hoddez

I am not an immigration lawyer, but this is my understanding: The H-1B is similar to the TN but each have their pros / cons: TN is much cheaper (~$50 + lawyer fees) and there is no cap / application deadline. You apply at the border so there is also virtually no waiting time assuming you get it.

The downside of the TN is that it is a "temporary" permit. Technically it's not actually a visa. It simply allows you to work and live in the US for a "temporary" amount of time. The temporary part is very gray in this case, since it is up to the discretion of the border guard to determine if you are using it in a temporary sense. Similar to the H-1B it can last for up to 3 years, and can be renewed after that.

The other big difference is that dual-intent is not allowed on the TN permit. In other words, you are not supposed to apply for a green card while you are on it. The H-1B is dual intent and therefore that's what you want if you want to get a greencard.

A pretty typical path for Canadians who start working in the US and decide to stay is TN -> H-1B -> Green Card. However, it is not easier for Canadians to get an H-1B. They must participate in the lottery like everybody else.

No sorry, the TN is only available to Canadians and Mexicans. It's based on NAFTA. https://travel.state.gov/content/visas/en/employment/nafta.h...

There is at least one caveat with multi-armed bandit testing. It assumes that the site/app remains constant over the entire experiment. This is often not the case or feasible, especially for websites with large teams deploying constantly.

When your site is constantly changing in other ways, dynamically changing odds can cause a skew because you could give more of A than B during a dependent change, so you have to normalize for that somehow. A/B testing doesn't have this issue because the odds are constant over time.

This is great stuff. Thank you! I'm actually about to release an Ear Training app for iOS and I would love it if you'd take a look and tell me what you thought. if you would e-mail me at [email protected] I'll tell you more

Wouldn't that mean you can't use any cloud data services with any company? Or even cloud hosting? What kind of customers forbid this?

Wow I didn't think it was supported on Android Chrome yet or any mobile browsers. That's awesome!

Yeah thats a good point. I don't like the alert either but I'm not the best designer/front end dev so i didn't bother with any custom ui for now.

nope, sorry but safari does not yet support the web audio api features necessary for this. I think they might be on the way.

Whoa cool thanks for sharing that. This could have been very helpful, and still might be. I'll have to check it out and see what they did differently and how the performance compares.

"I tuned my guitar succesfully with it" Cool! That's what I like to hear.

Thanks for the review and feedback. The performance delay may be due to the high volume right now because of the posting.

Also this is a very very simple algorithm that I will likely replace. As stated below, the cool thing (I think) about this is that its all javascript. All other online tuners require flash to use the microphone. This is still somewhat of a proof of concept/prototype and I hope to improve.

I tested this on a macbook pro with the built-in microphone and mostly Chrome. It's great to know that it works on a completely different platform.

Silently dropped the storage drastically in the lower plans. Knew the 1TB at $50 a month was too good to last forever. Now for that much storage you need to pay $2000

agreed. was thinking the same thing but you beat me to it. anybody who has ever evaluated an A/B test would look at this and immediately question the significance/validity of the result. if only somebody would actually back up their claim with ALL the data for once (i.e. variance and specific sample sizes)

1 is definitely an issue, but has a solution. I posted an article about this a while back with a simple example of the problem and a possible solution. The example is a staged rollout but it illustrates the same point. If you change proportions of tests over time as there are independent changes, you can get skewed results: http://blog.avidlifemedia.com/2011/12/23/advanced-ab-testing...