Contrasting Good WiMAX Analysis Vs. Bad

Wed, Apr 18th 2007 5:45pm — Derek Kerton

Today, I'm linking to two different discussions on WiMAX. The first shows the challenging, time consuming reality of breaking new ground with WiMAX, and the second shows how a research report can be sensationalist (and misguided) by pitting tomorrow's vaporware against today's real-world technology.

First up, a great summary of Sprint's current status with their WiMAX efforts at 2.5 GHz by Kevin Fitchard at Telephony Magazine. Fitchard correctly identifies the Herculean effort required by Sprint to advance WiMAX to the stage at which it can realistically be deployed in a nationwide network. That task is challenging because the reality is that the "standard" that is WiMAX is actually much more fragmented than certain PR departments would have you believe. The frequencies, MIMO characteristics, channel size, adaptive antennas, fixed/mobility are actually not defined by the WiMAX standards, let alone the fact that the standard isn't yet nailed down and interoperability testing is still distant on the horizon. All these factors could limit the global economies of scale that WiMAX has been promising.

But Sprint's Barry West believes that by moving aggressively, and early, Sprint can play a pivotal role in nailing down "de facto" standards before the WiMAX forum finalizes a spec - thus leading the 802.16e standards process. It doesn't hurt that powerful vendors Motorola, Samsung, and Nokia are all deeply involved with Sprint. With all the challenges standards bodies have had of late, a de facto standard seems less political and more attainable, especially when no WiMAX operators of the scale of Sprint exist to dispute Sprint's approach. Sprint's WiMAX gambit isn't a slam dunk, but it IS clear evidence that telcos aren't always stodgy, risk-averse followers propping up old business models. Sprint's 2.5GHz strategy is tied in my mind for "gutsiest US telco project" with Verizon's FIOS project. If Sprint succeeds, they will have a sustainable advantage that their competitors cannot easily copy (for lack of spectrum).

Next up is a report from InStat titled, "End Users Prefer WiMAX" over other options like Wi-Fi, EV-DO, and HSDPA (PDF link here). I'm not sure how InStat came to use this title based on their research, but I find it unfair that reality-constrained technologies like EV-DO are forced to compete with a PR-induced concept of what WiMAX may be when it arrives. That's like asking people what they prefer: The UK or Utopia. Hmmm...I'll take Utopia please. Wouldn't it be more fair to compare WiMAX powerpoints with the powerpoint versions of what EV-DO was supposed to be a year before it launched? The InStat research also found that the two top attributes when selecting wireless broadband were Availability and Reliability. How does that fit with their title, given that globally, WiMAX is available from a handful of base stations in Seoul?

The report details that "respondents interest in cellular data dramatically decreased when pricing was included in the description." So InStat pitted today's cellular data pricing against Sprint's estimate of what they will charge for WiMAX in end-2008? Is that reasonable? It's myopic to assume that the cellular carriers won't have different pricing by then, lowering it to compete with WiMAX, each other, and other technologies.

I suppose that I'm partly over-reacting here. The kind of questions that InStat is asking are, indeed, useful research questions. But they need to be taken in context, and interpreted very carefully. Analysts of this research data need to understand that the study is a "What WOULD users prefer?" question and not a "What DO users prefer?" Analysts also need to understand that reality ravages wireless technologies, so vaporware always looks better than today's choices. Analysts also should understand that markets are dynamic, and cellular won't stand still while we wait for WiMAX. The InStat people should know all these realities well, and should frame their research as such, but "WiMAX To Kill Cellular" gets more headlines than "In a 2008-9 WiMAX scenario as envisioned by Sprint Vs. commercially offered Feb. 2007 cellular data plans, users prefer WiMAX."

Thank you for reading this Techdirt post. With so many things competing for everyone’s attention these days, we really appreciate you giving us your time. We work hard every day to put quality content out there for our community.

Techdirt is one of the few remaining truly independent media outlets. We do not have a giant corporation behind us, and we rely heavily on our community to support us, in an age when advertisers are increasingly uninterested in sponsoring small, independent sites — especially a site like ours that is unwilling to pull punches in its reporting and analysis.

While other websites have resorted to paywalls, registration requirements, and increasingly annoying/intrusive advertising, we have always kept Techdirt open and available to anyone. But in order to continue doing so, we need your support. We offer a variety of ways for our readers to support us, from direct donations to special subscriptions and cool merchandise — and every little bit helps. Thank you.

–The Techdirt Team

Filed Under: wimax

7 Comments

If you liked this post, you may also be interested in...

Reader Comments

Subscribe: RSS

View by: Time | Thread

Propeller Head, 18 Apr 2007 @ 5:45pm

WiMAX and existing technologies
I don't think people care about the technology engine (WIMAX or EV-DO or whatever)...I think we care about speed and cost. And if Sprint's WIMAX is faster and lower cost, than that is what I want. And, I think you are right about the EV-DO people lowering their price, but what I've read, they can't compete with the speeds that Sprint is supporting nor the cost. My vote is on Sprint and their elephant in the room partner, Intel. And we all know how unsuccessful the Centrino platform has been.
[ link to this | view in chronology ]
An Industry Analyst, 18 Apr 2007 @ 5:45pm

Astonishing how much information you pulled from a
It's extraordinary how much you were able to ascertain about In-Stat's research methodologies, report content and conclusions by reading a one-page Executive Summary. Don't you think it's worth actually contacting the company before you critique their work? Funny how such a sloppy "journalist" can be so critical of what he believes to be a sloppy analyst.
[ link to this | view in chronology ]
Derek Kerton (profile), 18 Apr 2007 @ 5:45pm

You Had Me At Hello
In response to #2, I didn't pull my criticisms from reading a 1-page abstract, I pretty much had all I needed from the title, "End Users Prefer WiMAX". The title itself is an untestable hypothesis, and therefore I doubt that the good research InStat did actually supported the title. I'm not so much a journalist as an industry expert, which is why I know that end users have no way of comparing cellular data to WiMAX. Specific answers to your questions: no, it's not extraordinary - it was all in the linked PDF. No, we don't contact people before we critique, nor do we contact people before we compliment. What's wrong with calling it like it is? InStat publishes something into the public domain, and it's fodder for discussions and blogs: welcome to 2001. Also, my "journalism" was free, the report cost $2,995. Should it be held to a higher standard? As a last note: I'd bet that the research in the report is actually quite useful for anyone in the industry looking to measure consumer preferences for Broadband Wireless Services. But the criticism is that the Exec Summary and press release have been titled in a sensational way in order to increase circulation. But in this case, the sensational title actually degrades the value of the work.
[ link to this | view in chronology ]
Derek Kerton (profile), 18 Apr 2007 @ 5:45pm

Note Other Wild Claims
For commenter #1, sure, that's when you accept the PR ruse of comparing WiMAX powerpoints against 2007s version of EV-DO (which is actually a technology that was launched in Dec. 2001 in Korea). Comparing a future technology against a 2001 technology is unfair. Have a look at this current Qualcomm claim about the soon-to-be-released version of EV-DO. http://www.netstumbler.com/2007/04/02/evdo-rev-b-promises-more-than-9mbps-down/ I, by no means, am endorsing that claim, or saying that it is true. Quite the contrary. I am saying that the Sprint/WiMAX claims as well as the Qualcomm claims about future performance should be taken with a grain of salt. Future claims should be compared against competing future claims, and real world performance vs. real world performance. That's fair.
[ link to this | view in chronology ]
Alaric, 18 Apr 2007 @ 5:45pm

Re: Note Other Wild Claims
I agree wth Derek here and thank him for his honest assessment. I looked at the Instat PDF. There are some useful questions in there but its simply poor research methodology to ask the wimax vs everthing else that actually works question. Its a misleading and loaded question because WiMAX does not exist and it could very well perform far worse than expected or marketed. Derek is again correct on the current vs future performance issue specifically and in general. WiMAX is making its way by telling the world what it will be someday when it actually works and by comparing itself to older existing and working technologies. Lets settle this right here and now. There is no evidence whatsoever that mobile WiMAX will even be better than evdo today much less EV-DO in 1, 2, or 5 years. Furthermore, WiMAX proponents consistently compare future WiMAX with MIMO vs. existing EVDO/WCDMA without. That is simply meaningless. And its why only one major operator in the entire world, just one, has singed up for mobile wimax. And that one operator faced unique requirements (TDD, FCC coverage deadline) and that one carrier is probably going to be very very late with WiMAX because WiMAX is not ready, which means Sprint may still bail on WiMAX. Sprint also choose the one tech that had the slimest chance of being ready for deployment. Hmm interesting. They could not go to the FCC and say UMTD TDD or FLASH-OFDM are not ready but they can and they will do that with WiMAX when they ask for an extension. In the meantime they get invaluable intel sponsored marketing hype to help distract the market from their poor performance and issues integrating Nextel and maintaining nextel customers. Sprint can still dump WiMAX.
[ link to this | view in chronology ]
WirelessMan, 18 Apr 2007 @ 5:45pm

Wimax good on Powerpoint so far
Alaric & Derek, I agree with all what you guys wrote. We all can say we will have double the speed in 18 month time "Moors law", all the Wimax numbers stack nice in the future, if that future will really exist. And all the others can state the dsame figures for the future. Sprint had a different agenda, 1) please the FCC, 2) please the stock market, 3) delay it as much as possible, 4) If it fails in two years time they can say Moto & the Korean let us down. I recall that they have dumped other wireless technologies in the past when part of them was Worldcom they dumped Vayyo DOCSIS+ & Sprint dumped Andrew/Hybrid and these two half rollouts were on these same radio channels. even AT&T dumped the Ange project.l If any radio engineer take look at the radio link budget, he will tell you it will not fly, you can’t make an ecosystem work, if your filters are > 30 MHz wide, you will need different band units per market, this is were the compatibility and the economics go away. I think they have channels all around the 190 MHz between 2.50 to 2.69 GHz; you can’t have a 190 MHz filter and collect all the band noise, when there are other operators using it for something else including a TV broadcast. Designing a multi-band CPE is also hard when your bands are all around the place (distributed and georaphical ). These built in antennas in the laptop will have negative gain, when they have to be compatible with EVDO in 900/1900 MHz as well, and if the export version has to work in 2.3 GHz, 800 & 1800 MHz. And we know what Intel want's out of this, get it up so they can then shift all the processing power to their CPU chip, forget about those guys building dedicated ASIC for CPEs, these will be short lived, better do it in the Quad core CPU once they get the timing & syncs right. Also no were else in the world you have this much spectrum, other operators around the world need to think about n=1, not n=x like Sprint, I haven’t come across any inter-frequency handover or reselection scheme in the 802.16e and if a CPE doesn’t quickly reselect when it is losing the signal, it will cause trouble. The most annoying thing is actually these analyst, they tend to mix things like fixed and mobile Wimax numbers, present technology with future technology. They even know what customers want in the future and have numbers on people who will be employed, forecasts based on something that does not exist at all except in PowerPoint. I recall reading some KT / WiBro presentation predicting 270K jobs created in Korea in 2006 from WiBro, I wonder if they have plans to employ them for Sprint now. I think I might as well go and buy/sell Wimax spectrum in Secondlife now.
[ link to this | view in chronology ]
matt, 18 Apr 2007 @ 5:45pm

Speculation Drives Tech Industry
I agree that the comparison is unfair given that the two products are at different life cycle points, however it needs to be noted that companies such as Cisco have been selling the future for years. Analyst's need to help end users recognize how far away the future is.
[ link to this | view in chronology ]