Tuesday, June 11, 2002

802.11 - Planet Philadelphia conference 6/11/2002 - Session #4- Track 2: Education: Unwiring Innovations. Paul Goransson, Meetinghouse Data Communications Factoid - WLAN education market penetration is 25%. Is that true? That seems very high. Excellent slide on "Challenges" Proprietary security extensions. Vendor Solutions Lock -   these are very real issues. High implementation costs. High replacement costs. Feature set compatability. Speed to Market- so many markets - so little time, money, and where are the ROI's? Addressing multiple market segments. Meeting enterprise vertical market requirements. Delivering high quality solution. Georgia Designated Public Liberal Arts University -  there's a name for you... USG Proof on Concept: Avoid costly mistakes. Beyond the classroom and residence halls. Cautious engineering of the network - no rogue insecure AP's. Looks like a fairly robust network: Enterays RoamAbout R2 AP's. RADIUS servers. 65 AP's Omin-directional antennas with 50ft cable drop to AP's. 802.1X protocol required for all WLAN users. Covers the main campus, residence halls, 1-2 block perimeter area. Challenges: Immaturity of testing equipment. Traffic simulation - nothing available for real load testing. This is critically important - networks that don't scale are real disasters. Historical Buildings Camouflaged antennas. Roof top antennas - cables painted white to blend into architectural backgrounds. Penetration issues in older buildings with metal sheeting. Enforcing network policies. Big Issues over which security solution to implement - 802.1X, VPN, etc. From the user perspective all security layers should work with the same identification. Kevin Mazzatta, Wireless e-Business Services, IBM Univeral Access: e-Rate - ah ha. e-rate is very important in our (GCI) Bush School Access programs. Another very important funding source. IBM's New York School District project: 200 schools. 2 weeks/schoool to survey and install. 20 teams - 3 people per team. 6 month plan - erate has strigent time frame requirements from funding to implementation. Security - they used WEP 128. Build in time for politics - that is so true and often not considered. Technology is the least of the issues. Standardization for 1 million students - multiple operating systems, PCMIA cards, etc. Critical Success Factors: Administrative Commitment Constituency Support Experienced Project Management Sound Financial Planning Hardware/Software Standardization Defined and Measurable Results 12:15:03 PM    comment []

802.11 - Planet Philadelphia conference 6/11/2002 - Session #3 - Track 3: The Next Generation OSS for Wireless Data. Jim Geier, Principal, Wireless-Nets The big roaming issues: Open Access Roaming One access point to another with seemless reauthorization between access points. Note - we do not have seemless roaming here. I have to reauthenticate each time I move access points. From one IP domain to another. From one WISP to another with only one bill. "AAA" Authentication - proving who you are and the network proves who they are. Authorization - what is the user allowed to do on the network. Accounting - mechanism to account for usage and activity. Mediation and settlements. Single Bill - this is huge. WECA - working on a WISP roaming standard. PASS 1 - new organization. Also working on roaming standards. 3G mobile phone competition - outdoor connectivity. Complimentary. Tim Allwine, VeriSign Roaming = Interoperability + Transparency Hype in the press regarding 3G/WLAN interop Sees PDAs as a real emerging market. Vision - mobility and portability. Flexible billing options. OSS Requirements to support roaming: Subscriber provisioning Smart client download/updates RADIUS has been accepted by industry. Network monitoring Billing and Settlement Most plans today are bucket plans. In the future will be usage based - roaming issues. Interoperator settlements. Security Conundrum MAC address control lists are not maintainable. Authorization - all or nothing problem. VPN Requires client software Expensive Proprietary solutions compromise openess of systems. Dick Segress, President TTS-LiNX WISP Business Case Issues - Good Basic List - I'm surprised the list hasn't been presented more often at this conference. It's interesting - people started leaving during this presentation. And this is the one presentation I've attended that has REALLY looked at the business models. Know your demographics Geographic area (high tech/urban/rural) Income level of target customers. Pop density Hotspot/Enterprise/Corporate/Residential Know your offering Pricing Services Data Only? Voice/Video? Coverage Area Intended Deployment Strategy Build the system and hope the customers will come. Deploy system based upon a customer base. Vendor selection. Pre-Engineering. Own and lease or sell CPE - this is a biggie - cost of amoritization. Pre-market studies. Is the potential customer in a building interested in WLAN. Do the customers have an existing broadband agreement. Backbone - wireless/physical facilities. Financing. If you're on a limited budget - you have to build the market as you get customers. Do not do a "build it and they will come" model. Know your regulatory issues. Put it together Infrastructure costs. Market expectations. Service offerings. Marketing Costs. Marketing issues are very different based on the types of markets. He mentioned lack of hot spot signage - boy is this the truth. Admin costs. Provide/Outsource operating systems End user support. Hotspot support. AP/Network Monitoring. Implementation - be sure you follow your business plan. Site Survey Pre-Marketing/Marketing Hotspots/Residential/Business QoS Network Management. Bottom Line - does your business model provide an adequate ROI. There are going to be a lot of "covered wagons abandoned in the desert" because of bad business models. Adqueate financing is critical if you're doing hotspots. Right now the cash flow is not going to be as good as you think it is. Nationwide Access/Roaming Value added to Customer. Potential Revenue Source Larger Footprint. Q & A Revenue assurance and security are huge when implementing roaming agreements. Pushback to landlords on what they want - huge issue. Work with landlords who don't want to squeeze blood out of a turnip. Expensive landlords can kill the business model. Some buildings won't let you put the AP's in a visible location. 10:46:29 AM    comment []

802.11 - Planet Philadelphia conference 6/11/2002 - Session #2 - Track 1: New Directions in Mobile Technology. Moderator - Shelly Tyler Radler, Senior Analyst, Info Tech Joe Dodson - TeleSym Corp (Software based IP telephony) Oh good grief - this appears to be a demo of VoIP over DSL to 802.11 sigh....I was really hoping more for 2.5G/3G convergence with 802.11. More about seemless handoff between the two. Ah - Bethoven's 5th over ADSL to 802.11 VoIP - the height of listening. Actually - it didn't sound horrible. Brent Lang, Vocera Communications Kind of cool - I can at last see a use for Vo802.11 - they have a fully duplex voice device you wear on a neck cord that is voice activated and allows you to communicate internally within a company without having to dial - sort of a Star Trek communicator model. It also can be interfaced with a PBX so you can access the PSTN. This could have real uses - especially within an environment such as a hospital. Q & A What types of companies are interested in VoWLAN? Health Care and Retail markets - the pain point in those markets is extreme - better ROI. Diverse highly mobile workers that need to use their hands. What are some of the deployment issues? Duplex voice driver for an IPAQ. Getting connected to the Net - tons of IP address issues and SSID issues. Been there done that in spades. QoS? 802.11E spec will adress that. However, it only focuses on the wireless connection - AP to radio card. So end to end Qos is tricky. Good point. Thought: I still don't see widespread applications for VoWLAN. Cellular phones already do mobile telephony and do it at a reasonable (not great) QoS. VoWLAN has so many issues to address and the market is limited. I just don't see a viable business model. Although I do like the Vocera business model. As a matter of fact, the Vocera guy just said they're not targeting end consumers. I think VoWLAN is a specialized niche. 9:40:02 AM    comment []

802.11 - Planet Philadelphia conference 6/11/2002 - Keynote #2 - Dean Douglas, General Manager, Wireless e-business Services, Americas, IBM Global Services I've walked in late on this one, so it will take me a bit to get up to speed on where where we're at. Cute - he's showing on of my favorite ads - the geek sitting in a Venice plaza with the birds screaming buy and sell into thin air - handsfree. I think it's his glasses that have the wireless capabilities. Machine-to-Machine Communications - factory floors. Remote monitoring and control. On-Premise Management by Location - passive information flow. Wireless Payments - I'm a fan of this one - pay for vending maching transactions with your phone or PDA. The Gaylord hotel in Florida has wired all the vending machines on the facility. You use your room key to access the vending machine. This isn't really wireless but it's good practice for the real thing. Micropayments are the big issue. An East Coast university uses a swipe card method using the student id. They're testing the system in a "hostile environment." Kids and vending machines...Good test. The above examples are all swipe card based. From the IBM perspective, it doesn't appear as if there are wireless vending examples in the US. I bet there are but IBM isn't invoved. As mentioned above, micropayments are the big issue. A lot of talk and a lot of investment being made in public hot spots and enterprises. Here comes security again - IBM believes you can impement cost effective secure networks. Well duh! There are many different ways to create a secure network - VPN, Universal Gateways, etc.   7:57:04 AM    comment []

802.11 - Planet Philadelphia conference Alan Reiter's comments regarding another session: Hardware strategies for the enterprises I am listening to Lee Kedrie, convergence architect and evangelist for Hewlett-Packard, advertise the wonderfulness of HP.  This is supposed to be a panel about 802.11 for enterprises.  If it's too generic, I'm leaving.  Too bad I always like sit up front! "We're starting to see an ecosystem that's taking place."  Ecosystem has become one of the popular buzzwords.  "Traction" is another one.  Okay, it's too generic and too much of an advertorial for HP.  I'm outta here.... 7:37:46 AM    comment []

802.11 - Planet Philadelphia conference 6/11/2002 - Session #1 - Track 2: Growth and Profit Perspectives for 802.11 Network. Moderator: Goli Ameri, President eTinum, Inc. Panelists: Chris Kozup, Senior Research Analyst, META Group Inc; Mark Lowenstein, Managing Director, Mobile Ecosystem; Adam Zawel, Director Wireless/Mobile Enterprise and Commerce. The Yankee Group. Chris Kozup's presentation: Pervasive devices will outnumber traditional PCs by 2003. Question - how does Chris define a pervasive device? 100,000 worldwide hotspots by 2006. Note - See first presentation  - Cahners predicts 12,000. Hmmm big difference. Maybe Cahners is US only or maybe Cahner's stated 120,000 and I misread the slide - entirely possible.  65% of enterprises will deploy applications to mobile devices by 2005. 90% of enterprise applications can be synched. Most of the interactive apps are vertical. Yet - IT departments are experiencing grassroots demand for WLAN. A secure wireless LAN costs an additional 10%-30% above the initial implementation costs. Analyzing the price of postponement: Rogue AP assessments - time it takes to hunt down rogue AP's within the corporation. Potential security breaches. Lack of management. Limited features. Forcing standardization and compliance. The Global 2000 are seeing strong ROI for specific applications: Asset management. Point of sale/care Employee mobility. Inventory management. Chris thinks that WLAN should be deployed as a parallel technology with CAT 5/6 within new building construction. It's all about cost. Chris predicts that convergence between 802.11 and cellular technologies by end of year 2002. VoIP of 802.11 by 2005 - think it will take time for that market to mature. Absolutely True. Adam Zawel Wireless Opportunity Near Term High Revenue - Financial Services, Gov, Near Term Medium Revenue - Telecom Long Term Lower Revenue - Entertainment. Neat slide on horizontal opportunity by market segment. Another good slide on Value of Device/Network/Application - Lots of $$ opportunities with WLAN PDA apps - I would agree. Note to self: I need copies of these slides. 40% of enterprises think they will deploy wireless between 12-24 months (but they said that last year also). Hmmm - I'm not sure who this speaker is. He's from Alliance but is not listed. Market Opportunity - education. True. Offers low cost way to distribute computing power. Less concerns about security. Increased use of laptop as desktop. Team-based learning. Can't run wires to these desktops! Remember - educated students eventually become employees. Education focused on using the technology beyond surfing. Dynamic interaction beyond the yada-yada lecture. This is along the same lines as dynamic conference blogging. Backoffice - warehouses, etc. Font Office perspective - organizations ignore the ROI of an always connected employee. Home/SOHO - DSL/Cable + Linksys create CXO backlash. Where Enterprises will focus: consolidation of wireless voice services. Consider other enterprise services that can be served by 2.5/3G data. A company called RadioFrame. Comments: The Alliance guy - The biggest challenge facing scaling up is billing and infrastructure. Adam - Wireless carriers will be the ones to bundle services. Chris - internal deployements will of course be ruled by IT departments. Public hotspots - roaming and roaming/billing agreements are key to widespread adoption. Agreed. 6:27:40 AM    comment []

802.11 Planet Philadelphia Conference - Conference Observations: Alan Reiter is doing a much better job than I am of blogging the conference. The "geek quotient" at this conference is quite low. Only a few guys with pony tails, jeans, and nifty antennas. Way different than the EyeforWireless conference I attended in December of last year. Very few attendees have laptaps out - maybe one in 30 or 40. Wow - very surprising for a wireless conference.  Update - I take back that comment - probably 20% of us have laptops and we're all congregated around the few electrical outlets that are available. Makes for an interesting dispersion pattern in the conference rooms. There are a few more women here than usual. One of my favorite conference activities is to determine the percentage of men to women. The higher the conference's geek quotionent, the fewer women. It's my guess that this conference is 90% men and 10% women. 6:02:47 AM    comment []

802.11 - Planet Philadelphia conference 6/11/2002 - Keynote #1 - Dennis Eaton of WECA WECA membership is 51% North American and 40% Asia/Pacific. There's that Asia hotbed thing again. I love this - a WECA goal is the consumer can "open the box and it just works." Wouldn't that be nice? Significant price drop within the past 8 months and more product choices than ever. As of Friday of last week 367 products have been WiFi certified - 40 products alone have been certified since the end of May. Hubba Hubba. Product evolution - initially access points and PCMIA cards, now mini-PCMIA cards for laptop embedding, home gateways, dual function cards as minutarization (gads I mangled the spelling on that one). USB cards have become very popular. Print servers and compact flash cards are beginning to appear. WiFi is becomming promiscuous. Great description. Telco industry is becomming interested in Wi-Fi - no news there. Compliment to 3G services. What are the opportunities that are left? WECA never imagined that the growth rate would be this rapid.  How to build a business model - how to compete with the guys who are free. Quality of Service is one way. The big unknown is the intention of the big telcos - how are they going to tie in 802.11 with their current services and 3G. Business model failures - Mobile Star and Ricochet (which was not based on WiFi). In my opinion, Mobile Star failed because it had a dotcom business model - capital is free, basically give the service away, build out as far as possible. Rate of return is secondary. Poor Ricochet failed because of high capital costs and proprietary technology. Cahner's hot spot growth predictions: big explosion in US and Asia in 2003 and 2004. A good slide on hot spot predictions - 12,000 in 2005 - 40% in coffee shops - I couldn't see the rest of the slide. Airports had a big precentage also. A slide on the potential conflicts between SDARS (satellite based service providers such as XFM) and 802.11b. I think Glenn has this one right - WiFi (802.11b) should beat feet out of 2.4 GHz into 5 GHz (WiFi5 - 802.11a). Interesting question about "software radio's" Software taking over the MAC layer - Dennis sees this as 2-3 years out. I personally wonder about where the market is for software radios - radios are already so cheap - why move it to software? Other than flexability between standards, etc...This one I don't get. Someone just asked if WECA views Ultrawideband as a threat - Dennis answered no. Thinks it's a threat to Bluetooth because it's a PAN - not a distance WLAN. Will the FCC step in to regulate WiFi? Dennis thinks it's already so widespread that the FCC will not step in. I agree - you can't stuff the Genie back into the bottle. Dennis mentioned "smart antennas" that really boost the distance of WiFi. A definite must. Speaking of antennas, I've seen some very spiffy portable antennas that some of the geek attendees are using. 5:09:31 AM    comment []

© Copyright 2005 Stephanie A. Kesler. Last update: 3/5/2005; 8:52:21 PM.