Slashdot Mirror
Does Unisys Really Get It?
Joe Barr writes "There's an interesting story on NewsForge today about Unisys and its new-found love for Linux. In the story, Robin Miller interviews Unisys VP of engineering Anthony Gold and asks such delicate questions as how Unisys 'planned to make amends for its use of GIF patents against open source projects'? It's a good read, and in this day and age of software dinosaurs trying for peaceful co-existence with Linux, a very timely one."
GIF was developed as an image format by Compuserve (I think). They used LZW compression, which was a Unisys patented compression algorithm, but they never notified, let alone licensed LZW. Unisys, being the mainframe shop it was, didn't even notice for a long time. When they did, they pressed the patent issue.
The important point is that they didn't press the LZW patent because of it's use in GIFs, per se. They pressed it because it was an important patent to them at the time for other compression purposes. Unisys was big in air-traffic control and a variety of other communications applications that used LZW for dynamic, real-time, streaming compression (which is about the only application it was still good for when you look at the other algorithms available, even then). If they didn't press it on LZW in GIFs, they could have lost the ability to press it in those other lines.
Remember that uncompressed GIFs did not infringe, they asked only for licensing on commercial use, and as the article states they did make not-for-profit licenses available. Questions like "What if a licensed program made the GIF, but I editted it with Gimp" made no sense to them. (Besides, Gimp didn't exist then either ;)
I was a Unisys systems programmer at the time, and I didn't know much about "open source", so I suspect Unisys management didn't have much of a clue either. I seriously doubt they had (or even now have) any idea how the community reacted to that issue. No one ever really grasped that it was Compuserve's fault for pushing a patent encumbered format (with the patent held by someone else) as a "standard". Unisys was caught by surprise and acted in the way most any company would at that time.
Times change. I learned about open source (and TCP/IP, and Unisys' brand new C compiler, and how a 36-bit word architecture complicates a C libarary implementation) by porting a subset of NCSA's httpd server to OS/2200, partially because the GIF issue introduced comp.sys.unisys to open source. (iirc, this was when Mosaic was the browser and the "graphical web" was barely graphical at all.) Some time later, Unisys demo'ed their ClearPath model for us and said "they were even working on a web server for it". I wasn't there, but a co-worker donned a medieval french helmet and said "I doubt if we're interested. We already have one, you see!"
So Unisys and I went different ways (although I did get a "wayback-machine call" for Unisys help last week). I moved on into UNIX, self-taught by downloading Slackware and trying to get it to do anything on Token Ring. Unisys moved on to buy Burroughs and into the "datacenter push" of Microsoft. One of us apparently learns quicker ;) Still, the 2200 was seriously unstoppable from both a hardware and a software standpoint, and even attemting to use MVS or anything else on the "newer" IBM 390 made me want to run screaming to the hills.
If Unisys has woken up, they have the potential to do incredible things with hardware and helping them release that hardware from the constraints of Windows is a huge opportunity for both Unisys and the Linux community. If the haven't woken up, then they will once the see what Linux can do on their hardware. I was on vacation when they demo'ed the ClearPath and heard I had already written a web server on their previous (non-Intel) hardware and the day I got back a Unisys SE was waiting in my cube to ask "how'ed you do that!". He was fascinated and ate up everything about open source I had to offer (sadly, not much at the time).
Bottom line: What does it hurt us to give them a chance? If they screw up, it's going to be their fault. We've already shown that we can help any company that really wants to play nice. If we snub them for what we see as a past slight to open source, when they probably didn't know what open source was, we present ourselves as being petty about something that we still, as a community, fail to acknowledge was a situation not of their own making.