🚀 go-pugleaf

In article <Feb.27.16.48.24.1991.19818@presto.ig.com>, mcb@presto.ig.com (Michael C. Berch) writes:
|> As a policy matter, however, I would criticize AT&T for failing to
|> give notice of its patent claim at some reasonable time between 1985
|> and today, since it may have afforded the opportunity for the
|> developers of X to (1) evaluate the probable validity of the claim;
|> (2) evaluate its effect on current work; and (3) determine whether a
|> non-infringing implementation should be substituted.

About three to four years ago while I was a member of X3H3.6, the ANSC comittie working to standardize the X Window System I asked Bob Scheifler if he was aware of the AT&T patent covering layers which would seem to cover the X11 backing store concept as well.

Bob indicated that he was aware of the AT&T patent and that he felt that the MIT method of implementing backing store was not covered by the claims of the AT&T patent.

I don't think that you can fault AT&T here.  The X Developers were aware of the Pike patent at the time they were at least implementing the backing store portion of the X Window system.  This does make it difficult to say that MIT thought it all up themselves.

For the record I disagreed with Bob.  I believe that the MIT implementation of backing store is covered by the Pike patent.  This is a separate issue from whether the Pike patent is valid.  I was not building window systems at the time the Blit was built so without research I can't say if Pike really did something new in creating the idea of rendering occluded bits to offscreen memory.

One clarification to several other poster's comments.  The Pike patent covers the notion of backing store and not save unders.  Saving the area of the screen you are about to cover with a window so you can restore it without requesting the application to redraw the area is not covered by the claims of the Pike patent.  The innovation is in rendering the contents of the occluded portion of the window while it is still occluded so you don't have to request a redraw.  These are very different operations and t




he prior art of a save under will not invalidate the Pike patent.

Richard J. Greco
Intel Supercomputer Systems Division
rjg@ssd.intel.com