Request for grammaticality judgment

A quote from my Linguistics 100 lecture notes from April 14, 2014: “(graduate) syntax courses often filled with acrimonious arguments re: grammaticality”. Fellow linguistics students know what I mean.

The first rule of grammaticality judgments is to never trust your own. Therefore:

xkcd — the brilliant webcomic beloved by millions of readers every week — creator Randall Munroe delivers hilarious and informative answers to questions you probably never thought to ask. It’s time to meet your hero.

The above paragraph comes from promotional material for an appearance Randall Munroe is making in Berkeley on September 12. My question is thus: is that em-dash parenthetical permitted?

My own intuition says that you can’t split up an NP like xkcd creator that way.

Temperature dependence of quartz

The thin lines are various CPU core temperature measurements; in purple, NTP‘s reported PLL frequency, which tracks the apparent drift of the motherboard’s real-time clock, which is based on a quartz crystal oscillator—which is evidently very temperature sensitive (hence why temperature-compensated crystal oscillators (TCXOs) and oven-controlled crystal oscillators (OCXOs) exist).

pll_freq_vs_temp

English possessive -‘s as a clitic

A lot of linguistics-related posts recently…. this one will be short.

This is the title of the current top post on reddit’s front page (or mine, anyway):

Over a year ago I paid for the family in front of me’s groceries without thinking much of it. Today, I got this in the mail.

That’s a whole lot of separation between the head noun and the possessive marker.

(Stuff like this is why the possessive -‘s is considered a clitic rather than a case marker.1)

“more than I never expected”

TODAY in misnegations, we have “more than I never expected“.

The correct term is of course “more than I ever expected”. The cause of the misnegation is clear: the phrase represents something not expected, but oddly the negation is implicit in the word more, and so the negation rules of Standard English (which do not feature “negative concord“) do not accept a negation on ever or expected. As Geoffrey Pullum is fond of saying, our primate brains are sometimes simply not capable of handling such subtle intricacies.

A Google search for “more * than I never expected” returns 1,460,000 results. Most of these are not examples of this error though, as the wildcard went across a sentence boundary, matching grammatical passages where a sentence began with “I never expected” after a sentence ending with “more than *”. Google Search ignores punctuation entirely, so I don’t know of any way to refine the search further. The top few actual examples:

came away feeling they were more fearful of my experience than I never expected.

That’s something that’s inherently more fulfilling than I never expected when I started.

The third day I applied it was more longer and thicker than I never expected.

made me cry so badly more worse than i never expected.

The one I found that inspired this post: “a lot more to the PNG file format than I never expected” (myers.io).

Internet policy

IT IS fairly well known that on certain topics, such as the Internet, Congress simply does not “get it”.

This is bizarre, because it is certainly not for lack of good research data. The Government Accountability Office (GAO) released preliminary findings today regarding internet providers’ implementation of “usage-based pricing” schemes, otherwise known as data caps. It hits the proverbial nail on the head on several fronts.

Right at the start:

  • Access to broadband Internet is increasingly seen as crucial to improving
    • communications,
    • quality of life,
    • commerce, and
    • economic growth and innovation.

(emphasis added)

Internet access—and as they point out, specifically fast internet—is no longer just a luxury, but crucial for life. Going even further than that, several countries have recognized internet access as a basic human right.

The internet is also traditionally a platform for innovation, not of passive “use”. Lawrence Lessig and his acolytes speak frequently of an dangerous encroaching “permission culture“; this concept can be extended to internet service, which is increasingly becoming a one-way street in which users may only do what their ISPs allow them to (or alternately, what they pay [extra] for), rather than the internet being a platform for exploration and innovation.

Some focus group participants thought it was more fair to pay only for the data used – akin to utilities, such as water or electricity.

A quick note on this bit, which of course is dead wrong: as Dane Jasper (CEO of Sonic.net, my ISP) has pointed out, and indeed as the GAO points out (see next bit), Internet transit is essentially free. It does not cost ISPs more to carry more data.

Jasper’s example was that of steak, but to stick to the “utilities” concept, if you use more electricity/water, that does actually cost the utility more money. Utility rates are basically the marginal cost, in particular because utilities are regulated natural monopolies.

The marginal cost of Internet transit is essentially zero, and ISPs are not regulated natural monopolies with strict price caps either. Would that it were so.

Some wireless ISPs told us they use UBP to manage congestion; wireline ISPs said that congestion is not currently a problem.

The wireless ISPs’ reasoning is at the very least believable; radio is a tricky thing. The statement by wireline ISPs that congestion is not a problem is bizarre, firstly because it is well established that it is a problem, and that the problem is entirely of the ISPs’ creation. Secondly, if congestion is not a problem, the entire premise of the case for data caps disappears.

Some Experts Believe UBP Has Potential Drawbacks

• Some experts believe UBP may be unnecessary because the marginal costs of data delivery are very low, heavier users impose limited additional costs to ISPs.

As I was saying.

  • UBP may increase prices for some consumers and limit their Internet use, particularly for data-heavy content and applications.
  • As a result, UBP could limit innovation and development of data-heavy applications.

I have no idea who the experts GAO interviewed are, but they were clearly well-informed and thoughtful.

A thought-provoking example: there are currently no known uses for residential gigabit internet. None. Nobody has come up with any tangible immediate benefits of residential gigabit over, say, 100 Mb/s or even 50 Mb/s. With that in mind, why then are people so excited about Google Fiber, and the idea of gigabit fiber in general?

Hint: because they see opportunity. Access to gigabit internet removes roadblocks to innovations that have yet to even been dreamed of. But when they are dreamed of, internet speed (either download or upload*) will not be standing in the way. There is a reason why the most excited customers in Google Fiber’s Kansas City installation are small businesses and entrepreneurs.

*This is an important point. Google Fiber is symmetric: 1 Gbps down, 1 Gbps up. Most ISPs in America inherited DSL’s prioritization of download over upload, to the point where until very recently, even Verizon’s FiOS (fiber to the home!) had advertised upload speeds at only half their download speeds. (Verizon has since rectified the issue and now offer exclusively symmetric connections on FiOS. Their pricing is still complete and total BS though.) This asymmetry enforces the idea that an internet connection is a one-way thing: that data comes from the internet down the pipe to you, and not the other way around. But this need not be the case. Things like video chatting or even just uploading videos to YouTube (to leave out the obvious BitTorrent possibilities) rely on a fast uplink as well as downlink. If the innovations of the future internet make it more decentralized, as it should and must, ordinary people will require far better upload speeds than they have today.

Limited Choice Of Wireline ISPs Could Limit Some Potential Benefits of UBP

  • According to FCC, 54% of households are in census tracts that have more than two wireline ISPs with connection speed of at least 6 Mbps.
  • Some focus group participants said they would look to switch providers if faced with UBP but cited a lack of provider choice.
  • According to some experts we interviewed, this limited choice among wireline ISPs could provide less incentive for ISPs to offer more options in data plans to consumers.

And they even give a shout-out to the lack of ISP competition. As a sidenote, Dane Jasper has cited increased competition as his preferred solution to the net neutrality problem.

 

Copyright
This is a work of the U.S. government and is not subject to copyright protection in the United States.
The published product may be reproduced and distributed in its entirety without further permission
from GAO.

And even their copyright notice is awesome.