Skip to comments.Ken Olsen, DEC & the Horizontal Revolution (What CEO's can learn from the mistake of DEC)
Posted on 02/09/2011 7:25:47 AM PST by SeekAndFind
Ken Olsen, founder and CEO of Digital Equipment Corporation, died on Super Bowl Sunday. Coders of a certain age have been reminiscing about their first times with a PDP-8 and getting a PDP-11 to really sing. Me too. I wasted millions of ratepayers dollars at AT&T Bell Labs on VAX 11/780s. Woo-hoo. DEC was a great company.
But Ken Olsen made a classic mistake. He idolized IBM. Wanted to be them. DEC was the same vertically oriented company as IBM. They designed chips, wrapped plastic around them, wrote operating systems, wrote applications, and then marketed, sold and serviced their minicomputers. Soup to nuts. Chip to dip. (Any reference to salesmen as dips is purely coincidental.) But the world was soon going horizontal and would take DEC out at the knees.
In 1987, Ken Olsen famously hired the giant 963-foot, 70,000-gross ton Queen Elizabeth II ocean liner to dock in Boston Harbor to hold his DECWorld expo. DEC stock hit its all time high of $199.50, and Ken Olsen proudly announced DEC was hiring an additional 10,000 salesmen feet on the street, I believe was Olsens quote to gain market share and not only dominate the weak sisters in minicomputers, like Data General, Wang and HP, but to start to take on the big prize, IBM.
But think about it. This was 1987. The PC was introduced in 1981. Lotus and Compaq went public in 1983. Microsoft went public in 1986. DEC even offered, sorta, kinda PCs under the name Rainbow (it had a proprietary floppy disk format, yikes!).
But the PC didnt kill DEC, not directly. It was DECs business model that was their undoing. Vertical was dead. Olsen just didnt get the memo. Neither did IBM, really.
Instead, the computer industry, without pre-thought or oversight, organized horizontally. Intel microprocessors, Western Digital hard drives, Read-Rite disk drive heads, Microsoft operating systems, Lotus and Aldus and Adobe applications, and of course IBM and Compaq and Acer and Packard Bell and Dell and Sony and Toshiba computers (who could tell the difference?). The sales channel was not only retail like CompUSA, but distributors like Ingram and 800 numbers like Dell and mail-order catalogs. PCs had more feet on the street than Ken Olsen could ever hire.
By 1990 the stock was $57 and eventually, DEC was sold to Compaq (ouch!) in 1998 for a paltry $9 billion. Thats a slow sinking of a once-great ocean liner.
The lesson? Get horizontal. AT&T once sold phone service from soup to nuts, offered local and long distance, sold phones, repaired wiring. The Internet evolved horizontally. Cisco routers and Netgear switches, Comcast or Verizon broadband, Rackspace servers and distributed data centers, Google search and Apple iPhone Edge devices. There is always room for someone to enter and define a new horizontal layer and own a valuable service (think Facebook) without having to duplicate the entire infrastructure, i.e., you dont have to rebuild the QE2.
Horizontal makes things efficient. Each layer can go at its own pace of innovation (for example, several iterations of Pentiums during one Microsoft operating system cycle.) After Scale and Waste Abundance, Horizontal is my Rule #3 (out of 12 and a bonus Rule) in my new book Eat People and Other Unapologetic Rules for Game-Changing Entrepreneurs. Now you know how I learned this Rule!
Andy Kessler is a veteran Silicon Valley investor who switched trades and started writing books instead of running money. He is also a keen observer of human foibles.
That quote from Olsen was about mainframe computers. Do you have a room-sized computer in your home? No? Well, Olsen was correct, you see.
Ah, the good old days. My first job was being a gofer, printer monkey, tape-hanger, etc. at a local college’s computer center. They had a DECsystem 20. When I went away to college in 1984, our computer center was all VAX—at first an 11/785, 11/780, and 11/750, then they put in an 8650 and 8600 (I think) and ditched the 11/750. We had to do all our IBM work in batch, sent down a phone line to Virginia Tech and back again once an hour (as a student operator, I got to put the phone in the acoustic coupler and flip the switch!).
I always did think VMS was an excellent operating system from a user standpoint. I don’t know about the technical side of it, but I always liked using it.
Why would you need >512K RAM?
That statement was true at the then current prices. Many years ago DEC was the next big thing (along with Wang). FWIW, Mr. Watson’s company still builds mainframe computers and servers of all sizes. What does DEC build these days? And Wang? And Burroughs? (I date myself...)
Yes, but I vaguely recall a noted IBM scientist had said that we'll all be wearing 360s on our wrists, too.
RE: I always did think VMS was an excellent operating system from a user standpoint. I dont know about the technical side of it, but I always liked using it.
I developed on OpenVMS as recently as 2003-2008 ( before capitulating to evil Microsoft and becoming a .NET developer on Windows :)
It is one of the most robust, reliable, virtually unhackable operating systems I have had the pleasure to work with ( much better and easier to learn and use than UNIX too ).
Unfortunately, it has had its day ( and even I myself had to move out of it in order to put bread on the table ). There are still companies out there that use it for highly mission critical systems, but it is clearly a dying product.
During its heyday, it had the fastest microprocessor in town — the ALPHA Chips.
DEC sold itself to Compaq, which eventually was acquired by HP. LAst I heard, HP ported OpenVMS on INTEL microporcessors. I’ve never heard anything exciting about OpenVMS ever since.
I have to say thank you to Ken Olsen for providing me my bread and butter for years, right after I graduated from Engineering School. The VAX was one of the earliest computers I’ve used.
Just goes to show that the best engineered products do not necessarily become the most successful.
Say what you will about KO but he was worlds ahead of any of the new crop of bimbos we like to call CEOs. Like all company officials he made mistakes but he was right more than wrong. We wouldn’t be where we are today w/o the KOs of the world.
That guy was a piker. The computing power of a typical smartphone is several orders of magnitude greater than that 360. The palm pilot used a Motorola processer derived from the 68000 that powered the early Macintosh. I had no trouble learning assembly language for that platform, just as I had no trouble with the Mac, because the register layout and orthogonal instruction set were analogous to the 360, which I had learned as a teenager. The 360-on-wrist point was passed at least 20 years ago.
I thought I was completely bada** when I “sold” a small program to Quadram, an early PC expansion board manufacturer, for 192k of RAM to fully populate my Quadboard (4-function expansion board with RAM, parallel and serial ports, and clock). It took my system from 128K to 320K of RAM, which was a *lot* at the time (late-’82 or early-’83).
The thing that stopped me from buying DEC and going to SUN, SGI, Pyramid, and etc was the DEC sales force in the mid 1980s. Dealing with them was like pulling teeth to get technical questions answered. Their SEs were usually top notch, but my sales guy was a pain in the butt. If you were not a big customer, you got third rate service. I caught our sales guy, who had the appropriate first name of ‘Dick’, in a blatant, bald face lie. He response was, “Where else are you going to for a VAX?” Sadly my boss loved him as he got a steak dinner one or twice out of him.
Olson didn’t kill DEC, palmer did. everytime they made money, he’d whack them with more restructuring costs. he fired his best SW engineer, who went to microsoft and wrote WNT as a followup from VMS. when he decided to design WNT on the intel platform rather than the faster, cooler Alpha that DEC had, that was the death knell.
I can’t speak for the marketing side of things. But I think there were some technical issues that DEC failed to address quickly.
As microprocessors continued to improve in the 1980s, it soon became clear that the next generation would offer performance and features equal to the best of DECs low-end minicomputer lineup.
For instance, Berkeley’s RISC and Stanford MIPS designs were aiming to introduce 32-bit designs that would outperform the fastest members of the VAX family, DEC’s cash cow.
Constrained by the huge success of their VAX/VMS products, which followed the proprietary model, the company was very late to respond to these threats. In the early 1990s, DEC found its sales faltering and its first layoffs followed. The company that created the minicomputer, a dominant networking technology, and arguably the first computers for personal use, had abandoned the “low end” market, whose dominance with the PDP-8 had built the company in a previous generation. Decisions about what to do about this threat led to infighting within the company that seriously delayed their responses.
Eventually, in 1992, DEC launched the DECchip 21064 processor, the first implementation of their Alpha instruction set architecture, initially named Alpha AXP (the “AXP” was a “non-acronym” and was later dropped).
This was a 64-bit RISC architecture (as opposed to the 32-bit CISC architecture used in the VAX) and one of the first “pure” (not an extension of an earlier 32-bit architecture) 64-bit microprocessor architectures and implementations.
The Alpha offered class-leading performance at its launch, and subsequent variants continued to do so into the 2000s. An AlphaServer SC45 supercomputer was still ranked #6 in the world in November 2004.
Alpha-based computers (the DEC AXP series, later the AlphaStation and AlphaServer series) superseded both the VAX and MIPS architecture in DEC’s product lines, and could run OpenVMS, DEC OSF/1 AXP (later, Digital Unix or Tru64 UNIX) and Microsoft’s then-new operating system, Windows NT.
In 1998, following the takeover by Compaq Computers, a decision was made that Microsoft would no longer support and develop Windows NT for the Alpha series computers, a decision that was seen as the beginning of the end for the Alpha series computers.
DEC tried to compete in the Unix market by adding POSIX-compatibility features to the VAX/VMS operating system (becoming “OpenVMS”) and by selling its own version of Unix (Ultrix on PDP-11, VAX and MIPS architectures; OSF/1 on Alpha), and began to advertise more aggressively.
However, IMHO, DEC was simply not prepared to sell into a crowded Unix market however, and the low end PC-servers running NT (based on Intel processors) took market share from Alpha-based computers. DEC’s workstation and server line never gained much popularity beyond former DEC customers.
After Compaq merged with Hewlett Packard, they announced the port of OpenVMS to the Intel Itanium architecture.
This port was accomplished using source code maintained in common within the OpenVMS Alpha source code library, with conditional and additional modules where changes specific to Itanium were required. The OpenVMS Alpha pool was chosen as the basis of the port as it was significantly more portable than the original OpenVMS VAX source code, and because the Alpha source code pool was already fully 64-bit capable.
THAT WAS THE LAST I LEARNED OF DEC’s latest and greatest product — OpenVMS. I don’t think any significant organization is buying into this initiative any longer.
I remember telling them that some day, software portability would allow us to take our software and run it on non-DEC platforms and they would be dead. The DEC sales reps just laughed in my face. At the time, they had the Intelligence Software market captured on DEC platforms.
nice synopsis. the nice thing about DEC was that whatever the problem was, you knew who to call. if it was SW or HW, you made one call and eventually you got someone who knew what was what.
they had the best help desk out there. and their on-line help was good too.
RE: DEC sales rep, a young woman, walked out on us because someone was selling eel skin purses in our parking lot!
Who knew that a DEC sales rep would be a card carrying member of PETA??
My town is the home of DEC, and the big mill is still there, I have been inside it many times. The clock tower is interesting in itself, and is the longest continually operating clock tower in New England with the original mechanism. They gave a tour of it last year, and the tin reflectors behind the clock faces have signatures all over them from famous people thoughout the years, Henry Ford, things like that.
When I was in the USN back in the mid-seventies, the Navy, Detroit Diesel Allison and Rolls Royce were doing research (ICEMS, Inflight Engine Condition Monitoring System) to see if they could install sensors on our A-7 Corsairs TF-41 engines that would measure turbine temperatures, rpm, vibration levels and so on, in an effort to determine by analysis if an engine was going to fail prematurely. I was chosen to take part as a jet mechanic, and it was my first introduction to a real mainframe computer, a PDP-11. Lots of blinken lights and address switches, but most surreal to me was the paper tape. I found it hilarious and fascinating at the same time. It seemed so primitive, yet simultaneously, so advanced.
And so it went...:)
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.