31 thoughts on “WiFi On Mobiles: Fact or Fiction?”

  1. Present performance shouldn’t preclude future improvement. Boingo released its authentication stack as an open-source tool, meaning any firm could take that, integrate it, and have a pretty varied method of connecting behind the scenes to many, many different networks. (Of course, Boingo would like those hardware makers to also partner with Boingo to enable the account management and fees for those networks.)

    And T-Mobile has a variety of authentication experiments for devices going on. If you’re a Kodak EasyShare-One camera owner–as feeble as the first version of that Wi-Fi-enabled camera is–you just punch in a T-Mobile HotSpot account name and password in the camera, and it authenticates to T-Mobile locations to upload your pictures.

    Authentication will eventually become something that’s seamless and invisible. Standards have been floating around for years in hopes of finding agreement. But with apparently few independent or large-scale operators of Wi-Fi networks finding pure pay-for-play revenue, it’s tricky to get them to invest money that’s not in the core area.

    (Wayport runs 12,000 locations, but 4,000 are managed for AT&T, and 7,000 are McDonald’s, which are a kind of special deal in which McDonald’s and AT&T pay them money. T-Mobile may make little or break even on hotspots, but they gain ARPU and lifetime customers of voice service. And so on and so on.)

  2. Glenn,

    future might be very promising but the present is not all that exciting. The future doesn’t preclude the telcos from wifi, much like Dsl. Of course when worried about the problems they might face in 3g they might find ways to compete.

  3. Fiction: If you believe the carriers will let you run Wifi on your, you’re dreaming.

    Can’t give details but rumours are that carriers want phone manufactures to indicate to them when you switch to Wifi so they can charge you a service fee.

    And of course, you can probably count on not being able to load wallpapers or ringtones via your Wifi connection.

    It’s so stupid, I can’t wait till Wifi is pervasive enough for us to completely bypass carriers. It’s funny – carriers think customers identify with them yet custormers probably identify with the phone manufactures/model more.

    I don’t mind if carriers compete in the Wifi space but they should play fair and not restrict what I should be able to do with my phone.

    Phone manufactures aren’t off the hook either – they simply give in to carrier demands since carriers pay the bills.

  4. PC-Tel has the technology today and has licensed to the major carriers. It’s their roaming client for voice. It’s got all the bells and whistles (no pun intended) with appropriate QoS software, actually same as Skype, Google, Yahoo… (ie GIPS).

    I’m waiting to see when it comes to market, and how.

  5. CDMA phones with wifi/skype on Metro PCS would rule… you could pay on a month-to-month basis and use wifi while in the office. Around town you could use CDMA but while traveling you could revert back to wifi which wouldn’t be tooooooooo painful.

    I’m NOT a fan of the major carriers. Their 2 year commitments, horrible customer service, and price gauging could use some REAL competition and the more pain they feel the happier I’ll become.


  6. I’ve had various PDA phones with WiFi for a couple of years. I’ve used the WiFi capabilty maybe 2 or 3 times.

    Why? An always on, automatic, hassle free connection like EDGE or EVDO is much, much easier to use than turning on WiFi, finding a hotspot, associating to it, clicking through their session capture page, and then, what was I doing again?

    I don’t see WiFi on phones becoming anything big, although it’s a nice feature to add, and doesn’t hurt anything.

  7. WiFi on Mobiles: FACT

    Consumers want advanced services, WiFi is proliferating, handsets are getting better, operators want to increase ARPU … Win-Win for all parties.

    There are so many possibilities with WiFi plus cellular combination! Of course, WiFi on Mobiles will succeed!

  8. WiFi is far too short range. You can forget using it in a moving vehicle, but even walking at a moderate pace, you’ll be switching access points once every 60 seconds. I think this is going to have to wait until the next generation of wireless networking (WiMax, maybe) for it to be remotely useful to the public.

  9. Many people think that for Wifi on mobile phones to be useful, you should have the whole region covered by someone. But actually many if not most of the people make majority of their calls either at home or at the office, and the coverage will be there anyway (or already is today).

  10. Believe it or not, there is a world outside of the UK. despite what you think about Vonage, they offer a wi-fi phone in the UK that is starting to offer the beginning of competition to high priced mobiles. A number of months ago they announced a deal with The Cloud which is a national wi-fi network in the UK (7,000 hotspots) where the customer does NOT need a user name or browser. the phone simply works at the cloud’s hotspot–no input required. the cloud also is expanding its network across the UK in major cities–the first one being the city of london where a customer will be able to use a wifi phone indoors and outdoors across the square mile. that network will be built out within 2-3 months–considerably faster than anything in the US.
    An icnreasing number of wifi networks are being deployed and roaming is a possibilty. i am sure we will see dual mode phones within the next 12 months.
    Deals with networks like the cloud will start to compete with mobile operators–remember most mobile calls are actually made by customers when they are stationary. why pay 20-50pa minute with a mobile when you get equivalent service for one fixed price a month?

  11. Om–

    I’m currently testing a T-Mobile wifi/cellular phone. It has worked well with some minor glitches, motsly dropped calls when switching from the Wifi to cellular networks. I don’t know if this will be their business model in the future, but while testing all my calls made over the wifi network are ‘free’ and do not account towards my monthly minutes. There are still some kinks to work out but I see lots of utility to this product–especially for businesses. You could be wifi at very reduced or free minutes while in the office but then have access to cellular networks as you are out and about without changing phones or phone numbers…

  12. We tried several brands of wi-fi only phones and the battery life on all of them is much like that of a 1995 cell phone. For some reason the wi-fi phone makers have not found 2006 level batteries for their phones. Why is this? When they worked, they worked great – completely connecting to our Asterisk servers.

  13. This article hits on a number of interesting issues, and I would like to comment on a few:

    1) You will NEVER see economically viable VOIP over cellular. And I don’t care if it’s EVDO, or the descendent of EVDO. The key issue is that the voice part of the cellular network is highly efficient in terms of its use of bandwidth. EVDO will never be able to offer a VOIP call for less than the marginal cost of putting that call on the current cellular voice infrastructure. It may be technically possible, but would never make sense economically.
    2) Om is right, the key problem with Wifi, especially when we are referring to non-PC devices, is that the authentication and setup procedures are painful and difficult. Glenn is also right in his comment that people are working on this problem and it will get better over time. This is a software issue, and it’s much easier to solve than a physical constraint.
    3) Speaking of physical constraints, don’t expect for everyone to have the same wonderful EVDO experience that OM is having. Traditional cellular networks have very large cells (around 1 square mile). As more and more users come on line in your same cell, performance will degenerate linearly (you can find a warning of this in the user agreement, along with restrictions on high-speed activities like video streaming, file sharing, and, oh yeah, VOIP). And we are talking about “sharing” only 700K per channel. Enjoy it while you can. There is another reason you will never VOIP over EVDO – it’s extremely asymmetric (once again the bigger cell causing a problem). Upload speeds are not fast enough for 2-way communication.
    4) Lastly, several comments in this posting ignore the innovations happening in city-wide Wifi. These networks are highly symmetric (due to smaller cells), they do have the performance to support VOIP, they do support node-to-node handoff (at least the Tropos nodes do), and most importantly they are not limited to a “hotspot”. Om is right, device connectivity will suck for he near future. We need standards on these fronts. But, once again, the limitations are software issues, not physical ones. Watch Mountain View.

  14. Bill, you’re very wrong on point #1. Carriers will embrace voip in future high-speed packet cellular networks. IMS is only one of the reasons. IMS is part of 3GPP Release 5 and Release 6 specifications. The benefits of using IMS include handling all communication in the packet domain, tighter integration with the Internet, and a lower cost infrastructure that is based on IP
    building blocks and is common between voice and data services. This allows operators to potentially deliver data and voice services at lower cost, and thus provide these services
    at lower prices, further driving demand and usage. IMS will also play a role in the convergence of wireless and wireline networks.

    Furthermore, analyses indicate potential voice capacity gains for voip relative to WCDMA bearer channels, even though it is too early to tell exactly what the gains may be.

    One thing for certain is VoIP will bring other benefits, such as a consolidated IP core network for operators and sophisticated multimedia voice applications for users.

  15. The wireless carriers will eventually move to VOIP, but not with EVDO or even EVDO revA. Packets always win in the long run.

    WiFi setup and authentication will suck until there is a standard and roaming fees have been negotiated between all the major providers – T-Mobile, Wayport, Boingo. Don’t hold your breath, as this is a two-step process. Think SMS interoperability, data roaming on wireless today (… there was one?), IM interop and so on.

    EarthLink will soon be the biggest hotspot operator of them all (and assume EarthLink will probably buy Boingo, as Sky started Boingo and Sky still sits on the board of EarthLink) and could exert some pressure on the other hotspot operators to play nicely. The one useful thing EarthLink/SKT’s Helio can do other than lose hundreds of millions will be to build a WiFi phone that uses EarthLink muni WiFi/Boingo authentication. But then, will Sprint or Verizon really be excited about letting this Alien spawn ride on their networks? I don’t think so…

    The carriers are studying VOIP, trying to figure out how to crush it or co-opt it for as long as possible. They’ll really only play if they’re dragged kicking and screaming by WiFi only phones starting to see real uptake and usage. If the carriers wanted to promote Cellular/VOIP phones there would be great services out there today as the technology exists.

    The first real use of combo WiFi/Cell phones will be at home or in the Enterprise. Need not even have fancy intra-call switching technology, just a simple ‘strongest available signal’ algorithm. Combine this with ‘VOIP Call Forwarding’ available from landline VOIP (where you hand out your VOIP number, simul-ring your cell phone and perform human least cost routing by picking up whichever device you happen to be near) and the loss of cellular MOU starts to hurt…

  16. The main fallacy about dual-mode phones (and WiFi generally come to that) is that they relate to hotspots / city hotzones. They don’t – public WiFi is at best secondary, and at worst a distraction. Forget it, it’s unimportant.

    Dual-mode phones are principally about providing cheaper coverage either at the user’s home, or the user’s office. Everything else is tangential, as these are the two key environments for usage in which it “should” be possible to leverage existing infrastructure (broadband at home, LAN in the office) to deliver voice, using VoIP, more cheaply. This benefits both the service provider (whether that’s the carrier or a 3rd party) and the end-user.

    The main question is whether the WiFi bit is the same “service” as your main outdoor carrier, or whether its controlled by you the user (or your IP-PBX), or a third party service provider. Generally in the business environment it will be non-carrier, and in domestic settings it will be a hybrid fixed/mobile service provider.

  17. Alexander Muse asks…

    We tried several brands of wi-fi only phones and the battery life on all of them is much like that of a 1995 cell phone. For some reason the wi-fi phone makers have not found 2006 level batteries for their phones. Why is this?

    It’s not the battery, it’s the WiFi. Since 1995, cellphone standards developers have spent a lot of time and energy developing techniques in the air interface to maximize cellphone battery life. WiFi isn’t there yet – the design assumption was that the client device would be a laptop with a larger power budget than a handheld device like a phone, so less attention was paid to optimizing the radio technology for client device power consumption. Standards have started to emerge lately to rectify this (yes, the pun was intentional – I’m an electrical engineer by background), but they’re not widespread yet.

  18. My predication: VOIP over wireless/Wifi need at least another 5-8 years of evolution before it can compete against good old circuit switch voice over cellular.

    It will remain a fiction unless –

    Its very surprising to me that while discussing the future adoption of a particular technology, we miss the most important parameter of the equation: The customers!
    Mobile phone manufacturers and service provider need to fix the state of the standards and then deploy those solutions to fix the fix the usability problem.
    The reason I love circuit switch voice is because I don’t have to select/configure the name of the Base station that I need to make my next voice call.

    Quality of Service:
    We all know the current state of QOS in wired VOIP. Just add the mobility/hand-over/and battery issues and you have got yourself, one very unhappy customer. I personally don’t call friends and family with VOIP phones.

    So what if major US metros are fully covered with wifi hotspots, what about the rest of the world?

  19. Yes, you can use Asterisk Gizmo (or any other SIP provider), but that requires that you a) have a home linux server, and b) know how to setup Asterisk.

    I was able to get my E70 to make incoming and outgoing calls via the online PBX provider pbxes.com. I was amazed that it worked the first time…and I didn’t have to configure an Asterisk server at all! You do have some initial configuration to work through, but there are plenty of forum posts that give you all the info you need. It’s not plug-n-play, but it’s a LOT simpler than an Asterisk server. And I’m hoping this wifi functionality will fill the gap in the E70’s lack of 850mhz reception as well as T-mobile’s abyssmal coverage in eastern Long Island, the Colorado Rockies, and a few other areas I frequent. If I’m somewhere where I can hop on to a wifi network, I’ve now got mobile coverage!

  20. The first comment here (Glenn’s) points in the right direction: don’t judge future use or performance exclusively by what’s going on currently.

    The voice networks aren’t currently packet dominated, and thus not the current economically-favored voice call method, but they will be.

    People don’t do much heavy wireless uploading, but they will.

    My current Verizon evdo phone is fast enough for the kinds of uses I have today on that phone, that I NEVER use the wi-fi. Sure, if you’re going to upload your photos from the phone, higher speeds are essential, but the limited software and small-screen interface on the phone is going to make me wait get to the desktop anyway, not to mention the fact that I almost always want to manipulate, downsize, etc the photos anyway, done on the desktop.

    Has anyone here actually tried a Skype call on Pocket PC? You might reach the person you’re trying, but they’ll hang up since they can’t hear you. But it will get there.

    By the time the factors come in to place for a compelling handheld experience– a more solid handheld computer, 3G, 4G WiMax or whatever– we’ll see how dominant or relevant WiFi still is.


  21. Ken, I find Skype on Windows Mobile 5 sounds fine (at least over WiFi). The reason I don’t like to use it is because it kills the battery right quick. A lot of that might be the fault of lack of battery management in using WiFi. Sound quality is not the issue.

  22. Good discussion. Readers may also be interested in this story:

    Nokia tests Unlicensed Mobile Access phone technology
    7/28/2006 8:12:14 AM, by Ryan Paul

    Nokia launched yesterday its very first public test of Unlicensed Mobile Access (UMA), an innovative mobile communications technology that facilitates seamless handover between WiFi and cell networks…


  23. I am not a telecom specialist, so this is the comment of an “ordinary customer”. However I have been in technology for 30 years, so I know something important when I see it. For my personal use of technology I am a long way from early adopter as I hate technology that gets in the way and makes me less productive. So, I was apprehensive when I signed up for the UTStarcom WiFi phone when I signed with Vonage. It sounded “bleeding edge”. But I thought I could replace easily so why not. Today I got it out of the box and after a minor hassle with a fiddly battery connect, I turned it on and bingo it finds a wifi network and make a call. It was that hard. Call quality was good.

    I got Vonage to cut my landline bills. My hope was that it could also help with my much bigger mobile bills. It looks like it will. The big bills are international. How hard is it to wait a few minutes to find a wifi signal before placing that call to India? Vonage does not try to hose you on India calls – something normal like 17c pm. Yes, I could wait till my laptop is up and working and use Skype for free as long as my partner is set up at the same time.

    I am dubious about investing in Vonage, although I may appreciate them as a customer. Anybody know any good public stocks to ride this one?

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.