A recent feature in Network World discussed IT’s big flops over the past 20 years. Being in the industry during that time caused me to pause and reflect on this topic. Token Ring, for instance, was a pretty obvious one. I became an expert early on because nobody understood the thing and I'm embarrased to admit that I even wrote a book on it.
Another writer pointed to the failure of ATM. Well, certainly for LANs anyhow. I remember doing a live skit at an early Interop holding up a length of 2 x 4 and saying “This is your packet.” Then proceeding to pull out a Skill saw and cutting it into several smaller pieces proclaiming “This is your packet on ATM”. Naturally I was wearing a three piece suit and tie (not a good idea hovering over a Skil saw) but I did remember the protective goggles. I also remember giving away a free router at the end of the skit – from Sears of course. Ah, the good ol’ days.
ATM cell technology has actually thrived in the cloud of OC (Optical Carrier) circuits – the beauty is that you don’t have to wait for an entire packet to arrive at a router to start forwarding parts of it – i.e. cells. The bad part is the header and disassembly/reassembly overhead and thus for high speed LANs, it basically failed. But for highly latent networks due to distance and lots of hops like the Internet, it’s a winner. So I wouldn’t classify it as a absolute flop.
Another editor mentioned that OSI was a flop. Well, technically the ISO protocols were, not the layered model per se. But let’s not split hairs.
Yet another writer liked OS/2 but claimed it was out marketed by Microsoft. OS/2 actually started out as a text based OS and was a pig – I had first-hand experience with it and spent many hours down in Boca trying to figure the thing out.
We used to call it Oh-so-slow… divided by 2. IBM and Microsoft shared code but eventually split and Windows emerged, Microsoft slashed the price and cut deals with OEMs at time when OS/2 workstation was still selling in the stratosphere. IBM had a way-to-late fire sale and as the say, the rest is history.
Apple’s Newton, Microsoft Bob, and Windows ME were also mentioned. I’m not an expert on Newton, Microsoft Brains on Board it wasn’t, and ME wasn’t that horrible that I recall, unless you just happen to hate Windows in general.
Finally, security was mentioned, since the topic was hot 20 years ago, still hot today, and the flop is that we are still trying to find the Holy Grail of security. I tend to agree with that one.
I guess I’ll add one more that comes to mind: SynOptics. The company bet the farm on LattisNet – the pre-standard unshielded twisted pair wiring system that was incompatible with the emerging IEEE 10BaseT standard. SynOptics merged with Wellfleet followed by a name change to Bay Networks which was eventually acquired by Nortel and the rest is, well, history.