> The solution proposed by the author: two headers and error correction code (ECC) in every file."

When there are two possibilities, which one do you chose? Three allows the software to have a vote among the headers, and ignore or correct the loser (assuming that there IS one, of course).

Also, keeping the headers in text, rather than using complicated encoding schemes to save space where it doesn't much matter, is probably a good idea, as well. Semantic sugar is your friend here.

Come on, ISO, where are you? We all need the best (or alternatively, least-worst) glidepath now. When I retired, the argument was all about proprietary formats for formatted text, and this and that. USians seemed to want to take the lead on everything and thereby 'offer' formal Secretariat (and steering). Now there's something worth doing - fixit, folks - and non-proprietary, pretty-please.

A while ago I was averaging the cost of ink per mL and it worked out to about 10 cents per mL. Considering an average 1 Litre bottle of wine is about $10, that makes ink 10 times the price of wine.

All very valid caveats. My core point (which you also addresses) was that using a working copy instead of an export of the files puts tons of extra copies of your files and subversion management data under you don't need for web pages.

I do love the atomicity of your method however, and the trick of exporting from a working copy made my day. Thanks!

Side note: I humbly suggest (as someone else mentioned elsewhere) you use export instead of checkout for the live deployments.

I need a "DNS entry" or "config file" or something similarly trivial replaced. It's none of your fucking business why I want it. Just fucking do it.

And changing such "trivial" things as a DNS entry or a config file will never cause any serious problems that the aforementioned IT janitor would have to clean up after, right?

Probably, but personally I have no clue. I'm not actually a Hulu user, or even care enough to do more than read the /. article.

Sorry about that. I was mostly just trying to explain the silliness of hiding the important bits behind "sekrit" code which you don't/can't actually keep secret.

General theory for you:

The web browser displays HTML, which is easily displayable and copyable.

It can also run javascript, a programming language that runs inside the web browser and can easily access the web page you are viewing.

In order to "hide" the HTML they are sending your web browser, they instead encode it and send you a page which only has the encoded version and some javascript.

When the browser runs the javascript, that script decodes the HTML and sticks it in the page.

Net results: using View Source in the web browser only shows the encoded HTML.

Reason why it's stupid: Anyone can run the javascript and decode it. The only people who couldn't get around this with the most trivial of effort wouldn't be using the HTML in ways Hulu disapproved of in the first place. It's kinda like using a European keyboard on a US computer in an attempt to stop people from using it. Anybody who can type can still see the letters on the keys. The people who only used the mouse to control the music player aren't going to be able to do weird stuff with your computer anyway.

Reviewing the Network Neutrality Squad mailing list (http://www.nnsquad.org/archives/nnsquad/maillist.html) supports the above statements archives also shows a certain propensity for vitrol.

While entertaining, and relevant, this analogy does miss a couple important pieces:

You don't give examples for the income of the drinkers. If the tenth guy makes a million dollars a month and the ninth makes a hundred dollars a month, it changes the perspective.

Also, you're ignoring the scale of costs. If the tenth guy has enough money to buy pretty much all the beer he wants, even AFTER he pays for everybody else's drinks, why is he worried about the bill?

Finally, the risk of having the tenth guy leave is always risk. But if he doesn't help pay the bill, it's the same risk. If the nine guys have to buy cheaper beer or share fewer beers, that's what they can afford. It doesn't change the core idea that the cost should be split equitably, rather than literally equally. That is, you measure the burden based on it's impact to a person, not on some abstract idea.

I let little old ladies on the bus first. Why? because it's harder for them to stand around than it is for me due to the relative ages.

If we were being equal, they'd have to wait like anybody else. I think it's better to be equitable and respect that standing around a little longer is cheap for me and expensive for them.

I suppose the argument that you shouldn't attack rich people for percentage tax breaks because you don't understand how percentages work is a good one, and like I said, I like the analogy, but I feel like it isn't accurate in some important ways.