jgh, another view from a knowledgeable source on Ragingbull NT thread:
By: Petabit $$$$
Reply To: 23962 by Bex $$$
Friday, 8 Dec 2000 at 10:19 AM EST
Post # of 23985
Bex: lie, damn lies and statistics.
Which also happens to be the title of a great book.
You can spin market share any way that you want. By dividing the categories, you can be
the number one vendor in something. You can get an understanding of this from
comparing the product families:
Nortel's primary product (MOR) was launched in 1997 and carries 2.5G and 10G traffic
over 32 wavelengths. This is what is generating most of the revenue. They introduced the
new platform (OPTera LH 1600) in mid 2000, which can carry 40 channels of 10G today
(scaling to 160 next year).
So I'm very intrigued how NT could get 4% market share of 40 channel systems in 1999,
since they didn't have a product then. The number will be very different for 2000.
My point is, if you set the 'high-channel count' breakpoint to 40, Ciena will have enormous
market share. Set it to 32 and Nortel will win.
Nortel will ship a 160 channel next year, Fujitsu have promised a 196 channel system. So
I guess Fujitsu will have 100% market share in the above 180 channel systems!
The other point made in the article concerns 40G systems. I disagree with the points
made in the article, because they are still only looking at the line system. In the future you
will have to look at the networks as a whole - it will be too complex to try and manage just
the line systems. Which is what the whole Sycamore ODSI v. MPLS v. ASON is all about.
I have a lot of respect for Lightreading - under the gossip and spin is a lot of good
journalism. Peter Haywood is a smart guy. Everyone has their own agenda, and the
article reflects that.
P.
ragingbull.altavista.com |
