Can Palo Alto Afford the Vanity of its Ruling Elite? Part 1 of an infinite series: Intro and Fiber-to-the-Premises | A Pragmatist's Take | Douglas Moran | Palo Alto Online |

Local Blogs

A Pragmatist's Take

By Douglas Moran

E-mail Douglas Moran

About this blog: Real power doesn't reside with those who make the final decision, but with those who decide what qualifies as the viable choices. I stumbled across this insight as a teenager (in the 1960s). As a grad student, I belonged to an org...  (More)

View all posts from Douglas Moran

Can Palo Alto Afford the Vanity of its Ruling Elite? Part 1 of an infinite series: Intro and Fiber-to-the-Premises

Uploaded: Oct 31, 2013
"Ruling elite" in the title is meant to provoke thought about how far the priorities of the decision-makers diverge from those of the typical resident. One of the reasons that Palo Alto has such a huge infrastructure deficit is our ruling elite's spending on vanity items. In discussion of issues, delivering value to the residents routinely loses out to their need to have a service be "world-class", "world/national leader", "leading edge", "a lighthouse to the nation"? The immediate instance is "Palo Alto moves ahead with citywide fiber plan: City Council approves 'master plans' for Fiber to the Premise, wireless network" (PA Weekly, 2013-10-30).

In the business world, there is a warning "You can always spot the pioneers, because they are the ones with the arrows in their back" which has been popularized as the admonition to never buy Release 1.0 of a product, and to try to avoid any X.0 release. A more cautious version of this is "Wait for Release 2.1", under the reasoning that changes between Release 1 and Release 2 are often so great that you lose much of your investment in Release 1.x installations (integration, training?). If you are new to this, good summaries can be found in discussions of "First-mover Advantage" (Wikipedia), which is often instead a disadvantage.
Aside: I had a neighbor who invested in restaurants. I asked about this, saying that my impression was that most restaurants failed. He responded that that was true, but that often that was the result of the debt load from remodeling and initial marketing. If not for that extra cost of being a first-mover, the restaurant would have been operating profitably. To minimize the debt-load problem, what he did was match his potential restauranteurs to a location that would require minimal remodeling, that had equipment that could be largely reused (rather than purchased new), and that already had an established clientele for that type of restaurant.

It is not that our ruling elite is oblivious to this problem. Rather, some see bragging rights as priceless. If they were spending their own money, that would be one thing, but here we are talking about public monies. Others regard Palo Altans as so rich that they are morally obliged to subsidize the development of new technologies for the rest of the US/world (through the high costs of being an earlier adopter).

On to our immediate example: Fiber-to-the-Premises (FTTP) (previously -Home and FTTH for anyone doing historical searches). This has been under consideration since the late 1990s with a limited trial 2001-2005 (66 homes). A survey in 2012 found little real interest in FTTP: Few residents saw a need for its capabilities to the extent they were willing to pay for it, even at a moderately subsidized rate. The conclusion of our ruling elite was has been the answer is more subsidies.

I have been on the fringe of the FTTP discussions since about 2000, and have been dissuaded from being more involved because it is so thoroughly dominated by the true believers. My first question is "Is the higher speed of FTTP actually usable?" I have DSL capable of a download speed of 13 Mbps ( + favorable location). However, I rarely see this speed utilized except briefly and except in special situations (off-hour downloads of large popular files that would be cached on edge servers, for example, Windows updates). Mostly, I see speeds peaking at 1-3 Mbps with multiple gaps. My estimate is that there are only a few minutes a month when my current connection runs at over 75% capacity. The problem doesn't seem to be what FTTP would address--"the last mile" to my house--but rather the capacity of the remote web servers and the intervening network. The analogy is someone who ignores that the pipe between the water main and their meter is only a half-inch diameter and decides the way to get better water pressure is to install a 6-inch pipe from the meter to the house.
Additional data: Netflix's ISP Speed Index (Weblink) shows Google Fiber at 3.41 Mbps (average), far below its rated capacity, and with the various competing vendors and their technologies providing 1.2-2.59 Mbps, again suggesting that the bottleneck is not in the last mile.

My next question is "What would the much higher speeds of FTTP be good for?" The first answer is almost always "It would allow you to download a Hi-Def feature-length movie in a few minutes?" My response is two-fold. First, having to download a movie before viewing it was a short-term problem a long time ago--it was largely solved by streaming. Second, for quickly downloading humongous amounts of data to be compelling, that task must be unpredictable, the cost of delay high, and it must affect a significant portion of Palo Alto residents. None of the needs that I have heard the advocates present seem to come close to fitting these criteria.

The next answer I get is that having very high speed networks will inspire the creation of new software and services. But that begs questions about assembling critical mass of users and developers. And with the current emphasis on mobile devices and mobile apps, who is going to invest money and talent in such systems (that would be affected by whether Palo Alto has FTTP)? There seems to be no real analysis, only hope. And as many have observed, "Hope is not a strategy."

Answering these and similar questions would provide the foundation for the big question: What are the advantages and disadvantages of moving forward now versus waiting? From what I have heard, although there is a certain number of lessons-learned from the few other cities installing FTTP, too many were negative lessons. Consequently, Google saw itself needing to subsidize the learning with its experimental/trial deployments. And notice that Google deliberately chose to not deploy FTTP where many of its employees live.

I have heard (unconfirmed) claims that the City has spent over $2M on consultants for this over the years, plus staff time. The next step is "only" a few hundred thousand more, preparing for expenditures in the tens of millions.

In the most recent Staff Report (Weblink), I don't see any of the above questions being answered. What I do see are more invocations of vanity/prestige, for example, City's goal of becoming a "leading digital city" (pg 3), "Progressive cities" (pg 7).

There is already a Town Square Forum discussion going on under the news article (Weblink). What I would like to foster here is a more analytical discussion. That means no exhortations to be world-class (or similar), no reiterations of complaints that the US trails other countries in broadband speeds, no generic claims that City of Palo Alto government is incapable of running such a system (I am not judging that claim, simply placing it outside the focus of this discussion),... You get the picture.

The Guidelines (Weblink) for comments on this blog are different from those on Town Square Forums. I am attempting to foster more civility and substantive comments by excising violations of the guidelines.
What is it worth to you?


Posted by CrescentParkAnon., a resident of Crescent Park,
on Oct 31, 2013 at 2:01 am

> "You can always spot the pioneers, because they are the ones with the arrows in their back"

LOL, that's a great quote. Not a good idea to aspire to be world-class unless you have world-class expertise. Buying that is problematic, expensive of just a rip-off.

> for quickly downloading humongous amounts of data to be compelling, that task must be unpredictable, the cost of delay high, and it must affect a significant portion of Palo Alto residents.

What on Earth does this mean?

Just because Palo Alto is in Silicon Valley does not give it expertise in running a network service. So unless they can get something very basic and solid and non-cutting edge that if the people who run it decide to screw around with the city and they can easily fire them all and find new people to take over - forget it.

Instead of getting a bunch of lousy consultants who are all going to try to do what consultants always do - double-talk and triple-charge, a good thing to do would be to start with what the city is trying to do and see if you can get some tech people here to kick it around. HINT, when things don't make sense of the level of the discussion gets too high, then call BS. Stick to what a critical mass of people can talk about agreeably and explain clearly to others what they are thinking and doing.

It would be nice to have a city Internet. Google did something in Mountain View, and there was talk about ... some other city somewhere ... maybe start there, get as much info as possible. Something tells me the city don't know diddly about the interwebs. But worse, they don't know diddly about being tough and cheap either! ;-)

Posted by Douglas Moran, a Palo Alto Online blogger,
on Oct 31, 2013 at 3:11 am

Douglas Moran is a registered user.

RE: CrescentParkAnon: on "... What on Earth does that mean?"

If the need to download humongous amounts of data is predictable, you can accommodate lower speed by planning ahead.
If it is unpredictable but you can accommodate the delay by doing something else, then there is little value to the higher speed.

As to number of people for whom very high speed is valuable enough, the density of subscribers on a street, neighborhood, ... determines the economics of running the cable to them.

For example, is currently rolling out a fiber system in several place (Santa Rosa, SF) and has said that after those trials, it will consider installing fiber in other locations based on customer densities -- they currently rent their DSL/phone lines from AT&T and say that those fees are high enough to warrant stringing their own fiber in those denser locations.

Posted by Douglas Moran, a Palo Alto Online blogger,
on Oct 31, 2013 at 4:07 am

Douglas Moran is a registered user.

RE: CrescentParkAnon on the City's expertise needs

There is a lot of successful out-sourcing of network construction and management by private companies. And a lot that problematic and/or over-priced.

In the consultants' report done in the mid-2000s, I recollect the consultant failing to consider what in-house expertise would be needed to manage such contractors. Instead, their recommendation was to hire a contractor/consultant to manage the contractors. My understanding is that such arrangements dilute accountability, often yielding bad results.

Hoping that someone with knowledge/experience/... on this issue will chime in.

Posted by Norman Beamer, a resident of Crescent Park,
on Oct 31, 2013 at 8:30 am

Plastic bag ban is another example, in my humble opinion.

Posted by Susan Fineberg, a resident of Greenmeadow,
on Oct 31, 2013 at 10:36 am

Building an industrial waste treatment plant with unproven technology in the Baylands is another example. With nearly half a million spent on the analysis of concept, it's still a non-starter.

Posted by Boris Foelsch, a resident of Palo Verde,
on Oct 31, 2013 at 11:19 am

I can't get the DSL speeds you do, so I would be willing to pay more to get better speed. However, unless it's cheaper than cable, it's not a good value to me; I still use DSL. There is a need for higher speed than DSL can provide for many. However, there's an upper limit on what that's worth. I don't see any need to spend public money on subsidies for this. Having said that, I do hope that faster access would be available for less. I have been told by AT&T technicians that there is fiber into the box in my front yard. I just can't get anything really fast into my house without paying Comcast rates. I continue to get by with modes 3 Mb/s service. I can stream video and that's (barely) good enough.

Posted by Douglas Moran, a Palo Alto Online blogger,
on Oct 31, 2013 at 1:17 pm

Douglas Moran is a registered user.

RE: Boris Foelsch

I sympathize with your situation -- it sounds to be very close to mine before I switched from AT&T to And I know that there are many others with the same problem.

However, the *policy* question is
- Is there enough need across the City to justify FTTP? As Boris points out, although he would very much like the addition speed, it isn't worth what he suspects he would have to pay to get it.
- Are the extra costs of being an early adopter worth it?

Posted by Douglas Moran, a Palo Alto Online blogger,
on Oct 31, 2013 at 1:26 pm

Douglas Moran is a registered user.

One aspect I forgot, and that the City seems to have forgotten:

To be competitive with Comcast and AT&T, the FTTP offering would need to offer not just Internet and telephone, but also TV. Negotiating with the content providers (eg ESPN) is difficult -- There have been big licensing battles between the monster cable companies and the content providers with periods where the cable TV customers lost some of their channels.

In an earlier consultants' report, there were a few companies that might handle creating bundles of channels for FTTP, but they were small (because of the small size of the FTTP market) and didn't seem to have the experience or "market muscle" to deal effectively with the content providers.

Posted by Fred, a resident of Barron Park,
on Oct 31, 2013 at 11:45 pm

When I moved here 10 years ago and heard about this issue, I was shocked that a small city like Palo Alto would actually think it made sense to invest in pioneering such an effort. I guess Doug may have it right - instead of letting corporate and individual citizens be innovators and pioneers, the city councilors feel the need for the *city itself* to be one. We moved here from a similar small university town back east, which had no illusions about being a pioneer - it just tried to keep the roads paved and plowed, the grass mowed, and the library open late. Why Palo Alto can't just do the same baffles me.

Posted by Midtown, a resident of Midtown,
on Nov 1, 2013 at 9:47 am

So folks, what happens when Comcast offers Internet at zero dollars per month while PA is charging $50 a month for FTTP? Cable Coop of Palo Alto tried to run a cable and Internet service and failed. Comcast now owns all those assets including the cables. Just pay Comcast or AT&T to install FTTP and let them run it. Problem solved.

Posted by Laurent B.K. Mgbango, a resident of Duveneck/St. Francis,
on Nov 1, 2013 at 12:33 pm

[Post not in English: Deleted by blogger: ]

Posted by Chip in Barron Park, a resident of Barron Park,
on Nov 1, 2013 at 8:08 pm

Chip in Barron Park is a registered user.

Re: The problem doesn't seem to be what FTTP would address--"the last mile" to my house--but rather the capacity of the remote web servers and the intervening network.The analogy is someone who ignores that the pipe between the water main and their meter is only a half-inch diameter and decides the way to get better water pressure is to install a 6-inch pipe from the meter to the house.

Doug: You and any other engineer like me would understand this analogy, but I think you should explain it to all the other readers who are not. This is a great analogy, but only if you understand it.

Posted by Jeff Hoel, a resident of Midtown,
on Nov 3, 2013 at 1:11 pm

Re CrescentParkAnon,

Google implemented a Wi-Fi network in Mountain View. It wasn't very fast, or very reliable, or very supported. Since the service was free, it certainly wasn't a proof-of-concept that it could be financially successful. Personally, I'm interested in FTTP but not wireless.

I try to keep up with what's happening with municipal FTTP networks -- in Chattanooga, Lafayette, UTOPIA, etc. And I tell Council what I know.

On 05-14-13, Council's Technology-and-the-Connected-City (TACC) Committee directed City Manager Keene to put together a Citizen Action Committee. TACC didn't provide any further direction, but such a committee might include tech people. As of 10-28-13 the Citizen Action Committee had not been put together. Eight people have applied to be on the committee, including me. City Manager Keene has said he is free to choose anyone he wants, and also to seek out people who didn't apply.

Posted by Eric, a resident of Leland Manor/Garland Drive,
on Nov 3, 2013 at 2:13 pm

At our house, we're currently paying $55/mo for sonic DSL (free nation-wide calling, caller ID, etc.) - but we're too far from the central office to get much above 2Mbps down. I would happily pay double this for higher speed if I knew the money were being invested locally. My perception is that Comcast/ATT regularly squeeze people with every possible up-sell with the goal of pulling in $150/mo or more. I don't see much incentive for incumbents like them to lower their prices over time, or provide better service.

The challenge for the city will be to arrange a very basic service coupled with VOIP (+TV?) and this is where things get tricky. I think a good example to look at is Salt Lake City which has been running a substantial project for more than 10 years. According to their FAQ, 50Mbps up/down currently costs around $65/month. One barrier is that getting hooked up costs $2,750 (or $300 down and $30/mo for 10 years). Note that according to the Salt Lake City Tribune (Sep. 16, 2013 article), the average $65/month is a substantial reduction compared to when the service first debuted in 2012 ($299/mo). If residents had confidence that the $3000 they are spending would be worth it in the long term, I could imagine a lot of people signing up.

I hope the commissioners studying the options have solid data on the economics here. I sure don't claim to but here's some rough example numbers... Let's say you sign up 20% of residences - call it 5000 homes; if you charged $100/month, that's $6M/year. Could that come close to supporting all of the costs? How much of the cost would need to be spent on new equipment vs. network operation / support calls / billing? Is the potential revenue high enough to interest an existing provider from handling the service side?

Equipment costs are certainly lower than they were 10 years ago when Salt Lake City first launched their project. And CPAU already has experience running the dark fiber ring. So the question is whether it's possible to add aggregation switches / other equipment in a cost effective way and how to negotiate contracts with service providers to take care of things like peering, network operations, billing?

Posted by Jeff Hoel, a resident of Midtown,
on Nov 3, 2013 at 2:35 pm

Re Doug Moran's comment about TV:

Lafayette's municipal FTTP network offers TV services. It gets much of its TV content through the National Cable Television Consortium (NCTC), which is able to negotiate discounted prices for its members. When Lafayette first applied to join NCTC, they were turned down, probably because Cox is an influential member of NCTC and Cox competes with Lafayette's municipal network. But Lafayette took legal action and eventually NCTC relented.
Web Link

Longmont's FTTP business plan, written with help from consultant Uptown Services, says that it doesn't need to provide traditional TV services for its municipal FTTP network to be financially successful. It plans to offer a 1-Gbps symmetrical Internet service to residences for $49.95 per month. A referendum, Measure 2B, is on the November 5th ballot. It asks whether the city should finance the FTTP network with bonds; if the measure fails, the build-out will still proceed, but it will be on a pay-as-you-go basis and it will take a long time.
Web Link

Posted by pat, a resident of Midtown,
on Nov 3, 2013 at 3:21 pm

This is the best discussion of FTTP/FTTH/whatever that I?ve seen since the topic came up at City Hall. Thank you, Doug.

The problem with so many of the city?s grand plans is that

(1) the people with the big ideas don?t know anything about the implementation or even if it?s possible to implement

(2) they hire consultants, but since they themselves don?t know much about the topic they are easily taken in by the consultants

(3) they don?t know how to hire or manage consultants -- see Web Link

(4) they use hope as a strategy instead of writing a business plan,

(5) while they?re spinning their wheels and spending our money on illusions, the basic needs of the city like road maintenance go unheeded and unfunded.

If the city can?t even manage a construction contract like the Mitchell Park Library, how can it possibly manage something as complex as the fiber business?

I?ll add another good quote to complement Doug?s: ?Hope is a good breakfast, but it is a bad supper.? Francis Bacon

Posted by Jeff Hoel, a resident of Midtown,
on Nov 3, 2013 at 3:37 pm

Re Midtown,

Over the years, most municipalities that have implemented successful FTTP networks have first taken the step of asking the telecom incumbents to implement something that meets the community's needs -- in the opinion of the community. But the telecom incumbents usually don't agree to do this. In 1999, the City issued an RFP for a "Universal Telecommunications System." The RFP got no bids -- which is a good thing, because the minimum requirements were not nearly ambitious enough.

If you meant that the City should pay a telecom incumbent to build a FTTP network that the City would own, and run it in the public interest, I see no evidence that a telecom incumbent would agree to that. If you meant that the City should pay a telecom incumbent to build a FTTP network that the incumbent would own, and would run on the incumbent's terms in the incumbent's interest, that makes no sense to me.

Comcast will never offer Internet service for zero dollars per month, but if it did, it could be sued for predatory pricing. Anyhow, what sort of Internet service would Comcast hypothetically offer for zero dollars per month? In Kansas City, Google is offering a "free" Internet "basic" service of 5 Mbps down and 1 Mbps up (once a $300 connect fee is paid) for up to 7 years. It's also offering a 1-Gbps symmetrical Internet service for $70/month. Google expects Google Fiber to be financially successful, so that tells me that Google believes that a lot of its customers will prefer its $70/month 1-Gbps service to its free service.
Web Link

The cable "assets" of Comcast and AT&T in Palo Alto are mostly copper (coax for Comcast, twisted-pairs for AT&T), with a tiny bit of fiber. It wouldn't cost them significantly less to implement FTTP than it would cost the City.

Posted by Marrol, a resident of Embarcadero Oaks/Leland,
on Nov 3, 2013 at 3:38 pm

[Trash-talk. Deleted by the blogger.]

Posted by Fed up, a resident of Midtown,
on Nov 3, 2013 at 4:03 pm

I have lived here for the entire duration of the fiber to the home/premises debate, and am fed up. Fed up with the delays, and fed up with the service from the private providers. I had Comcast for TV and internet, and the TV service was so awful that I switched to DirecTV. By awful, I mean unwatchable. Many channels appeared with "snow" on screen, making the picture so obscure as to be not visible. Other channels had so many compression artifacts (macro blocking, video skipping, audio dropouts, etc.) that they were also not watchable in any enjoyable way. And Comcast did try to fix the issues - switched out my set-top box numerous times, replaced the coax cable inside and outside my home, and even said they made some repairs at the pole across the street. The result: no improvement whatever. Same problems on same channels, after all "repairs" were done.

Compared to the service I had from the much-maligned Cable Co-op, DirectTV is inferior. Much more expensive, fewer of the channels I care to watch.

As for the Comcast internet service, it is unreliable, not very fast, and expensive. By unreliable, I mean that my neighbors and I have experienced multi-day outages (no internet connectivity at all) on more than one occasion. The last time this happened, Comcast tried to make me pay for the repair to their equipment at the pole, which is what they determined to be the issue, after I went 5 days with no service. Only after I complained to the California Attorney General did Comcast waive the charge.

So, would I be interested in a city-provided service? Yes, if it would be reliable, fast, and within 10% of the (very high) costs I pay now.

As for the argument that streaming is a substitute for downloading, that depends on your use case. I may not choose to watch an entire streamed movie in one sitting. I may want or need to download it for later viewing, or partial viewing now, and more later. I may want to view it more than once. And I may be downloading a large amount of data other than entertainment. Please don't feel entitled to decide for me whether streaming or downloading is appropriate or necessary.

Posted by Douglas Moran, a Palo Alto Online blogger,
on Nov 3, 2013 at 6:19 pm

Douglas Moran is a registered user.

RE: Fed up
This comment is an example of why it is so hard to give credence to the arguments of the advocates for FTTP.

First, the final paragraph seriously misrepresents my argument. He omits "largely" from my statement "... largely solved by streaming." And he seems to ignore the context of my statement as a response to the FTTP advocates' frequent portrayal of downloading movies as a compelling justification for FTTP.
Advice to advocates: Misrepresentation of skeptics is a strong signal that your case is weak or deceptive.

Second, most of the argument is an example of the Fallacy of Hasty Generalization. Just because he and some others have poor service doesn't mean that enough other premises in the City have bad enough service for them to also support FTTP (Similarly for the opposing direction: Just because a resident is in an area that has good enough service doesn't generalize to there not being enough demand for FTTP elsewhere).

Third, there is an implicit argument that because the incumbent (Comcast/DirectTV) provides poor service (to him and others) then whatever the City's implementation of FTTP provides will necessarily be better. Again, this is a common fallacy.

Too many advocates (Palo Alto and elsewhere) don't seem to appreciate that when they lead with logical fallacies, misrepresentations and other improper argumentation, then many people will dismiss what follows, if they even bother to continue to listen.

Posted by Eric, a resident of Leland Manor/Garland Drive,
on Nov 3, 2013 at 7:44 pm

>> Third, there is an implicit argument that because the incumbent (Comcast/DirectTV) provides poor service (to him and others) then whatever the City's implementation of FTTP provides will necessarily be better.

I think the argument would go like this... We trust the city to maintain our water supply, sewers and roads. And for the case of Palo Alto we trust the city to provide electricity and gas service. If fiber connectivity were considered similar infrastructure, could the city provide that as well in a cost-effective manner?

The answer may lie elsewhere - have other municipalities/projects had success in running such an endeavor? Profit/loss are one measure; customer satisfaction and cost of service are still others.

The main benefit is that when a corporation provides poor service, you may have limited recourse. If a municipality provides poor service, the leadership can eventually be held accountable to the public. Again, lessons from other projects could be very instructive.

Posted by Douglas Moran, a Palo Alto Online blogger,
on Nov 3, 2013 at 8:59 pm

Douglas Moran is a registered user.

On extrapolating from other services (water, electricity...):

An interesting question. Products and services can fail when they are handled by organizations whose cultures and skills are a mismatch. I had this experience several times in high tech. FTTP would seem to have enough differences from the Dark-Fiber utility that this issue needs serious consideration (as Eric suggests).

I have skepticism about the City's management of FTTP from dated experience. In 2000, I was in a tech startup on California Ave. We wanted to connected to the City's dark fiber which passed right by our office, with an access point less than 300ft away. However, dealing with the City was impossible -- we wound up getting multiple DSL lines. Before and since then, the City has acknowledged this problem multiple times and said they were going to fix it. But I kept hearing stories of serious problems (thru 2008 when my sources aged-out).

Posted by Jeff Hoel, a resident of Midtown,
on Nov 4, 2013 at 12:57 pm

Doug's citing of Netflix's ISP speed index data is misleading. In this case, the "bottleneck" is not what Netflix's servers are capable of providing. It's only what Netflix thinks the customer can use. The graph doesn't say, for each ISP, how much last-mile bandwidth the customer has, and that's crucial to the point Doug is trying to make. The graph also doesn't say, for each ISP, what kind of video is being sent.

I think the graph is saying that if last-mile bandwidth is not the bottleneck, then Netflix will stream at 3.71 Mbps, to send the highest-quality standard-format video it has. But if the last-mile bandwidth is the bottleneck, then Netflix will reduce the quality of its video to a bandwidth that the customer's connection can handle.

It's not clear whether the "Google Fiber" last-mile bandwidth is 1 Gbps (the $70/month option) or 5 Mbps (the "free" option) or some of each, but for this data, it might not matter.

Netflix actually recommends having bandwidths of 7 Mbps for "super HD quality" and 12 Mbps for "3D" quality.
Web Link
I assume that Netflix thinks that if your Internet service has these bandwidths, then no bottleneck elsewhere in the Internet will prevent you from benefiting from them.

Other streaming video companies have similar requirements.
Web Link
Netflix will reduce the quality of its video stream to what it thinks the customer's last-mile bandwidth can handle. If Netflix thinks the customer's last-mile bandwidth is insufficient for HD, then it won't even try to send an HD video.

Conceivably, a premises might want to watch more than one streaming video simultaneously. In 2010, the average number of TV sets per household was 2.93 (although the average number of people was 2.54). (Google "more TV sets" AND "2.93" AND "2.54".)

This 2012 article talks about television sets with 3840 x 2160 pixels -- 4 times as many as the 1080p HD format.
Web Link
These TVs might require a streaming bandwidth of 250 Mbps or 500 Mbps for best quality, or 16 Mbps or 32 Mbps for reduced quality. Or the video might be downloaded at less than real-time speed and watched later. I'm not saying that every living room will need one.

My interest in FTTP is not for streaming video, but I think it can help to make municipal FTTP successful financially.

Posted by Jeff Hoel, a resident of Midtown,
on Nov 4, 2013 at 2:40 pm

Re Doug Moran (6),

I'd like to learn more about the experience of the California Avenue tech startup where Doug worked in 2000 when it looked into getting a dark fiber connection from the City. Why does Doug say that working with the City (in this context) was "impossible"?

I don't know what the City's terms were in 2000. Today, as I understand it, the customer has to pay an initial fee just to talk with the City. Then the customer has to pay for all the City's installation expenses up-front. And the customer has to pay a monthly rate that depends on the length of the backbone fibers its using. (It's more complicated than that, but those are the basics.)
Web Link

In this 12-18-12 audio interview, Josh Wallace, Key Account Manager with the City's dark fiber network, says there are about 70 dark customers and hundreds of dark fiber connections.
Web Link
In other words, some customers didn't find working with the City to be "impossible."

What was the address of the California Avenue tech startup? Where exactly was the nearest access point? Was the fiber between the startup and the access point going to be aerial or undergrounded? If undergrounded, was conduit for pulling the fiber already available? Where was the other end of the sought dark fiber link? PAIX in downtown Palo Alto? What was the City going to charge for the installation? For the monthly fee? Was the startup competent to light the dark fiber?

Doug mentions the fact that the dark fiber passed right by his office. But, of course, that doesn't make any difference. You can only connect to dark fiber at its access points.

The Fiber Internet Center, located at 2635 Park Blvd, is a reseller of the City's dark fiber connections. It's only a couple of blocks from California Avenue. Did the startup ask them whether they had any good ideas?

If municipal FTTP had been available on California Avenue in 2000, would the startup have signed up for it instead of getting multiple DSL lines?

I don't think that the startup's experience with trying to get a dark fiber connection in 2000 is a reliable indication of how hard it would be FTTP once the City starts to offer it.

Posted by Douglas Moran, a Palo Alto Online blogger,
on Nov 5, 2013 at 1:54 am

Douglas Moran is a registered user.

The two preceding posts by long-time FTTP advocate Jeff Hoel demonstrate a basic mistake common in Palo Alto politics: He chooses to argue rather than trying to convince.

RE: The Netflix example: My point was that it didn't seem to show that people wanted and needed more speed. One would expect that stats based on users of Netflix streaming would have a disproportionate number of subscribers to higher speed networking, and one would expect them to skew the average upward. Yet there we are with a 3.41 Mbps average that falls closest to Netflix's lowest tier: DVD quality (3.0 Mbps). With three steps above that -- HD (5.0 Mbps), Super HD (7.0 Mbps) and 3D (12Mbps) -- why isn't that average higher? Even if most subscribers are for the free (5 Mbps) service? And if most Google Fiber subscribers are choosing "Free", isn't that evidence against "a need for speed"?

RE: Dark Fiber (second of those two messages): I cited a 13-year old example as a basis for a skepticism that remains to this day. Hoel wants details, including ones already given. Really? This comes across as petty harassment.


The persistent failed leadership of a long line of City Councils has created a system where issues are decided more as a trial by ordeal than a rational decision process. One can often win by dragging out the process for so long that people with other perspectives give up from exhaustion and frustration. Staff and developers are particular good at this game, but are hardly the only interest group doing it (Please, please dear readers, do not list them -- they are too many and an unwelcome tangent).
History: In 2007 I wrote an essay (Weblink) on this attempting to have this issue be part of the Council election. Fail (and multiple times since). I have made several attempts to talk to the current City Manager about this, but he has been uninterested.

However, attempting to simply outlast everyone else can be a risky strategy because every now and then a breath of sanity intrudes and the failure to make a convincing case in a reasonable period results in a proposal being shelved. I say "shelved" because Palo Alto rarely kills anything, it just becomes undead, rising up periodically to terrorize the populace.

Posted by CrescentParkAnon., a resident of Crescent Park,
on Nov 5, 2013 at 3:53 am

> One can often win by dragging out the process for so long that people with other perspectives give up from exhaustion and frustration.

Yes, a major component of the "Palo Alto Process", nothing happens until the "riffraff", i.e. residents are completely fatigues and out of it, and more and more of them learn not to bother to venture into areas where they are told they don't belong.

I don't really get ... is more internet speed being argued against here, or is that a tangent as well. I thought the point was that the city's core competence is not at running an ISP?

Posted by CrescentParkAnon., a resident of Crescent Park,
on Nov 5, 2013 at 3:59 am

By the way ... I see no "follow this discussion" ability here on these blogs ... is that deliberate, an oversight or what? It would seem to be useful for a blog/conversation to be able to follow new posts without having the check in regularly and go searching for it. PAO seems to be obscuring a lot of conversations instead of improving its interface to find them and comment.

There is a major long conversation somewhere about people's reminiscences of Palo Alto growing up that shows up and then disappears now and again ... but that can be "subscribed to" ... why not blog postings? Or did I miss something?

Posted by Douglas Moran, a Palo Alto Online blogger,
on Nov 5, 2013 at 4:25 am

Douglas Moran is a registered user.

RE: CrescentParkAnon (2 comments)

1. I am not arguing against more Internet speed. What I am arguing is that there doesn't seem to have been adequate consideration of what the value of additional speed is to residents (and businesses) and whether that value is worth the cost.

2. On "following" the blog: The PAOnline IT people are aware that this feature is missing and that it is desirable. However, I haven't heard when they hope to have it available.

Posted by Crescent Park Dad, a resident of Crescent Park,
on Nov 5, 2013 at 12:10 pm

FTTP is a colossal waste of city money. I pay $60 for 25Mpbs via Comcast and am very happy with the service (we use DirecTV for video).

[From the blogger to readers: Please no more of this type of comment. This blog is not intended to be a (bad) surrogate for an opinion poll (on the ultimate decision), but rather to help shape the discussion with facts, analysis and reasoned perspective.]

Posted by Jeff Hoel, a resident of Midtown,
on Nov 5, 2013 at 2:42 pm

Re Doug Moran (7),

I'm just trying to understand why you said that when your startup tried to get a dark fiber connection in 2000, "dealing with the City was impossible."

I don't know why Netflix users on Google Fiber aren't demanding more than 3.41 Mbps average. Perhaps most don't have TVs capable of showing HD or 3D -- yet.

I didn't say that most Google Fiber subscribers in the Kansas Cities are choosing the "free" 5/1-Mbps service rather than the $70/month 1/1-Gbps service. On the contrary, I said that enough people signed up for the $70/month service to convince Google that Google Fiber could be financially successful there.

This 2013 primer, "What Fiber Broadband Can Do For Your Community," explains what FTTP is good for. On page 11, there's a discussion of "Why We'll Always Need More Bandwidth."
Web Link

In 2002, the City commissioned a phone survey by DataCycles that found sufficient community interest in FTTP services to make a FTTP network financially feasible. In 2012, the City commissioned a phone survey by RKS, reported here:
Web Link
The intent of the survey was to find out whether people would be interested in FTTP services even if they had to pay a substantial up-front connect fee. On PDF page 45, it says 16% supported building out FTTP and would be willing to pay a $3,000 connect fee, and 20% were willing to pay the connect fee even though they did not support building out FTTP. In other words, 36% were willing to pay a $3,000 connect fee. The experience of other municipalities is that if you get a 36% take rate, you can be financially successful even if you don't charge an up-front connect fee. The survey didn't find out what the take rate would be if there were no connect fee, but I think it would be higher than 36%.

If enough people sign up for FTTP services, then FTTP will be financially successful.

Posted by Douglas Moran, a Palo Alto Online blogger,
on Nov 5, 2013 at 5:17 pm

Douglas Moran is a registered user.

Jeff Hoel,
1. On dark fiber circa 2000: What I remember was that I got the basic info for our IT manager to pursue and he eventually gave up, reporting that the City seemed geared to dealing only with large, slow-moving organizations (such as H-P).

2. On Google Fiber: I didn't say that you claimed that most of KC was choosing "Free" -- I used "Even if" to mark the enumeration of an alternative.

3. The numbers you cite from the Staff Report (Weblink) (pg 4)don't seem to be for those who would be willing to pay to get FTTP at those price. The report has the question worded differently, in effect asking if the City should enable residents to become subscribers. I couldn't find the exact questions in the document, and as we all know, minor differences in wording and question order can produce wildly different responses.

Posted by anne, a resident of Another Palo Alto neighborhood,
on Nov 5, 2013 at 5:44 pm

You know more about this than I do. I'm confused as to why and how the City is involved.

Why is the City trying to do this? Are they directing our utilities to do this? Did our utilities say it was a need? Who told them it was a need?

Did they assess said need in the context of our goals and needs? Is the money from something like this coming from the same pot as a safety building? I\'m happy with my Sonic service. Sure, more would be nice.

I guess I just don'tt understand this -- are there actual PlANS to spend money this way? Does the public have a choice?

Posted by 18 year resident, a resident of Barron Park,
on Nov 5, 2013 at 11:17 pm

How can residents get the city council to stop forcing this issue on the public?

Posted by Anne, a resident of Midtown,
on Nov 6, 2013 at 11:29 am

I love the series title as it reflects my opinion of our City Council - more interested in vanity projects than maintaining existing infrastructure properly. FTTP sounds like yet another boondoggle to me and I don't want my tax dollars spent on it. I want our City Council to maintain our existing roads and other infrastructure so we don't need special bond measures. I've been Treasurer of my homeowners association for close to 15 years. As HOA board members our fiduciary duty is to maintain our common area and protect property value, not to develop enhancements we can't afford. HOA's must use Reserve studies to project expenditures 30 years into the future, thereby collecting enough for operations and maintenance while keeping the monthly dues affordable. Of course I realize that an HOA is a very different entity than a municipal government, but I think the principle is a useful one and that spending money on dubious projects like FTTP when our roads are in deplorable condition is putting the cart before the horse. Do we need a Measure E to prevent FTTP from going forward?

Posted by HUTCH 7.62, a resident of Old Palo Alto,
on Nov 6, 2013 at 12:26 pm

Really wish the ruling elite would fix the roads and sidewalks around here

Posted by Jeff Hoel, a resident of Midtown,
on Nov 6, 2013 at 5:20 pm

Re Anne, a resident of Midtown,

If the citywide municipal FTTP network is paid for, in the long run, by the customers who subscribe to its services, would you be OK with that?

Chattanooga's citywide municipal FTTP network now has more than 50,000 customers. Its business plan said it could be financially successful with only 28,000 customers.
Web Link
They predict that by next year, the municipal network will have more customers than Comcast has (in its footprint). That's not a "boondoggle." That's the opposite of a boondoggle -- the epitome of practicality.

Posted by Douglas Moran, a Palo Alto Online blogger,
on Nov 6, 2013 at 5:56 pm

Douglas Moran is a registered user.

I am calling a belated halt to further debate on whether FTTP is a good idea / ... / boondoggle.

The intended focus of this blog is how to get a rational discussion of the issue, not to debate right/wrong.

That said, I don't believe that FTTP is inherently a boondoggle/... However, the recommendation to the decision-makers (Council) and the statements by those decision-makers predominantly represent logical fallacies: the Bandwagon Fallacy (we should have it because other cities do) and Emotional Appeal (Appeal to Pride, Appeal to Envy). This strongly indicates that the FTTP decision was not made rationally.

Follow this blogger.
Sign up to be notified of new posts by this blogger.



Post a comment

Sorry, but further commenting on this topic has been closed.

Stay informed.

Get the day's top headlines from Palo Alto Online sent to your inbox in the Express newsletter.

California must do a better job spending cap-and-trade revenue
By Sherry Listgarten | 5 comments | 2,511 views

Got the Munchies at Hardly Strictly? Your Weekend Guide.
By Laura Stec | 2 comments | 1,992 views