cavrak@uvm-gen.UUCP (Steve Cavrak,113 Waterman,6561483,) (07/02/89)
The other day I watched a friend port about 400 lines from Turbo Pascal on a PC to Lightspeed on a Mac ... Besides the horror of discovering that the Mac has a problem with the line-feed character, and that the European quote caused Lightspeed to hang, Wes discovered the long list of "features" that "does not compute" (pre initialized "variables" for instance; something strange about the undocumented "byte" type, etc.). The conclusion was that this is "Fortran" all over again, and that programmers have to adhere to "standards" while the manufacturers lure them away with convenient features. If we want the new toys, that's the price. Steve (p.s. I, for one, want the new toys!)
dat0@rimfaxe.diku.dk (Dat-0 undervisningsassistent) (07/06/89)
cavrak@uvm-gen.UUCP (Steve Cavrak,113 Waterman,6561483,) writes: >The other day I watched a friend port about 400 lines from Turbo Pascal on >a PC to Lightspeed on a Mac ... >Besides the horror of discovering that the Mac has a problem with the >line-feed character, and that the European quote caused Lightspeed to >hang, Wes discovered the long list of "features" that "does not compute" >(pre initialized "variables" for instance; something strange about the >undocumented "byte" type, etc.). >The conclusion was that this is "Fortran" all over again, and that >programmers have to adhere to "standards" while the manufacturers lure >them away with convenient features. >If we want the new toys, that's the price. I don't get the point in your posting. Are you pro or con ? As far as I can tell your above example illustrates quite well why a standard is to be prefered. And why it is well to stick to it, if you want to port your programs easily. If you have to use non-standard features, stick them neatly away in small procedures, so they are easy to find and replace. Kristian Damm Jensen (dat0@diku.dk) Institute of datalogi, University of Copenhagen (DIKU) Universitetsparken 1, DK-2100 Copenhagen \, Denmark Kristian Damm Jensen (dat0@diku.dk) Institute of datalogi, University of Copenhagen (DIKU)