Wifi is still not a good WAN (municipal) technology. It has never succeeded. Many NYC public parks offer free wifi and I don’t know anyone that uses it.
4G/LTE/etc is engineered for uncontrolled, outdoor spaces. It works more reliably than Wifi in that environment.
Wifi’s UX is poor in that environment – you’re already connected to 4G but connecting to Wifi requires effort. And one assumes that if one is on the street, one will be walking out of range of that Wifi within a few minutes. I don’t understand the use case.
(The sitting-in-Starbucks use case makes more sense.)
Politicians like the sound of Wifi. Users walking down the street, by and large, ignore it.
Minor win: it's fantastic for tourists. Roaming kills, and it's great to have access to free wifi (maps, booking sites, e-mail, messaging, whatsapp calls, &c) in a foreign country.
Same for Wellington's waterfront wifi! It was very helpful for quick Skype calls and looking up information, especially when internet anywhere else in the city was very expensive.
That requires either sending login information over http (thankfully rare in 2016) or breaking https, which is much harder nowadays with modern browsers-- most of the tricks sslstrip[1] used no longer work due to key pinning, etc.
Are you sure there's a significant set of people who have wifi devices but no data service? I am under the impression that the poor buy data-capable phones over laptops. Even if they can't afford data for their phone, the phone can accept free 4G if it were to be offered.
In NYC a lot of teens carry iPod or cellphones without data plans. They basically just attempt to connect to whatever free wifi is available. I can see this being a game changer for them since they'll always be connected now.
You, like almost everyone else responding to me, have missed the idea that smart phones without data plans still are data capable. (The number of teens without any sort of smartphone carrying around an iPhone that gets wifi is negligible.) The point is that the claim "we should support wifi because the poor don't have devices that can access 4G data" is mistaken.
Yes, there may be technical issues surrounding swapping SIM cards that makes a municipal 4G network unworkable, but this isn't the issue I'm addressing. It's discussed by one of the other commenters. And in fact, this downside would effect the poor the least since they could just pick-up a free municipal SIM card and us it continuously (since they often can't afford a network dataplan), while those who can afford network data plans would constantly need to swap SIM cards.
$30 android tablets with wifi are a pretty good option for being connected.
I have a few obsolete iPads with cellular modems paid up which I "loan" to people who need them. Modern tablets on wifi or even obsolete cellphones on wifi would be better.
(By "loan", I mean I pay the person to take care of some task for me and provide the device so they can do that, but it keeps them and their children online while they sort out their financial situation. Being a teenager without social internet in the US sucks badly at this point. You become "that kid".)
With regards to 4G, any spectrum which can be used for 4G has been bought for billions by the telecom oligopoly. If they wished the poor to have access to social connectivity, they would have done it. At least in the US we politically can't spin up a "basic level" telecom system as a government service. It isn't in our national narrative. So, engineering be damned, it's WiFi or nothing. And lets hope we can keep the telecom oligopoly from jamming Wifi. (Looking at you Verizon and T-Mobile with your LTE-U plans.)
I'm only one person, but I (and my family) have computers and smartphones but no data. It saves a whole lot of money to be on a prepaid plan when you have wifi in most places you go.
Right, but your smart phone is data capable, i.e., can accept a 4G signal. You don't have a dumbphone alongside a computer with wifi.
mwsherman's original question was "Why would municipalities offer free wifi rather than free 4G?". jws thought it might be important for folks who didn't have data plans. But my point is that if someone has wifi, they also very likely have a smartphone that's capabale of receiving a 4G signal, even if they cannot afford a data plan from their cell network provider.
How would you arrange it practically though? You would have to distribute sim cards, and people would have to swap those out with their normal sim card rendering them unreachable by phone. That's ofcourse assuming their phone isn't carrier-locked. Wifi seems a lot more practical unless phones are redesigned to cover this usage pattern.
> mwsherman's original question was "Why would municipalities offer free wifi rather than free 4G?"
This isn't quite how I read the comment, but here are some pretty good answers, I think:
1. They would have to either purchase suitable licensed spectrum (literally hundreds of millions of dollars in NYC) or find an existing carrier with which to form an MVNO relationship with generous enough terms that the city could afford to "resell" the service for free.
2. 4G would probably require NYC to deal with distributing phones or SIM cards, and then users would have to figure out how to unlock their devices.
> But my point is that if someone has wifi, they also very likely have a smartphone that's capabale of receiving a 4G signal, even if they cannot afford a data plan from their cell network provider.
That's a bit of an assumption. I'd like to see some numbers to support it. I know a bunch of people who have wifi capable devices but not devices capable of 4g. Or they have older smartphones that are not 4g capable.
I don't need to do anything to connect to my providers public networks. The keys are (I'm guessing) just stored in the SIM card and I'm automatically connected when there is WiFi available. LTE being super fast and all, I don't mind not using my data plan and having a low latency, high bandwidth connection without even lifting a finger.
You also seem to overestimate how many people can afford a data plan like we (probably) have. Or just young people. Or people who don't use the Internet that often at all. It is probably not ment for the average HN-reader.
>>The keys are (I'm guessing) just stored in the SIM card
They're not. They keys (and many other things) are stored in "carrier settings" that your provider pushes to your smartphone. In the past we had to set things up like APN & etc manually for the internet to work but not anymore with the settings being pushed automatically (if the provider supports it).
On iOS you can view the current version of pushed settings by seeing the Settings > General > About -> Carrier option.
"Carrier settings updates are small files that can include updates from Apple and your carrier to carrier-related settings, such as network, calling, cellular data, messaging, personal hotspot, and voicemail settings". [1]
The issue with LTE is that spectrum is limited. With 30MHz of spectrum, ~2 bps/Hz spectral efficiency and 3 sectors per tower you can provide at most 90Mbps of total downstream capacity per tower. With a density of 10k+ people per km2, a tower serving a one km2 area is easily overloaded. With towers costing hundreds of thousands of USD, things get pretty messy.
Wifi doesn't replace LTE, complements it. No doubt that the UX is still crappy, but SIM-based authentication can solve that.
Also, unlike LTE, wifi spectrum isn't licensed, anyone can set up a hotspot without paying billions to the government. Sure, we could all solve that by switching the entire world to an unlicensed, pay-as-you-use system, but that's unlikely to happen any time soon.
In other words, spectrum is the major constraint in urban areas.
The Wi-fi Alliance has created an standard called Hotspot 2.0, which aims to fix "Wifi UX" in these scenarios. From Wikipedia: "The idea is for mobile devices to automatically join a Wi-Fi subscriber service whenever the user enters a Hotspot 2.0 area, in order to provide better bandwidth and services-on-demand to end-users, while also alleviating mobile carrier infrastructure of traffic overheads."
PassPoint uses 802.1x and client certificates/PKI to authenticate. I have the Boingo Wi-Fi service client cert installed on my iPhone and it automatically connects to any -Passpoint network via a WPA2 connection and authenticates with the installed cert. It's really nice for traveling.
Many cities have free WiFi in whole downtown areas, on many streets.
The use case is by now more than fleshed out.
Users connect to it once, and done. It’s cheap, fast, and simple.
You don’t have to worry about falling back to roaming when you walk out of the range of the network, as your phone automatically stops downloading that huge file as soon as WiFi stops.
But most phones have no way to mark a specific 4G network as "good roaming", cheaper than normal.
> But most phones have no way to mark a specific 4G network as "good roaming", cheaper than normal.
Actually, this doesn't quite seem to be the case. When I went to Denmark a couple of months ago my phone didn't consider it roaming, since I have a contract where I have the same prices there and at home.
Yes, so the network is marked as "equal to normal". Like O2 and eplus do.
But there’s no way to mark a network as better than normal.
For example, if you connect to WiFi, Android will recognize it as a non-mobile network, and start doing updates, etc. If you start listening to a song via Play Music on WiFi, and then go out of range, it will automatically downgrade the quality, or stop playing and ask you if you want to continue.
If you are on "NYC 4G", and go back to your own 4G network, there’s no way for Android to stop streaming that FLAC song, because it doesn’t know a way to tell that one 4G network is equivalent to WiFi, but another is not.
> For example, if you connect to WiFi, Android will recognize it as a non-mobile network, and start doing updates, etc. If you start listening to a song via Play Music on WiFi, and then go out of range, it will automatically downgrade the quality, or stop playing and ask you if you want to continue.
Since the wi-fi radio is different from the cellular data radio in the phone, it's trivial to set different policies depending on which radio the phone is using. The settings to allow differentiation exist because we know implicitly that a wi-fi connection won't be draining your expensive data plan. The phone doesn't actually look at a connection's bandwidth/signal quality/etc to do this.
I'm trying to state this is a way that would just get massively downvoted.
It doesn't feel to me like internet access works the same as other utilities. The need for water per individual hasn't changed much in 80 years? 100years? more? The need for electricity has gone up a little since say 1940. Natural gas, and other utilities. Maybe even streets also seem relatively static?
But internet? So NYC installs gigi wifi today. In 10 years Korea and Japan and Singapore and China will have tera-wifi or peta-wifi and NYC would likely have nothing. Getting public tech services upgraded at or near the speed that tech accelerates seems highly unlikely. Having public wifi would seem like it would remove a huge incentive for anyone to come in and offer better faster service because competing with free will be hard for a very very long time.
Am I wrong?
Also, please don't devolve this into ISPs suck in America. I fully agree with that statement. I'm just not sure public wifi is the answer.
For example in Japan it was competition that got internet faster in general. Specifically Yahoo/Softbank Japan kept undercutting the competition with faster and cheaper internet and all the other ISPs were forced to respond. That doesn't seem to happen in the USA because of poor regulations (local government approved monopolies)
I'm just curious if public wifi is really the correct solution or if it's actually going to end up much worse in the long run.
Removing incentives to come in and compete is of no consequence when no one is doing that anyway. Gigabit fibre to the home has been all over the world for years (more than a decade?) and the vast majority of Americans are sitting here being treated like shit by Comcast/AT&T with connections that work barely or not at all at peak streaming time.
The free market has had its chance; we've gotten nothing but the token effort from Google Fibre in a tiny handful of markets.
I'm not sure public WiFi is the answer. I'd much rather see municipal fibre from homes to PoPs where consumers can ask to be patched to various ISPs. But competition just isn't doing it.
The U.S. market is smothered by local monopolies[1] and a global duopoly. Your argument would hold water if the free market has been given chance at all.
Local monopolies were granted by municipalities, certain officials and not the prospective end users. The very first winning bid from a cable company carried the monopoly clause. Once the monopolies had been granted, the market was no longer free; basically the market was (and is) no more.
I'd argue that a free market for cable / broadband access never ever existed.
> I'd argue that a free market for cable / broadband access never ever existed.
I agree. This is because that market was never regulated. A free market for POTS/DSL access did exist (and to a large extent still exists). Federally regulated mandatory line sharing made it happen.
It does not benefit the public good to run multiple, competing sets of infrastructure to carry commodities.
However, -as we see in things as disparate as Instant Messaging networks, remote-controlled lightbulbs, and wired Internet access- private business places far more value on capturing and walling in sections of the population than they do helping to facilitate a vibrant, competitive marketplace that promotes customer choice. From a business perspective, this lock-in makes sense. This isn't the best thing for society, though.
I think GP is thinking further back to ol' Ma Bell. If Bell hadn't been broken up, and competition-promoting regulations put in place, there would have been no need for municipal exclusivity agreements--Bell would have been the only carrier out there in the first place.
It's possible that Cable could eventually have grown to compete technologically, but in a world where Bell was still a single unregulated entity, it strikes me as unlikely that cable could have survived independently.
In a sense, yes. I would argue that the cabled infrastructure is not a natural monopoly at this stage of technology development, but the pre-existing regulatory framework and the local municipalities treated it as such. The effect was a market where local monopolies were bought and sold. Which leads to no free market for buying and selling last mile Internet access.
To build a parallel tax-funded infrastructure like the fine article says, would be just silly in those monopoly markets (instead of enabling commercial competition first; not sure if New York in particular has any meaningful competition, so this may not apply there).
I honestly don't care about fast public wifi. I will be happy if NYC (where I currently reside) has reliable 5 or 10mbps wifi. Fast enough to read websites - but not to watch movies or download massive size files.
This will actually address your point about private competition. City might offer free 10mbps wifi - whereas private companies will offer paid 50 - 5000mbps.
> The need for water per individual hasn't changed much in 80 years? 100years? more?
Unlikely. 100 years ago we didn't have nearly as much water infrastructure and appliances. It's difficult to find data on past years though but when looking at other countries that have less amenities than the United States the water usage is staggeringly different [1].
I would be shocked if, per person, the water usage hasn't gone up by double since 1915. Don't forget the infrastructure also has to handle a far denser population as well.
> The need for electricity has gone up a little since say 1940.
Define little. Almost everything we do today involves electricity whereas in the 1940s that was far from true plus the population has grown significantly since then. WorldBank seems to indicate it's been increasing (they have data since 1989) [2].
> Natural gas, and other utilities. Maybe even streets also seem relatively static?
Are you sure? Where are you getting this data? According to the EIA[3] natural gas usage has gone up about 5x since 1949.
> Getting public tech services upgraded at or near the speed that tech accelerates seems highly unlikely. Having public wifi would seem like it would remove a huge incentive for anyone to come in and offer better faster service because competing with free will be hard for a very very long time.
While I agree with you there is also a huge burden on getting around publicly allowed monopolies for the last mile (since the wifi needs some sort of data connection) so I'm not convinced any private companies are going to do this anyway due to that unless the government is going to offer some serious incentives or someone invents a better wireless technology which doesn't require a physical internet connection to the access point.
I believe your parent post was referring to household usage (which determines infrastructure needs in a city), while you are referring to economy-wide usage.
Energy usage per household seems to be essentially flat for the past 40 years according to Residential Energy Consumption Survey[1].
As for water, I can’t find any links at the moment, but I would guess that per capita indoor residential water use is also flat or slowly declining as appliances become more efficient.
Per household is kinda irrelevant for his point, is it not? You need the additional infrastructure to handle a more dense population and his point was regarding the upgrading of utilities.
I'm curious to see what sort of bandwidth "gigabit wifi" translates to in real life. In my experience with public wifi, it isn't a competitor to a wired ISP, its more of an alternative to 3G (and not as good as a 4G data connection). I'd be very surprised if a city government had found a way to make wifi feasible as a primary home or business internet connection when none of the companies working in the space couldn't.
For all the "American ISPs are terrible" rhetoric, they are profit driven and possibly evil but not incompetent. I feel like if NYC actually had a technology here that could put a home ISP out of business, the current ISPs would have bought it up first. NYC's wifi will be great for residents and tourists trying to check their email or do a quick Google search from their phone, but sustained gigabit connections for every client is way too good to be true.
>If NYC actually had a technology here that could put a home ISP out of business, the current ISPs would have bought it up first.
The capital allocation strategy for current ISPs has been to secure legislation prohibiting competition and sue threats out of existence. They might have invested in R&D as a hedge against that strategy not working, but that would seem to defeat the purpose.
Japan, and South Korea are mature, but China still has a lot of catching up to do.
What I think is needed, instead of passing lots of regulation and competition hurdle, Government should mandate All new housing to have shielded ( CAT8 ? ) Ethernet cables prewired. Much like electricity and water pipes. You dont buy a new flat and expect to rewire all electricity and water system. You pretty much expect them to be there.
It is the same thing with telephone wiring.
Then all new home / building will have a central point of connection, ISP will only be required to connect to that point, and doing some config to enable internet.
Doesn't realy allow for competition does it and in the USA and other developed countries people don't all live in massive high rise blocks.
Unfortunately its not uncommon in the UK for developers to forget about providing ducting for telco/cable and also working with Virgin/BT to put in the cabs and last mile.
There is a point, even with bandwidth, where you get reducing benefits to scale. just because we've had an insatiable appetite for faster and faster speeds doesn't mean that will always be the case. honestly is there much you wouldn't be able to do with a symmetrical 1gbps link? eventually we will stop worrying about faster speeds probably once we get to 1/1gbps or maybe 10/10gbps even with all the applications we can dream up it will happen.
Free wifi is very popular, though it makes me wonder why free wifi providers don't provide data the same way cellular networks do though 4G/LTE/etc, albeit in a limited area?
I don't know about the technical limitations but it seems a solution that would cover a much larger area, and most peoples phones already have mobile data enabled.
Are phones simply not set up to handle multiple mobile data networks ?
Without significant changes to most phones, you would need to issue SIM cards to all participants. It's not impossible to change, but the provider would need to convince 3GPP and OEM vendors to do so.
Because doing so would require them to either purchase licensed spectrum (way too expensive) or start a virtual carrier (an "MVNO") and then either provide their own SIM cards and hope that users can unlock their phones, or somehow enter into a roaming agreement with the big carriers.
It's radically easier to just offer Wi-Fi. There are also new Wi-Fi advances designed to reduce the friction of finding and connecting to a Wi-Fi network. A modern Wi-Fi network is also capable of dramatically greater speeds.
How would a city handle the pressure from telco's if they offer free, reliable wifi? Wouldn't all residential's switch to the free wifi, making telcos lose hundreds of thousands of customers?
Has there been a case where free wifi for a city has been implemented reliably where residential stopped purchasing internet from a telco?
Free public wifi, unless intensively engineered in cases where problems develop, can't promise 100% coverage to everyone's residence. Some people will get a windfall that effectively replaces their SP's service. Some will get only unusable connectivity.
SPs have a simple way to compete: Offer service that provides a lot more bandwidth than the free service. They should do that anyway, now that Google Fiber has announced a push into some major cities.
Would people in the "good coverage" area be able to run repeaters for this kind of WiFi provider so that, for example, their friend across the street could be covered? I'm not familiar enough with large-scale WiFi infrastructure to know whether they have mechanisms in place to prevent such practices (or if they'd want to).
Companies absorbed, dead links, contracts that had to be met at midnight, one booth with wifi, zero chance of getting anything like this out into NYC wasteland (where millions live). Soon the flashing giant ad monitors will all be strewn across midtown, not a byte will make it to deep Brooklyn not a one.
Do some googling on this one, follow the trail of disappearing companies involved, in the end, Google took it all over, that's the footnote, its all Google running the show now. But hey, more screens, that's Times Square. I'm ok with that. :-)
I hope the data they gather on people connecting to each gateway as they move around the city is anonymised (as much as it can be) and then made open and freely accessible. That'd be fascinating.
The physical "wifi booths" they are installing have 2 - 55" flat screen TVs that will run ads on them.
Back of the napkin math time: that's $41mil a year generated over 8000ish ("over 7500" is what the official announcement said) booths, which is $5125 or just shy of $500/mo in revenue per booth to hit that target.
I'm not in the physical display advertising business, but that would generally seem pretty doable given some quick research into NYC ad costs.
This is what I dislike most about free wifi. It's completely lacking in security and most are not aware of the issue. the use of mobile apps makes it worse. At least with a browser, the browser will warn if something suspicious is happening. But there's no way for a user to know if a mobile app is using https and using properly (not ignoring certificate errors).
Question:
How secure is LinkNYC Wi-Fi?
Answer:
The LinkNYC Private network is one of the first
free municipal Wi-Fi services in the country to
offer an encrypted network connection between your
device and the hotspot, securing all wireless
communications between devices and the Link.
I'm not sure about the feasibility of the encryption protocol but as an anecdotal note, a buddy of mine worked as a software engineer at Control Group and I almost considered interviewing there when I was looking around recently, so they are investing in a significant number of engineers at the very least.
"The full network will install more than 7,500 public hubs throughout the city, each replacing a pre-existing phone booth. Once completed, the hubs will also include USB device charging ports, touchscreen web browsing, and two 55-inch advertising displays. The city estimates that ads served by the new hubs will generate more than $500 million in revenue over the next 12 years."
Genuinly interested in understanding your viewpoint. If you have some time, please elaborate. I have always been under the impression that municipal fiber to consumers has the best benefits in the long run but I might have missed something.
Municipal Fiber is a solution looking for a problem. The overwhelming trend is toward mobile. Many people, particularly lower income people, have a smart phone but no computer. At the same time, municipal Fiber is incredibly expensive. Chatanooga spent $330 million plus on their network, to wire up about 71,000 subscribers. Plus, maintenance of that infrastructure is very expensive. Given that most states and municipalities are broke, and have neglected crucial electric and water infrastructure that already exists, they can't afford to take on the cost of a Fiber network.
The wireless technologies we have today don't deliver anywhere close to the throughput of fiber. There's no way municipal WiFi will support every household running a couple of Netflix streams at the same time in the evening, something that's (occasionally) possible on cable infrastructure. We've already seen from dense apartment buildings that WiFi does not deal well with more than a handful of active users at close range, even on separate SSIDs.
Some form of radio communication might be the answer, but there is very little unallocated bandwidth and it's bandwidth that we need.
If we're going to spend limited public resources on building network infrastructure, the target use case shouldn't be households watching a couple of Netflix streams in the evening on their iPads. We should instead build the technologies that can deliver the broadest connectivity at the lowest cost, so people can apply for jobs, kids can do homework, send pictures to their friends or parents, etc. Wireless wins big on flexibility, and can be deployed at a cost that might allow cash-strapped municipalities to offer it as a free service. You'll never achieve that with high-cost and high-maintenance wired infrastructure.
In the case of LUSFiber, part of the process included voting on a bond issue. It passed, and the build began. The effort has been successful enough to create a profit, which has been put back into the build, allowing them to reach more homes and businesses than originally planned, pay the bond off early, etc. (Source: Talking to a half-dozen of their business and residential customers in April, 2015).
LUSFiber is, lacking a better term, solvent.
Which cash strapped municipalities using public funds are you referring to?
Every New Yorker can already get good enough internet to apply for jobs and do homework, both from an ISP and a local public library. That need is already well served by current infrastructure; there's no point in duplicating it. If current options are too expensive, than subsidies would seem much simpler than new infrastructure.
Municipal fiber is not proposed in places that lack internet service, it's proposed in places that lack good internet service. People are upset that aging and oversubscribed cable company infrastructure isn't good enough for their media consumption desires, and municipal fiber is the solution.
Where we have problems getting people online at all is in rural areas, where 1) there are no municipalities and 2) distances are far too long for WiFi. There, the best option is low-cost point-to-point antennae on high ground. The hard part for rural Wireless ISPs has been finding someone to sell them transit at a fair price.
I've got better internet in rural New Mexico, due to a subsidy to a local ISP to put in fiber, than I do in a town in Massachusetts that is full of people in the tech industry.
The CISCO Wifi Certifications says 10 hosts per AP as a rule of thumb - you can probably push it to 15 or so but you would have to limit the connection speed.
Things work great with LUSFiber in Lafayette, LA. Municipal Fiber is a solution to the problem of communication services monopolies. Which is why ISP lobbyists throw a lot of effort to get states to pass laws prohibiting municipal service.
4G/LTE/etc is engineered for uncontrolled, outdoor spaces. It works more reliably than Wifi in that environment.
Wifi’s UX is poor in that environment – you’re already connected to 4G but connecting to Wifi requires effort. And one assumes that if one is on the street, one will be walking out of range of that Wifi within a few minutes. I don’t understand the use case.
(The sitting-in-Starbucks use case makes more sense.)
Politicians like the sound of Wifi. Users walking down the street, by and large, ignore it.