802.11 channel 14

This article in Daily Dot’s Kernel magazine, “The mystery of WiFi channel 14”, was recently posted in /r/TIL. This led me to discover that there apparently exists a conspiracy theory as to why WiFi channel 14 is not permitted in the United States.

This made me irrationally angry.

I have hence dug up the relevant FCC and even ITU regulations to explain why it’s not mysterious at all that channel 14 is not permitted.

Let us begin. What is channel 14? The 802.11 channels go like this:

(source: Wikimedia Commons. CC-BY-SA 3.0)

Channel 14 is all the way on the right there, spaced 7 MHz farther apart than usual.

It’s not just channel 14, for the record. As Wikipedia points out, channels 12 and 13 are generally not used either, because anything operating more than very low power will almost certainly cross into the restricted zone.

Wait. The restricted zone? Yes, there’s a restricted zone starting at 2483.5 MHz extending to 2500 MHz, per 47 CFR §15.205. No Part 15 device can transmit in this zone. The question becomes why. Part 15 doesn’t say.

So we look in the FCC allocation table (PDF warning). There’s conveniently an entry for 2483.5–2500 MHz:

Screen Shot 2016-05-29 at 12.43.46 AM

Let’s break it down. First, notice that this frequency band is allocated to a bunch of radio services mentioning satellites, in particular space-to-Earth communications. We then get a shitload of references to various notes. Let’s go through each of them. (Some of these notes do not appear in the FCC document and need to be searched for in the full ITU Radio Regulations. As of writing, it’s available here.)

5.150 defines the ISM bands. In particular, one of them is 2400–2500 MHz.

5.398 “In respect of the radiodetermination-satellite service in the band 2483.5-2500 MHz, the provisions of No. 4.10 do not apply.”

4.10 “Member States recognize that the safety aspects of radionavigation and other safety services require special measures to ensure their freedom from harmful interference; it is necessary therefore to take this factor into account in the assignment and use of frequencies.”

5.402 “The use of the band 2483.5-2500 MHz by the mobile-satellite and the radiodetermination-satellite services is subject to the coordination under No. 9.11A. Administrations are urged to take all practicable steps to prevent harmful interference to the radio astronomy service from emissions in the 2483.5-2500 MHz band, especially those caused by second-harmonic radiation that would fall into the 4990-5000 MHz band allocated to the radio astronomy service worldwide.”

US41 “In the band 2450-2500 MHz, the Federal radiolocation service is permitted on condition that harmful interference is not caused to non-Federal services.”

US319 “In the bands 137-138 MHz, 148-149.9 MHz, 149.9-150.05 MHz, 399.9-400.05 MHz, 400.15-401 MHz, 1610-1626.5 MHz, and 2483.5-2500 MHz, Federal stations in the mobile-satellite service shall be limited to earth stations operating with non-Federal space stations.”

US380 “In the bands 1525-1544 MHz, 1545-1559 MHz, 1610-1645.5 MHz, 1646.5-1660.5 MHz, and 2483.5-2500 MHz, a non-Federal licensee in the mobile-satellite service (MSS) may also operate an ancillary terrestrial component in conjunction with its MSS network, subject to the Commission’s rules for ancillary terrestrial components and subject to all applicable conditions and provisions of its MSS authorization.”

US391 “In the band 2495-2500 MHz, the mobile-satellite service (space-to-Earth) shall not receive protection from non-Federal stations in the fixed and mobile except aeronautical mobile services operating in that band.”

NG147 “In the band 2483.5-2500 MHz, non-Federal stations in the fixed and mobile services that are licensed under 47 CFR parts 74, 90, or 101, which were licensed as of July 25, 1985, and those whose initial applications were filed on or before July 25, 1985, may continue to operate on a primary basis with the mobile-satellite and radiodetermination-satellite services, and in the sub-band 2495-2500 MHz, these grandfathered stations may also continue to operate on a primary basis with stations in the fixed and mobile except aeronautical mobile services that are licensed under 47 CFR part 27.”


So. The reason channel 14 is restricted is because it conflicts with the allocation of 2483.5–2500 MHz to the mobile-satellite and radiodetermination-satellite services. In particular, that band is used for space-to-Earth communications, meaning any WiFi operation in that band would almost certainly overpower any satellite signals.

mystery solved…


Edit: One last note. On top of clearly doing absolutely no research whatsoever, the Daily Dot leaves us with this: “Though the channel is banned the consequences of using the restricted channel are not specified. It is considered a felony due to its illegality though it seems unlikely that the FCC will come knocking on your door.” The FCC takes its regulations extremely seriously: even in the Amateur Radio service, repeated willful interference can often lead to a forfeiture order well into the five figures. Should you ever cause interference to one of the satellite services mentioned above by operating a WiFi access point on channel 14–enough to be noticed, anyway–you most certainly can expect a knock on your door.

Diversity in tech: a different view

“Innovation and disruption do not come from homogeneous groups of people.” —PepsiCo president Brad Jakeman

Credit for this post goes to Selena Larson, who wrote on the problems that some had with the 2015 Anita Borg Institute Grace Hopper Celebration of Women in Computing conference. In particular, she pointed out that diversity efforts in tech focus a lot on gender to the exclusion of race, and that gender diversity in tech is often code for “more white women in tech”. (At least, that’s what I read out of it.)

This all hit very close to home.

I was compelled to take a second look at the UC Berkeley demographics information available in Cal Answers, this time at degrees awarded (i.e., actual outcomes) rather than admissions and student body makeup (inputs). The following table is a full listing of degrees in EECS and CS awarded by UC Berkeley in this millennium, broken down by race and gender.


In the past 15 years, seven black women have graduated with a computer science degree from UC Berkeley—as the administration likes to tout, the predominant source of elite engineers in Silicon Valley. Seven. Out of a total of several thousand. In more than half of the years—and in one entire six year stretch—not a single black woman graduated out of either EECS or L&S CS. In no year were there more than two. Hispanic women make up not much more.

We are not making any progress.

Is there a pipeline problem?

…and what if there isn’t?

This is the admissions data for UC Berkeley EECS and L&S CS for 2015 vs. 2014, made available on August 3.

Compiled by the author. Source for data in italics: Cal Answers.

A few thoughts:

On the pipeline issue, which gets so much press these days. The gender ratio among applicants has improved. In fact, there is a 34% increase in female applicants to EECS and a mind-boggling 55% increase for L&S CS. It remains, however, quite high.

The main thing about this chart that makes me think there must be something other than pipeline issues at work here is the yield rate.

While the admissions process improves the gender ratio somewhat, the ratio among SIRed applicants remains abysmal for EECS. There is a stark contrast between EECS, which actually has a worse gender ratio among SIRs compared to last year, to L&S CS, which has a substantially better one (with an astounding 120% increase in female SIRs). L&S CS has a 8.4% point improvement in female yield ratio, but for EECS there is a year-on-year decline. Clearly something about L&S CS is making it more attractive, because its female yield rate is now almost double that of EECS. We have to ask ourselves: why? There are many possible explanations for this disparity, and none of them involve pipeline.

An explanation offered on a Facebook discussion thread for this chart was that female admits are rarer and stronger, and thus generally have a wider selection of schools to choose from, pushing down their yield rates. This is a convincing theory, but without interviewing all 86 women who declined the EECS department’s admissions offer, there can be no proof. But if this is indeed the explanation, it does not reflect well on the department.

A note on acceptance rates: female applicants are accepted at a higher rate than male ones. The first instinct is that perhaps this is to balance out the much lower yield rate for women, since admissions departments generally target a certain number of matriculants, not admissions. Yet this is clearly not the case: for EECS, in order to match the 5:3 ratio in yield rates, the gender ratio for acceptance rates would need to be 0.60, not the 0.78 that actually happened.

Further evidence: looking at L&S CS, the admit ratio is nearly identical, despite their yield rate ratio being much better. Thus for L&S CS the admissions department was a net positive for gender equality, but for EECS it was actually a net negative.

The headline statistic is basically this: the EECS department had 3x higher year-on-year growth in female admissions over male ones, but by the time in came to SIRs that increase completely disappeared.

From all of this, the one thing we can conclude is that gender disparity in the yield rate appears to be a major problem blocking progress. And that’s something that can’t be blamed on the pipeline.

Stryver and Carton drink… a lot…

In his book A Tale of Two Cities (1859), Charles Dickens specifies of Stryver and Sydney Carton,

What the two drank together, between Hilary Term and Michaelmas, might have floated a king’s ship.

(Part II, Chapter 5)

Now, how much is this? Just how alcoholic were these two famously Bacchanalian figures?

First, “between Hilary Term and Michaelmas”. I interpret this to exclude Hilary term itself, so this is strictly the duration between the two. That makes it, using rough figures for Hilary term taken from contemporary standards at Oxford University, something around six months (March 25 to September 29).

But how much alcohol is necessary to float a king’s ship? I have no examples of a king’s ship. What I do have is the HMS Victoria, a first-rate ship-of-the-line launched by the Royal Navy in 1859, conveniently also the year of publication for A Tale of Two Cities. The HMS Victoria was the largest wooden battleship to ever enter service. It had a displacement of 6,959 tons (we’ll round to 7000). This is the minimum amount of alcohol necessary to float it, then, as per Wikipedia,

Another way of thinking about displacement is the weight of the water that would spill out of a completely filled container were the ship placed into it.

Obviously a completely filled container with less than 6,959 tons of wine and punch would be incapable of floating the HMS Victoria.

\[\frac{7000\text{ tons}}{6\text{ months}} \times \frac{907\text{ L}}{\text{ton}} \times \frac{\text{six pack}}{2.13 \text{ L}} \times
\frac{\text{month}}{730 \text{ hours}} = 680 \text{ six-packs per hour}

This is approximately \( 5 \text{ six-packs per person per minute} \).

Considering especially that the beverages of choice were “wine and punch”… this is a lot of alcohol.


Yes, my mind turns to Malibu Media again. Their effect on lost productivity in the United States must be vast, especially considering that their lawsuit count is fast approaching that of the infamous Jonathan Lee Riches.

TorrentFreak has an article out on the Tashiro case, reporting that Malibu has succeeded in issuing a subpoena to Comcast for Six-Strikes and DMCA notices. They wonder why Malibu Media would do this when they claim to already hold a smoking gun, the packet capture. (Update: upon second reading, I realize they never actually say this. Nevertheless…)

The answer is simple: the packet capture is not the smoking gun Malibu likes to claim it is. It only establishes the following: at certain times, a particular IP address, running a particular torrent client, sent to an IPP/Excipio server the pieces of a particular torrent. In particular it does not tie anything to a person, which is now Malibu’s problem as they try to decide whether to pursue the wife, the husband, or both.

I must admit however that I am completely unaware as to how six-strikes or DMCA information would establish anything further, as these systems are also IP-address-based, i.e. only the subscriber is informed, regardless of who actually infringed. I suspect this strategy is similar to the “Exhibit C” one that has been stopped by the courts, and indeed TorrentFreak reaches essentially the same conclusion. The additional request for the Tashiro household’s bandwidth usage statistics confirms that.