|
Getting your Trinity Audio player ready...
|
The medium of choice for wired telecommunications networks is fiber optics. Fiber can support much higher bandwidths than copper-based alternatives, and it’s no more expensive to deploy. It’s inherently more reliable. It’s the future. But the telecom incumbents are reluctant to abandon their copper-based networks as long as customers are willing to put up with them.
Fiber to the premises networks, which use fiber-optic cables to provide internet access directly to users of an internet service provider, extend all of the virtues of fiber to homes and businesses. A municipal network would offer not only superior bandwidth at reasonable prices but also would guarantee net neutrality and privacy, which are not guaranteed by the private sector.
More than two decades ago, Palo Alto was one of the very first communities to start thinking about creating a citywide municipal communications network based on fiber optics. In 1996 as a first step, the city council decided to deploy a “dark” fiber network — just fiber-optic cable connected to nodes, or common network boxes. With this system, customers lease unused “dark” fiber strands within the cable and then “light” the fiber strands with their own electronics.
The city’s original $2-million investment was paid back long ago, and by now the network has amassed about $26 million, which is sitting in a fiber fund ready to be spent on next steps.
I moved back to Palo Alto in 1998, in part because the buzz that the city was thinking about taking the next step: deploying a citywide municipal fiber to the premises network. A community group of advocates held informational meetings and spoke at meetings of the City Council and the Utilities Advisory Commission. In 2001, the city deployed a trial network to 67 homes to see whether staff could make it work (they could) and whether people would like it (they did).
From 2002 to 2004, the city commissioned a series of studies by consultants that found that a citywide municipal network was feasible, and there was enough interest in the community to make it financially viable. But then the project fell apart when the city couldn’t figure out the details of doing the financing.
Next, from 2005 to 2009, the city tried (unsuccessfully) to form a public-private partnership to create a fiber to the premises network. Unfortunately, that involved years of closed-door negotiations between the city and the prospective private partner, during which time advocates had no role to play, so community advocacy sort of dispersed.
In 2010, the city tried (unsuccessfully) to persuade Google to deploy its first fiber to the premises network here. In 2011, the city tried (unsuccessfully) to get a stimulus grant for the network. In 2012, the city looked into user-financed network, in which customers pay a substantial connect fee (of, say, $3,000), but the Utilities Advisory Commission, in a 4-3 vote, thought it was a bad idea, and staff never even asked what council thought.
In both 2013 and 2014, council voted to make “Technology and the Connected City” (a term that included both fiber and wireless) one of the city’s top three priorities. But in 2014, when Google announced that it was considering deploying fiber to the premises here, most of the city’s attention shifted to that possibility. Google then “paused” (gave up).
In 2017, staff proposed, and council approved looking into, a much tinier next step — just connecting dark fiber to more nodes (at a cost of $15 million) with no real plan or commitment for connecting nodes to premises.
Next month, staff will tell council that all its work on this next tiny step was in pursuit of the wrong idea, so it wants permission to start over with another version of the system. I think it would be much better for the city to do an engineering design of the citywide fiber to the premises network we want, in sufficient detail to make realistic cost estimates, and then commit to building out the network, in phases if necessary.
How could Palo Alto have gone from being so visionary in 1996 to being so clueless in 2019?
One reason, I believe, is that the city went out of its way to take to heart the failure of a municipal telecom network in Alameda. But Alameda made mistakes we can avoid. Its network was hybrid fiber co-ax (an inferior technology), not fiber to the premises. And it bet on a private-public partnership, but then the partner flaked.
Meanwhile, there are now 213 municipal fiber to the premises networks in the U.S. Many offer 1-Gbps symmetric residential internet service. Prices vary, but in Longmont, Colorado, it costs $49.95 per month (to those who signed up as soon as it was available). It’s not rocket science. We can do this.
I have launched a website, munifiberpaloalto.org, hoping to inspire the Palo Alto community to let the council know that a citywide municipal fiber to the premises network is important to them, and that the council should figure out a way to make it happen. On the website, I’m hoping to post a list of people who support citywide municipal fiber to the premises. If you’d like to be on the list (and your home or business is in Palo Alto), please let me know.
Councilman Greg Tanaka has launched a petition on Change.org (“Bring fiber optic access to Palo Alto!”) for the same reason. The petition wants to “Hold the people who are in charge of this to be accountable for their actions and deadlines.” I think the council should hold staff accountable, and the public should hold the council accountable.
Jeff Hoel, a retired electronics engineer, has been advocating municipal FTTP since 2002; he can be reached at jeff_hoel@yahoo.com.
Is the $50/month price in Colorado subsidized by taxpayers or is that a self-sufficient business? I thought pricing was the entire reason that previous city-sponsored efforts have failed.
Can we please let this die and stop. Palo Alto has never shown that it can run anything in a professional and cost effective manner. Why does anyone think that this would be different?
You have a bounded customer base and have pressure from commercial providers. Take a look at google fibre and their withdrawl from expansion. Does Palo Alto really think they can do if better/faster/cheaper?
/marc
>> How could Palo Alto have gone from being so visionary in 1996 to being so clueless in 2019?
You missed something important. First, for personal needs, you can get 36 Mbps down/8 Mbps up -through DSL+WiFi (!) Plenty good enough for videoconferencing and watching 4K video streaming simultaneously. That is (dual-channel) vDSL – a recent version of the “slow” technology using old twisted-pair phone company copper wires. YMMV, depending on your location. But, you can also do better with some other non-DSL options. For reference, 4K video streams are usually 11-15 Mbps, with some very high-speed test streams up to 25 Mbps.
Second — not many people need that much to their homes anymore, because businesses (including nonprofits of course) all locate their public servers at hosting facilities/services – AKA “in the cloud”. Why drag 1-10 Gbps to your house when you can locate a (virtual or real) server somewhere with direct nationwide high-speed networking? “The cloud” really has greatly reduced the need for upstream bandwidth from the home. That is the real reason we don’t have it. Not many people need it. What -requirements- are there for 1 Gbps symmetric speed to the home or small (retail/service) business?
Please let this die. I’d rather have the private sector compete for my business, and based on experience the city will screw this up and it will suck up a huge part of the city budget that’s needed for essential services. This will be a boondoggle.
@resident, Longmont’s municipal FTTP network, NextLight, is not subsidized by taxpayers. NextLight is ahead of schedule for paying back the bonds that financed the network.
I think that pricing is NOT the entire reason the City has not moved forward with municipal FTTP so far. For example, in 2004, the hang-up was what financing option to use. On 07-07-04, UAC considered several financing options.
https://www.cityofpaloalto.org/cityagenda/publish/uac-meetings/4129.pdf
@marc, Most people think the City’s electric, gas, and water utilities are run in a professional, cost-effective manner. If you don’t, be thankful that City Council and the Utilities Advisory Commission are willing to listen to what the public has to say.
Google Fiber gave up when they realized that Wall Street expected a faster return on investment than utilities (like FTTP) can deliver. That’s no reason municipalities shouldn’t have municipal utilities.
Fiber is not a utility in the traditional sense of the word like water, power, wastewater, etc. The fact that Wall Street wouldn’t get behind Google’s fiber, as Jeff just acknowledged, says it all. If it’s too risky for investors then it’s too risky for us taxpayers. Let this die.
“Most people think the City’s electric, gas, and water utilities are run in a professional, cost-effective manner.”
That’s somewhat of a stretch. The City’s electric, gas, and water utilities are steady-state flywheeling on a long-established, relatively low-tech infrastructure. A fiber network is neither. Good performance delivering electric, gas, and water does not necessarily translate to good performance delivering SOTA communications services.
A new fiber network requires a substantial capital outlay, hence a bond issue. What risk can you cite to prospective bondholders? What interest must you pay?
If the wise guys (ATT, Comcast, …) aren’t rushing in to mine this putative bonanza, the wise fool asks why.
On the other hand, why not form a company, raise the capital, build the network, and operate it at a decent profit yourself? It’s the Silicon Valley Way.
The City of Palo Alto current fiber network is successful and makes a profit. As Jeff Hoel said, it has a $26 million dollar surplus. It’s run by the same City of Palo Alto Utilities department that provides electricity and gas at rates lower than what PG&E provides.
I support this. I think it can provide higher speeds, better service, privacy at a lower cost than the commercial alternatives.
Internet service to the home has become a utility. The city through the Utility department can provide this service like electricity and gas. It’s not clear to me that for profit utility services are viable, just look at what happened with Google Fiber and PG&E. Municipal services are a better model.
“The City of Palo Alto current fiber network is successful and makes a profit.”
That’s the city’s famed “dark fiber,” which is arguably a large factor in creating our jobs-housing predicament. It amounts to a passive wholesale operation which tech-savvy subscribers can use on a light it or leave it basis. That’s a long way from a viable retail operation with customer support for tech-naive subscribers.
I agree that we could do much better, but I think the day of fiber to the premise is gone. Having worked since the early 90s on fiber field trials and deployment across europe and the US the concept is wonderful but the cost is too high per homeowner. 5G is far more economical to deploy and will bring equivalent bandwidth for what most homeowners require. ( please let’s not use this thread to argue about the radiation that some are worried about). In the end 5G will allow local connectivity and provide movie downloads in the order of seconds compared to today. It is also focusing on mobility and connected cars and homes, fiber in your home will still require a broadband router anyway, I’d prefer to skip the cost of connecting homes by any physical cable at all and just move to wireless. And this can be done on a per homeowner basis and a new 5G router. Verizon is already deploying this in limited areas in California and soon in San Mateo county for trial sites.
As a telecommunications engineer with technical and historical economic understanding of natural last mile monopolies (wired telephone, gas, water, electrical power distribution, roads etc., fiber to the home has been recognized as stupid economically for last thirty years, with many failures (Google latest). Nor does it provide any compelling technically compelling feature.
I argued with Mr. Hoel, five years ago, that it was stupid. Unfortunately at the time Google was pursuing it which I pointed out was a mistake, which Google now gets and has stopped further fiber expansion. So, if Google has figured it’s a mistake why not Mr. Hoel? Google stupidly hired a bunch of fiber engineers who weren’t willing to jeopardize their jobs by pointing out the imposibility of success of fiber to the home since they surely understood why fiber to home had failed previously. If not they weren’t good engineers who analyze their failures. Google has had similar corporate management failures in many areas, it’s still a one product company, 80% revenue still comes from search ad revenue. Unlike Amazon AWS has clocked Google on Cloud computing.
The correct solution is a micro (block level on power or light poles) WiFi system using the WiFi 5ghz band, poke to home roof top repeater to in house, usually in 2.5 GHz band. The 5ghz WiFi band has excess capacity because it has worst wall penetration, and 6db lower antenna capture gain, but better on rooftops and allowing directional beaming, gain with smaller antennas.
As aside about the hypocritical Telecom logic, 5ghz was added to the free WiFi band because it was a junk band useless for longer range communication; that is, it had no commercial value. Even 2.5ghz WiFi band was parked on the Microwave oven ISM band which was thought at the time as a junk band useless for communication due to interference from ISM non-communication devices
The pole, light top nodes may be fiber but can also be fed for cheaper price from cable. The city might be able to sue the cable guys for access at real cost plus some reasonable profit since the cable and twisted pair DSL are common carriers. ISP’s pay less than $1 a month per customer for long distance internet access. Their real cost is less than $5 per month per customer. The installed cost of cable is less than $500-$1000 per customer but amortized over 20 years, is only a few dollars per month. The maintenance cost is higher (but less with wireless solution). Cable monopolies justify high prices, $5000 value per customer which they get if system sold which is a high intangible value representing the impossiblity of anyone competing and the $60 a month they can gouge.
Historically, the only way to break last mile Telecom monopolies has been through wireless technology on last link to user. Satellite TV was the only viable alternate to cable and has a significant market share despite being late to market. Similarly cellular telephone technology broke the wired telephone last mile monopoly. Many players could offer some wireless broadband solution but it’s not in their direct financial interest, hence the monopoly, lack of competitors.
However, Google might do it and pay for it with ad revenue as a more benign search monopoly keeping an internet access monopoly in line. Such, a system might improve internet access improving Google”s ad returns, and applying competitive pressure on cell phone spectrum prices. In principle we need more free bandwidth at cellular frequencies since there now is no justification for bandwidth to be owned as long as operating in a band required type accepted, smart radio technology that uses the band most efficiently and least disruptive. The highest utillity per mhz in the world is the free 2.5ghz WiFi band.
More free bandwidth to gain access to Telecom nodes which can be provided by multiple providers will drive down intangible cable and cellular wireless spectrum value by increasing competition.
Sorry, this got long winded but it’s a complex system problem, involving radio and communications technology, monopoly economics, FCC regulatory policy, and missinformed political issues.
GigE fiber is necessary in today’s technology. Certain services are not that usable without. For example, hard disk backup to the cloud or surveillance video. We can’t wait for a private company to pick this project up and if the city picks up this project wisely (with all the doubts based on pass experience) it would be great. We can learn from other cities.
GigE fiber is necessary in today’s technology. Certain services are not that usable without. For example, hard disk backup to the cloud or surveillance video. We can’t wait for a private company to pick this project up and if the city picks up this project wisely (with all the doubts based on pass experience) it would be great. We can learn from other cities.
> A municipal network would offer not only superior bandwidth at reasonable prices but also would guarantee net neutrality and privacy, which are not guaranteed by the private sector.
This claim is egregiously misleading.
Compromises of user web privacy almost always occur at the remote end of a web link, at entities such as Yahoo or Google or Facebook or the like, not at the customer end where the city fiber would be. A city fiber system would provide little actual defense of user privacy. Nor would a city system mitigate net neutrality violations occurring on the 99.999+% of the web that would be beyond its purview.
I still use dial-up. People these days are in too much of a hurry.
Who cares about social media and whatever POTUS or the Kardashians are tweeting?
I also use a flip phone. Modern day life is for the birds.
Posted by Expecting resident,a resident of Old Palo Alto
>> GigE fiber is necessary in today’s technology. Certain services are not that usable without. For example, hard disk backup to the cloud
Except that so much of that private data is now in the cloud already, at Google or Apple iCloud or Amazon or wherever. It is already uploaded, and, people download, at higher speeds, what they are using currently. How many private pictures of your cat’s face do you need to have local to you? Making the cloud your primary storage is the current response to the telecom monopolies discussed by Wayne Holcombe.
>> or surveillance video.
A bigger problem. You can do one high-quality stream with the uplink bandwidth I discussed earlier, but, if you need 10 or 20 streams, you do have a problem. Is that a common-enough requirement to be a successful business model? I don’t know. If it is, then perhaps physical security companies should be driving this, with personal internet access just an added bonus.
@KR – the private sector has been competing for your business. Wait – no. Xfinity has a strangehold over my neighborhood and ATT is only offering a 10mbps plan. Both providers list 2yr contracts that double in price when ended, and have a 1TB data cap. Of course the city may bungle it up – but the private industry has proved itself to not be reliable.
@Anon – 4k at 11-15mbps? Netflix recommends 25mbps, and considering 1080p alone goes up to 60mbps, for those looking for very high quality streaming 36mbps will not suffice. Yes, you could download it locally – but when a 4k BluRay is ~100GB, you’re looking at a ~1hr download in optimal conditions.
@Old School God forbid someone has other use cases.
Posted by Citizen, a resident of Charleston Meadows
>> @Anon – 4k at 11-15mbps? Netflix recommends 25mbps, and considering 1080p alone goes up to 60mbps, for those looking for very high quality streaming 36mbps will not suffice.
Actually, my wireless router and internal app measurements agree– Netflix uses ~11 Mbps often, and 15 Mbps sometimes, for 4K w/wo HDR. Yes, this is much less than the minimum 25-30 Mbps content suppliers stated initially. And, you can find beautiful 25 Mbps 4K videos on YouTube. But, Netflix seems to have switched to HEVC, which can be very, very efficient with pre-processing– i.e. recorded content. Doesn’t always help that much with realtime encoding.
http://www.dtcreports.com/weeklyriff/2018/04/26/netflix-well-leave-no-codec-or-customer-behind/
For the same quality level, pre-processed recorded video can be more efficient/use less bandwidth, than live/realtime video for, e.g., live sports.
“…when a 4k BluRay is ~100GB, you’re looking at a ~1hr download in optimal conditions.”
If you believe there’s a viable business opportunity here, why not form a company, raise the capital, build the network, and operate it at a crazy high profit yourself?
Make the call. Investors are standing by.
AT&T was pulling fiber through our area last week. My wife and I talked to one of the linemen as they were pulling it passed Duveneck to a pole on Hamilton. The speeds they were talking 600 meg to 1 gig should be competitive. I’d be willing to bail on comcast in a heartbeat for those speeds.
@Wayne Holcombe, Thanks for using your real name.
The economics of FTTP have changed a lot over the last 30 years. I think it makes sense for the Palo Alto to focus on whether it’s economically feasible today and in the future. Broadband Communications Magazine’s interactive database of FTTP networks in the U.S. and Canada lists 1,176 FTTP networks, of which 218 are municipal.
http://www.bbpmag.com/search.php?s0=1&cols=-co-me-st-ve-gr-te-se-ty-in&st=&ve=&gr=&te=&se=&ty=-mun&qco=&qme=&qan=&qus=0&qmu=&qsu=&qpa=&qin=0
As of last year, in the U.S., FTTP was being marketed to 40.8 million homes, of which 18.4 million homes had FTTP connections.
https://www.lightwaveonline.com/business/market-research/article/16675952/ftth-now-second-to-hfc-in-north-american-broadband-homes-passed-fiber-broadband-association
I tried to find evidence that you argued with me five years ago about this, but I was unsuccessful. You (as “Wayne H”) posted a reader comment about a 03-18-13 article,
https://www.paloaltoonline.com/news/2013/03/18/palo-alto-reboots-its-fiber-dream
but I didn’t. (If you’re thinking of another article, what was it, and what name were you using then?)
@An Engineer, Sure, Yahoo, Google, Facebook, etc. can compromise user privacy. But so can ISPs.
AT&T used to charge extra not to spy on your data.
03-27-15: “AT&T’s plan to watch your Web browsing — and what you can do about it”
https://arstechnica.com/information-technology/2015/03/atts-plan-to-watch-your-web-browsing-and-what-you-can-do-about-it/
Then they decided not to do that.
09-30-16: “AT&T to end targeted ads program, give all users lowest available price”
https://arstechnica.com/information-technology/2016/09/att-to-end-targeted-ads-program-give-all-users-lowest-available-price/
But no law says they couldn’t do it in the future.
The Trump-era FCC rescinded the net neutrality rules adopted by the Obama-era FCC. And Congress is having trouble getting them reinstated.
05-24-19: “47 Democrats cave on net neutrality after GOP calls bill ‘dead on arrival'”
https://arstechnica.com/tech-policy/2019/05/47-democrats-cave-on-net-neutrality-after-gop-calls-bill-dead-on-arrival/
So you want a city that
could not build a library faster than Sunnyvale built a stadium
could not design a bike bridge in the time that East Palo Alto Built one
could not run its own Internet connection
I dont think the city has shown they have the savvy to provide fiber.
“@An Engineer, Sure, Yahoo, Google, Facebook, etc. can compromise user privacy. But so can ISPs.”
Thanks for the prompt reply to my May 20 posting. So, if CPAU becomes an ISP, it can compromise user privacy. Presumably for money. But it would never do that, would it?
You daily provide your personal info to potential and proven compromisers like Apple, Yahoo, Google, Facebook, Instagram, YouTube, … whatever your channel to them is, CPAU fiber or 5G or ATT. So going fiber is no cure.
Don’t forget (or learn here, if case be) that compromisers, whoever they are, need your net address to respond to when you request something from them, so you provide it to them, and that address is easy for them to log for future use and sharing with other compromisers. IPV6 is consigning ISP-level DHCP with the dinosaurs. You’re doomed in any service scenario.
Jeff, I had long arguments with you in some local, fiber (I think) user group and not in Palo Alto online. Not sure I can find the links.
However, referening “umm” comments if Google can’t succeed with fiber to the home why can Palo Alto do better, rather as umm suggested they have almost always done worse than private businesses which is not to say large private monopoly type businesses are guilty of gouging and keep prices up but they keep out small businesses and city’s like Palo Alto due to economies of scale difficulty in getting users to change to new system since they charge below the price threshold users will bother considering all if their other monthly fees. Any competitive “cabled” system is not economically viable unless you get over 50% sign up or unless you offer a compelling function that people will pay more for. Gigabit speed is nice but not really compelling.
4k TV requires 11mbs but really is just sold on marketing hype. With realitively close 30 degree viewing angle or group viewing standard HD TV has better resolution than eye acuity. 4K is waste, of course the TV guys won’t tell you that.
Cell wireless overcame the sign up percentage problem since a cell tower needed only a few percentage sign up (hundreds users) within it’s several mile urban radius or 10 mile suburban or freeway distance. But also several per cent of consumers we’re willing to pay a $100 a month (over 25 years ago) for a wireless telephone service which was a pretty compelling function for this who needed/could afford it. So, cell wireless solved both economic problems with wireless technology of both economic “cabled” technology.
Like I mentioned earlier the only viable alternative is some micro cell wireless technology which is technically viable from the regular cellphobe providers but probabably won’t since it’s not economically favorable for them.
One government solution is to force, regulate Comcast and ATT to provide service at real cost plus 20% which is probably around $5-$10 a month which is the real hidden problem. Any new service has to be able to provide service below this cost, not less than $50-$60 per month since Comcast can drop prices to match.
Posted by Wayne Holcombe, a resident of Mountain View
>> Any competitive “cabled” system is not economically viable unless you get over 50% sign up or unless you offer a compelling function that people will pay more for. Gigabit speed is nice but not really compelling.
>> Gigabit speed is nice but not really compelling.
Uplink, or, downlink? AT&T and Comcast are optimized around downlink bandwidth, and, adding satellite and OTV in, downlink bandwidth for user content isn’t, or shouldn’t be anyway, a big issue.
The main difficulty is uplink bandwidth. Not that many people need it, so, your 50% (or whatever) threshold is difficult to overcome.
-Unless-, as was mentioned earlier, more than 50% of people decide they need online video surveillance. Crime is much lower now than when I grew up, but, if things get really bad again, then I guess everyone will want to have 10 Nest Cam IQ streams going at high quality. Then everyone will need 40+ Mbps bandwidth uplink constantly.
https://nest.com/support/article/How-much-bandwidth-will-Nest-Cam-use#Hello-video
>> 4k TV requires 11mbs but really is just sold on marketing hype. With realitively close 30 degree viewing angle or group viewing standard HD TV has better resolution than eye acuity. 4K is waste, of course the TV guys won’t tell you that.
Uh, no. Actually, the difference between HD and 4K is clear. Quite a few people just don’t care, though. They just want to see talking heads anyway. For them, standard TV was/is just fine. People who always thought standard TV was awful and would only watch film movies in a theater– HD is the -minimum- necessary for high-quality 16mm (e.g. Panaflex) or standard 35mm movies. 4K HDR is good enough for all 35mm film. But, you are probably not a super-visually oriented person, I’m randomly guessing.
Either way, we are agreed that you only need 11-15 Mbps for 4K movies because HEVC is so good. What isn’t clear is what you need for live sports– the codecs are not as efficient in realtime. It would be nice if you could actually see the baseball in flight, for example, when watching a baseball game. Higher framerates, better action cameras, HD or 4K. I’m not sure where the sweet spot is, and, what bandwidth is required, but, current practice hasn’t caught up yet with the human eye.