zader

Good catch. I figured HN would reject it if it had already been posted. Was surprised when it went through, actually.

Estonia down.

What are the "big issues" that would be caused by forcing SSL? It sounds like visitors from Iran (and other countries that block SSL sites) might be unable to access HN. What else?

(To clarify, I'm interested in the general topic of whether it's a good idea to have all-SSL websites.)

Site appears to be down. The Google cache is here:

http://webcache.googleusercontent.com/search?q=cache:tAI5zmC...

But the meat of his presentation appears to be this video, available on Vimeo:

http://vimeo.com/45485034

For anyone who wants to download the Nightly build for Firefox 15:

Windows version http://www.mozilla.org/en-US/products/download.html?product=...

Mac version http://www.mozilla.org/en-US/products/download.html?product=...

OSX version:

http://www.mozilla.org/en-US/products/download.html?product=...

Helpfully, the final paragraph of the article seems to contain its takeaway message:

"So, comparing A/B testing and multi-armed bandit algorithms head to head is wrong because they are clearly meant for different purposes. A/B testing is meant for strict experiments where focus is on statistical significance, whereas multi-armed bandit algorithms are meant for continuous optimization where focus is on maintaining higher average conversion rate."

Like any business owner, at the end of the day I care more about conversion rates than about statistical significance.

I'm not a scientist trying to advance the sum of humanity's knowledge. I'm a business owner trying to find the shortest path between what my customers need and what I can profitably offer them.

In a way, statistical significance strikes me as a bit of a fool's errand, because significant results in one context may not be generalizable to another, which means even if we know for a near certainty what worked best, it's hard to apply that knowledge reliably in the future.

Of course, with MAB we could still wait for statistical significance if we want it, before turning off variations that are performing worse. And we can certainly still try to draw conceptually useful conclusions by designing our test variations in ways that facilitate easy comparison.

But with MAB and Myna it sounds like we can pretty well count on higher conversion rates at the end of the day, as well, and that counts for a lot in a business context.

I'm grateful to the VWO team for writing up their analysis and findings, and being so frank about the relative advantages of A/B and MAB. Their summary above tells me what I need to know.

What do you think of Myna, in these respects? Does it suffer from the same disadvantages as other bandit optimization approaches?

http://mynaweb.com/docs/

Interesting story.

For anyone else wondering what a "poison pill" is, in this context:

http://www.investopedia.com/terms/p/poisonpill.asp

Yes, Nicaragua would be the protectorate, but the terms of the contract for the economic zone ultimately can only be enforced by someone with a large enough military to impose justice of the last resort -- or at least have the opportunity to do so. Cities can't do that. Legal clout comes from military cloud. Cities have no military clout. Therefore, they have no legal clout, except as provided for by some country.

Countries have military forces that can ensure the settlement of international disputes.

Yes, I'm interested. This article was relevant to some community building efforts I'm involved with, and I'd love to read more veteran insights on the subject.

"So, if you are a Whole Foods shopper, please consider honoring the boycott, at least for a short period of time. The impact will be very evident, and almost immediate."

I have a different idea.

If you are a liberal who shops at Whole Foods, and you read John Mackey's article with his ideas about health care reform, why don't you think about whether the solutions he's proposing -- and has successfully implemented among thousands of happy, healthy Whole Foods employees -- might be worth some serious consideration.

Instead of, you know, rejecting his free market ideas out-of-hand with insulting ad hominem attacks and threats of a hard-line ideological boycott.

Tolerance, civility, honest debate, etc.

Just sayin'.

Because Whole Foods has been amazingly successful as a company in recent years. And they do have thousands of employees. And I'm fairly sure that those employees like being able to keep the majority of their health care dollars, instead of having them thrown away on overly comprehensive health insurance policies they don't want or need.

The article assumes that any cancellations must be unjust, and then uses math to show that there are more cancellations than one might suspect. But if the cancellations aren't unjust in the first place, then the article is all smoke and no fire.

News flash - If you include fraudulent information on your insurance application, your insurance company may cancel your coverage when you need it most. Of course, if you filled out your application in good faith and they try to cancel it for frivolous reasons, you can sue them in court.

What exactly is the problem here?

Most of the urgency in his article comes from comparing fraudulent insurance applications to game show contestants and underage gamblers. Using these kinds of analogies obscures the more fundamental fact that insurance companies can only legally cancel your policy when you have supplied demonstrably fraudulent information.

That hardly seems like a scandal, or anything that warrants additional government regulation in what is already an incredibly heavily regulated industry.