technical documentation is carefully documented how the menues should look like, where ne items should be inserted. Then the AppKit removes a lot of decisions from the programer. If however NeXT can keep it's developers disciplined, is another question... Ronald ------------------------------------------------------------------------------ "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." - Bernhard Shaw | rca@cs.brown.edu or antony@cogsci.bitnet
rca@cs.brown.edu (Ronald C.F. Antony) (12/20/89)
>Rather, I was trying to show that putting Ether on the motherboard and not >also including enough open architecture features was probably not the best >design decision. I think what NeXT was considerd was the LAN's in the businesses. About 50$ more per system does not matter (50$=my estimation of the ethernet controler chip) if you consider the price of the machine. However if ppl discover the use of the network without expecting it (they buy the machine without networking in mind) they will consider that it is a real bargain (think of the 500$ you would have to pay for a ethernet card...). This again should be good for the sales. I hope of course, that when it becomes more pressing to have solutions for the issues you raised (networking cities/states over the phonenet/fiber optics) that there will be solutions in form of adapters or add-in boards. Ronald ------------------------------------------------------------------------------ "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore all progress depends on the unreasonable man." - Bernhard Shaw | rca@cs.brown.edu or antony@cogsci.bitnet