Scott McNealy's one time right-hand man at Sun has expressed his dismay over the impending acquisition of a company he helped build. Masood Jabbar, Sun Microsystems' former head of sales, told a Silicon Valley conference he was "deeply, deeply" saddened by the prospect of Sun's acquisition by database giant Oracle. "We built a …
Mirror Mirror on the wall
"It's the subtle change that's happening inside your organization that eventually cripple it and most people don't see it coming it," he said calling this condition a cancer. "You must change before you have to."
Right, so how many VP's did Sun decide to create during the last 10 years?
Sun (not "SUN") innovated a lot more than you give them credit for. Sometimes it wasn't about inventing something, but knowing when to ride with a good idea. Sun was a proponent of open systems when that was hardly the popular trend in computing. Sun invented the first practical network file system and remote procedure call mechanism. Sun was a champion of UNIX, and one could make a strong case that Linux would not even be a blip on the graph if it wasn't for that. Sun took the idea of RISC computing and rode it to the top. Sun championed distributing computing back when their competitors couldn't even figure out what "The Network is the Computer" meant.
They also innovated in areas you're probably totally unaware of. Remember Token Ring? Didn't think so. IBM was pushing their own funky network topology; Sun countered with the work that took the Ethernet standard from 10 megabits to 100 megabits to a gigabit.
Sun also had victories against the likes of SGI with true innovation: their GX card blew the lid off of low end graphics and started pushing SGI more into the niche market. (The GX card team went off to start a company you may have heard of called NVIDIA).
You can dismiss Java if you want, but it changed the industry.
Seymour Cray, AFAIK, never had one second of interrelationship with SGI. He was long gone before SGI bought and pretty much killed Cray -- firesale-ing Cray BSD to Sun in the process, spawning Sun's most successful midrange product line ever. So who did the most justice to Cray's legacy?
There's more -- the industrial design of the SPARCstation 1 and the associated SBus standard was groundbreaking as well. And, although late to the open source game, they moved very quickly to become one of its most active participants: OpenOffice, Solaris, Java, Glassfish, Grid Engine, MySQL...
SGI did some very nice things in graphics, and a few fillips in high-end performance, but their mark on the industry is nothing like Sun's.
If you don't think SGI, and HP, and IBM, weren't at the same time trying to kill Sun, every minute of the day, you truly have no idea what the industry is about.
What happened to those that didn't perform--keelhauling?
I hate to correct such a great man...
...but watermelons grow on the ground. They do not hang.
Re: Remember Token-Ring?
I call killing the superior, more innovative product on that one too!
I used to take great delight in pulling the odd wire at random out of a socket on my network as a demo while my Ethernet-encumbered colleagues shat bricks at the sight of this. I sometimes wonder at the sort of throughput we'd be seeing these days if TR had been developed further, given that even 4meg TR kicked the living shit out of 10meg Ethernet in the old days.
Anyhow, SUN and Ethernet didn't kill off TR, IBM managed that on their own by not opening the the thing up until it was too late and the opposition had got their feet well under the table. They had the lead and they threw it away in an orgy of fat-arsed complacency. The speed, price and compatibility of the 3com TokenLink III cards (once IBM saw the writing on the wall) was a pointer to what might have been, but it was too little, too late.
<Yawn> Try looking forward instead of spending all your time lookig in the rearview mirror. I'm sure there are plenty of people that laughed at your points like ".....first practcial network file system...." - nice qualifier there - or "....the work that took the Ethernet standard from 10 megabits to 100 megabits to a gigabit..." - no kudos to Xerox, then? Just because our World does revolve around the real Sun, doesn't mean everything that Sun claims it did to make the computing world revolve around it is true. But your ending - "....If you don't think SGI, and HP, and IBM, weren't at the same time trying to kill Sun, every minute of the day, you truly have no idea what the industry is about." - is spot on, if a little limited. Sun's blinkered vision concentrated on killing SGI and let such riches as what became nVidia spin off. Sun was late to the x86 and Linux party because, after SGI, Sun spent far too long trying to kill Microsoft and then Linux with the same tactics it used against SGI. Sun didn't adapt well, didn't realise Linux could not be treated like just another vendor OS, and didn't make the correct decisions on vital products. Everything else is forgotten becasue the latter meant Sun didn't make profits. Sure, there will be plenty of Sunshiners for years to come who will get all misty eyed and speak of Sun's "great history of innovations", but they'll be sharing the park benches with those saying the same about Cray, SGI, DEC......
Don't forget about another relatively small smin off - Cisco.
...and yes, NFS was and is an important tech. NFS is critically important in the enterprise, while SMB is more relegated to the desktop (relatively speaking that is). RPC is extremely important, and every OS vendor has copied this technology from Sun.
I don't know what Frank Gerlach would consider a genuine innovation, but I recall the impact Java had on my users when it was introduced. I was an SA for a lab full of quantum thermodynamics modellers at the time. The reaction when I showed them the "3-d molecule" Java demo on Sun's website (the web itself was only three years old at the time) was most gratifying. As far as these guys were concerned, Java was definitely an innovation.
Gee Matt, you'd almost think you had a bias against Sun.
Yes, I did qualify "first practical file system," because other people had tried previously, such as MIT's implementation of a single-writer network file system. The one that worked, the one that got widespread use, won.
"no kudos to Xerox, then?" Well, in this particular case, Sun did the work to develop and promote the 100 megabit extension to Xerox' original work, and carried it to success against Token Ring, which, apologies to TeeCee, had some serious implementation issues, such as jitter and the world's funkiest connector. The point here is that TR was being promoted as *the* networking future by the world's largest computer company, and Sun came up with and successfully promoted a cheaper, more open and more robust solution that carried the day. Yes, part of the problem was that it was a closed architecture; that's part of what made Sun's alternative more compelling.
My only intent was simply to respond to Frank Gerlach's rather short summary of Sun's past that there is a bit more to Sun's history than Java. Slamming this as "looking in the rearview mirror" seems to be more picking a fight than anything, isn't it?
I think that was one of those "joke" things I've heard so much about.
Don't forget about another relatively small smin off - Cisco."
ummm....in what way was Cisco a spin-off from Sun? I don't remember any connection at all. The two founders of Cisco both came out of Stanford - but not Sun itself. One of them worked at DEC for a bit I think.
"Don't forget about another relatively small smin off - Cisco."
more a Xerox Palo Alto Research Center link than Sun
The Stanford/Cisco/Sun link.
He designed the network board that with the router software that was Stanford's router, which became more or less Cisco's first product.
Like these new icons