To: jag, patl, owen, rxb, mackay, powell, hvr Subject: Re: Status of NeWS --text follows this line-- Date: Fri, 1 Nov 91 10:41:55 +0100 From: knut@forit.forut.no Knut, I just quit my job at Sun because Sun's managment is living a lie. Their words say they support NeWS, but their actions testify to just the opposite. There are many people in Sun who want quite badly to kill NeWS, and they have been trying to do that while at the same time giving lip service to NeWS, so they don't get fired. McNealy and other top level executives say they support NeWS, and say they will fire anyone who tries to kill it. So everyone pretends to be behind it to keep their job, while continuing to sabotage it. It is a very unpleasant and unfortunate situation. Pat, Since the early days of Sundew (1986) my hopes for NeWS and TNT has been very high. IT is good to see some from within Sun still is advocating the system in netland. The activity on comp.windows.news has been low over the last months. Bad rumors frequently pops up in my mailbox. The company is criticized for not pushing NeWS hard enough. NeWSPrint and FrameMaker are good signs. More is needed. A bewildered market needs convincing applications. Where has HyperNeWS gone? Sun was going to buy HyperNeWS from Turing and develop it into a wonderful environment. But the VP in charge of making the decision missed the deadline, so he gave himself another deadline, and then dumped it (so it didn't look like Sun dumped HyperNeWS just because he missed the deadline). Can developers risk their future on NeWS/TNT. Well one of the main reasons Sun isn't dumping TNT is that Frame would sue them if they did. So I don't know how much of a future there is for NeWS. I don't know when the managers at Sun will decide it's worth getting sued by Frame in order to get rid of NeWS (or figure out a way to dump it that they won't get sued). The sad fact of this industry is that right or wrong in highly technical matters seems to be settled by marketing success or failure. Remember Cobol. X10 and then x11 did - regardless of technical merits - hit the marketplace just when "open windowing" was desperately needed. It has by now a position similar to what COBOL had in the late 60-ties. Its documentation is said to encompass some 6000 pages. I have no idea of the source code volume. It is a heavy burden for both human and machine resources. Personally, I hope Sun still believe in its role as a defender of conceptual economy and its power to maintain a leading role in establishing industrial standards based on scientific sound principles. If good and bad is mainly defined by marketing powers computing is more like politics and less high technology founded on science. Sun Microsystems is in business to sell hardware, and SunSoft is in business to supply software for Sun Microsystems to run, and maybe make a few bucks on the side selling the same old shrink wrapped shit to whoever else will buy it. The main goal of SunSoft's main customer is to sell as many tons of hardware as possible, and big bloated software sells big bloated machines. Read the DOE/DOMF announcement. Draw your own conclusions. I believe we have seen in the past that software dinosaurs are difficult to hide under the carpet. It is simply not enough to feed them more cpu cycles. If Sun gives up on actively promoting NeWS and TNT as an API, the stand of the company in the advanced industry and scientifically inclined market will not be the same. NeWS in many ways represents the essence of software enginering research over the last 20 years applied in the fields of interactive graphics. In my opinion it has great potentials for object oriented (OO) databases applications too. In distributed applications, the NeWS/PostScript interpretative OO methods offers an new opportunity to use object instances and their class definitions as dynamically defined standards for information interchange. Hence, we could exchange data and their meaning as one entity (objects) in open network environments instead of interchanging data based on a priori agreedments on their interpretations. This will give us much needed flexibility. That's one of the reasons we were given for Sun dumping HyperNeWS: HyperNeWS is *very* object oriented, and if we come out with a product *that* object oriented, *so* soon, before the 100% vaporware DOE/DOMF has had a chance to crustalize into millions of lines of code and link into every application on the desktop, that would confuse the customers. Sun is "commited" (in the institutional sense of the term) to delivering a totally object oriented environment in the future. HyperNeWS is much too far in the present to be compatible with those goals. I hope rumors on tipping NeWS and TNT in Openwindows3.0 is simply not true, - or is it? False rumors. TNT 3.0 is included with OpenWindows 3.0. It will take them quite some time and several lawsuites to get rid of NeWS. I would appreciate if Suns officially announced commitment with regard to NeWS/TNT for the upcoming releases and the Solaris announcement could be summarized and posted. There was a recent Solaris announcement that mentioned NeWS and TNT. It is just a statement of fact that NeWS and TNT are included with OpenWindows 3.0. But if you hear anything saying that sun *supports* NeWS and TNT, that is just a lie. Knut Skog Professor of Comp. Sc. Senior research scientist PS In order not to serve speculative interests I do not post this to the net. Feel free to use this statement of opinion as you like. DS Thanks -- please don't post this reply either. I feel I owe you this explination of my opinion, because I have told you different, more hopeful things in the past, based on my opinion that was formed around the lies I was told. Of course, this is just my opinion, who am I to know what's *really* going on? The people in charge certainly aren't telling.