Path: utzoo!utgpu!news-server.csri.toronto.edu!rpi!dali.cs.montana.edu!uakari.primate.wisc.edu!zaphod.mps.ohio-state.edu!wuarchive!rice!uw-beaver!milton!basiji From: basiji@milton.u.washington.edu (David Basiji) Newsgroups: comp.sys.next Subject: Re: New HP machines Bad News for NeXT?? Message-ID: <1991May23.181115.6929@milton.u.washington.edu> Date: 23 May 91 18:11:15 GMT References: <1991May20.143857.1071@hulaw1.harvard.edu> <35275@athertn.Atherton.COM> Organization: University of Washington Lines: 43 dlw@Atherton.COM (David Williams) writes: >>predecessor of NextStep was developed under HP-UX at HPLabs >>(under the name RMG), and then given/sold to Stepstone. >Yup, developed by ONE awesome hacker by the name of Charles Young. This guy >wrote down to the metal (optimized for each particular graphics card) on >the HP series 300 680x0 boxes. This stuff just blazed, thought supposedly >the internal code made wizards cringe (I can't comment as I'm 1. not a >wizard and 2. never really looked at the code). This stuff was >incredible...not only could you wire up an application a la NeXTstep but >you could have a DAG drawn of the applications functions and when you ran the >interface you could see it step thru the functions. You could have code >running and arbitrarily instantiate a graphical interface widget a meter, a >vu meter, a thermometer and graphically "hook" them to the running code and >see the output change. >Charles works at Hp Labs in the medical instruments group and just hacked >this baby together to allow them to prototype the front panels of >instruments being developed. Apparently there was some similar work done >at SGI by Haberli(sp)?. >If Charles were turned loose on a NeXT I'm sure we would get the Visual >design/Visual programming that is missing from Interface Builder. By that I >mean it is GREAT that IB lets you graphically lay out the GUI...but why >stop there? Why not be able to graphically express the rest of the program? >Just wire it up! >David Have you folks ever seen LabView for the Mac from National Instruments Corp? It's similar to your concept of wiring up the program. It's optimized for data acquisition from experimental apparatus and its subsequent analysis, but I've managed to get some decent instrument control out of the thing (with both of us kicking and screaming the whole way). You simply wire up your graphical objects in the correct order and context and off you go (hah). Its biggest drawbacks are its speed (sloooowww), its lack of objects and hardware for external control of devices/data and its needless complexity at times. It's a very impressive effort but it looks and drives like it was developed seriously at first and then rushed to market, possibly by a different development team. With more development and run on a decent machine, it would be very formidable indeed. Dave Basiji