Even with two-factor authentication, desktop and mobile apps such as Twitter’s own TweetDeck and Twitter for iOS would most likely still store credentials and be able to hijack a Twitter account without the second factor.

How should you secure a high profile Twitter account while still allowing your social media person to do their job?

1. Use a randomly generated password for the Twitter account that’s not used anywhere else and is only written down on paper.
2. Use a dedicated computer for any use of Twitter that involves typing in the password or using twitter.com. Secure that computer in every way possible. Make sure it can’t access any website except twitter.com.
3. Only tweet through a custom, dedicated Twitter API app. Control access to that app carefully, with two-factor authentication, IP address restrictions, and, if it’s not too much hassle, two-person approval of every tweet. Encrypt the OAuth access tokens, the secret that authenticates the app with Twitter.
4. How do you initially authenticate the app so it can obtain access tokens? Do so manually, by writing down the access token key and secret and typing it into the server hosting the dedicated app.
5. It will still be possible to reset the password for the account through email. Guard the email account for the Twitter account very well, make sure the domain name cannot be hijacked, and that the email account is only ever accessed from the dedicated computer.
6. Regenerate the access tokens periodically.

The security of the Twitter account now relies on being able to secure the dedicated Twitter computer, as well as the server hosting the dedicated app. The computer should be easy to secure: it will hardly ever be used and can be powered off most of the time. The dedicated app server is more difficult to secure, but you can implement two factor authentication on it and do all kinds of other tricks without requiring cooperation from Twitter.

It is one of the oddest spending habits in Washington: This year, the government will spend at least $890,000 on service fees for bank accounts that have nothing in them. At last count, Uncle Sam has 13,712 such accounts, each with a balance of zero.

These are supposed to be closed. But nobody has done the paperwork yet.

So even now — as the sequester budget cuts have begun idling workers and frustrating travelers — the government is still required to pay $65, per year, per account, to keep these empty accounts on the books.

Kevin Drum points out that this is an infinitesimal amount of money compared to the overall size of the federal budget. However, as small as it is, even the $890,000 is probably overstated. The clue is here:

First, a federal agency gives out a grant. It doesn’t just write a check; it creates an account within a large, government-run depository. The grantee can draw money out from there.

The $65 per year fee is not paid to a private bank like the fee you might pay for a checking account. The account is with the Payment Management System, managed by an agency of the US Department of Health and Human Services. PMS charges grant-making agencies a fee, based on one of two fee plans. This means that the net cost to the government of zero balance accounts is not $65 a year, which is simply transferred from one agency to another, but the incremental cost for PMS to manage these dormant accounts. Since they have no activity, that figure is probably a lot less than $65, and may even be zero.

In his chapter on climate change, he makes the point that one reason climate change skepticism is so tenacious is that the statistical data about climate patterns really is a bit on the noisy and ambiguous side. The reason you can know that the skeptics are wrong isn’t so much because the data is so overwhelmingly persuasive, it’s that the data is overwhelmingly persuasive in light of the underlying science of how greenhouse gas emissions would cause climate change. Absent the causal theory about the greenhouse effect, simply looking at a chart of world temperatures and the correlation with CO2 emissions wouldn’t prove very much. The empirical data is important because it’s in line with the predictions of a persuasive theoretical account.

It’s worth recalling that global warming was first predicted at a time when the Earth was actually cooling:

If man-made dust is unimportant as a major cause of climate change, then a strong case can be made that the present cooling trend will, within a decade or so, give way to a pronounced warming induced by carbon dioxide.

Why “stupidly”? Before they were removed from this list, Kickstarter required Flash for video in desktop browser even though they could display videos in an HTML5 player in mobile browsers.

Under prevailing American tax law, a firm that wants to pay a dividend to its shareholders needs to pay with after tax dollars while a firm that wants to pay interest on a loan gets to pay with pre-tax dollars. Since the statutory corporate income tax is pretty high at 35 percent, this is a pretty big deal.

The difference in the tax treatment of interest and dividends is, of course, not 35 percentage points.

If a US corporation $100 of operating profit in interest, that money is not taxed at the corporate level. The debtholder, however, has to pay ordinary income tax on the money, at a top rate of 39.6% for individuals making more than $400,000 in 2013, so the debtholder gets $60.40.

If it is instead retained as profits and eventually paid out as a dividend, the profit is taxed at a top rate of 35% at the corporate level, so $65 is paid out. That income is in turn taxed at a top rate of 20%, so the stockholder ends up getting $52, for an overall tax rate of 48%.

This means there’s an 8.4 percentage point difference between debt and equity in how a dollar of profit is taxed from when it is made until an individual capitalist can receive it. That’s still a hefty subsidy for debt, but not as high as a naive reader of Matt’s post might think.

The figure is different for people in different tax brackets. It gets really complicated if debtholders and stockholders tend to be in different tax brackets or if you consider investors such as pension funds that are taxed in weird ways.

“GMT” is a natural choice because mean solar time at Greenwich was the basis for civil time for a majority of the history of widely coordinated timekeeping. GMT is still perfectly well defined, but it is no longer the basis for civil time. Instead, civil time in almost every country, whether by law or in practice, is based on Coordinated Universal Time (UTC), a time standard based on atomic clocks. If you think that’s too long, just say “Universal Time”, that’s close enough.

What’s worse, GMT is no longer precisely measured. Nobody can tell you what the time is in GMT right now with any precision, unless you happen to operate an astronomical observatory, while millions of clocks around the world, including Internet connected computers, GPS receivers and “atomic clock” wristwatches, can tell UTC.

Conclusion: Don’t say “GMT”. It makes you sound old, unscientific and imperialist. Say UTC, Universal Time, Coordinated Universal Time, or Zulu time.