zardeh

That makes me even more curious:

what is the market share of adwords things (my understanding this is mostly going to be people offering services) that is impacted by AMP? An ecommerce site won't support AMP so won't be affected, and are news sites and blogs going to be using adwords to advertise their news and blog offerings that much?

E: (and then does it even matter if you're ads are still shown on your AMPed site so you get paid per hit?)

so you still have to pay for a click even if the user never really visited your website

Don't you mean so you get paid for a click even if the user never visited your website?

Ah, I entirely misread that as "probably a better comparison is" instead of "than".

Whoops.

Why? In this case, the majority of these employees are likely not in tech roles, but in various parts of the fulfillment pipeline. That makes them more similar to FedEx or Walmart.

Edit: Disregard this, I can't read.

It's not a tangent. Ms. Yates apparently feels the president does not have the legal authority to make orders of this nature when President Trump is in office, but that he does have the authority when President Obama is in office.

There's a very simple answer to this:

She did not believe that the Obama EO was unconstitutional, whereas she believed that this one was. Then, everything makes sense. You can rant about how "there's no credible argument", but I'm inclined to believe an Attourney General over someone on an internet forum with regards to the law.

So tell me this: what should she have done if she believed the EO, as applied, would be unconstitutional?

Edit: also the judiciary is moving swiftly, in the past two days, the EO has been practically gutted in the courts.

but the affirmative action one is pretty damning

The supreme court agrees that affirmative action is constitutional (within certain bounds), see the recent Fisher v. UT case, or Bakke for the original example.

Last year, the vetting process somehow managed to grant visas to 9500 terror linked individuals, all of whom are missing. When the visas were revoked, there was no way to locate them. (Source: House oversight committee hearings & Mainstream media ~ 2015 / 2016)

No, since 2001, 9500 "terror linked individuals" were given visas. Unfortunately, I can't find any explanation of what "terror linked" means in this context. Each year, the US grants approximately 8-9 million visas. So, in 15 years, across over 100 million issued visas, approximately 10,000 may have been given to individuals who were "terror linked", where that could mean "confirmed terrorist", or it could mean "their uncle once went to a market that we believe is a known terrorist hotspot". Again, there's no explanation, anywhere, of what "terror linked individual" means in this context.

Please stop editorializing.

That's interesting. I totally see this as a valid argument (and I don't disagree that this might be politically motivated on the judge's part). However, it seems to me that

any clear conviction that the law was unconstitutional

is the wrong way of going about it. I would think you'd want to be pretty darn sure that the actions you were carrying out were constitutional before doing them. That is, I'd much rather people not enforce a law that might be unconstitutional, but isn't, than enforce a law that is unconstitutional. Err on the side of caution.

And they are allowed to be vilified for their ambivalence. Saying "I don't care" doesn't absolve you of any guilt.

Well, cookiecutter came first, has a huge library of supported templates, and works across multiple languages.

That story mentions Bishops and priests who are very likely to not be Syrian Citizens, and instead be capable of just flying back to their home countries.

I'm by no means saying that every Christian will be more capable of leaving, or not be a citizen. But if 30% of Christians are citizens elsewhere, and only 3% of Muslims, then you have 70% of 10% of the poopulation as potential Christian refugees, and 97% of 90% of the population as Muslim refugees, which would make 7.4% of asylum seekers Christian, as opposed to 10% of the population.

If (and that is a potentially big if) Trump continues the Obama admin plan of responding to all petitions that have >100K signatures, it will be interesting to see what the response to 'hey you should let this young female student who was already here back in, since she poses no danger to the US'.

But would they be more likely to be refugees?

That is, I'd expect that a Christian in the middle east would be more likely to have citizenship or family in a western nation than your average Syrian.

Now it could be, but 'they are being specifically targeted' doesn't immediately imply to me that 'they should be a disproportionately high number of refugees'.

That story mentions Bishops and priests who are very likely to not be Syrian Citizens, and instead be capable of just flying back to their home countries.

You may be, but trump didn't say anything about Muslim immigration, he said a and again, I'm quoting his website:

a total and complete shutdown of Muslims entering the United States

A shutdown of entry, to use your own words.

When it comes to constitutionality, SCOTUS precedent is not good or bad, it's simply is reality.

It depends, there are SCOTUS rulings that (most) people consider "bad". These create precedent.

Him being quoted by Breitbart means nothing. I mean Obama has been quoted by them. Clinton has been. Pretty much every single prominent Democrat politician has been. The only thing that is exposed by saying "breitbart" is that you disagree with the right.

No, I was pointing out that I had already addressed this specific example when I stated that "The only people I can find who believe it might be constitutional are non-constitutional laywers quoted by Breitbart." Banzhaf is the non-constitutional lawyer who was quoted by Breitbart. Now, you're quite correct that I don't find Breitbart to be a reliable source of news (although construing that to 'I disagree with the right' is a bit of gymnastics), but its also orthogonal to my point.

I'm saying that anyone who says that it's unconstitutional on it's face is full of it.

That also very much depends. If we're talking " a total and complete shutdown of Muslims entering the United States", that is on its face unconstitutional, since it includes American Citizens. If we're talking a registry, it probably isn't. If we're talking about refugees who are Muslim, then there's the grey area. But if you take him at the words he used, it is unconstitutional, and most of the blogs have caveats that say something along the lines of "a ban on American Citizen Muslim's returning from abroad is unconstitutional on its face, so we'll ignore that and talk about refugees and immigrants"

1. "It would raise complicated questions, but might not be unconstitutional, if only because of bad precedent"

2. Isn't referring to a ban on Muslim immigration, but on a registry, which is a whole different beast

3. Banzhaf isn't a constitutional Lawyer, he's the one quoted by breitbart, btw, and what he glosses over is the very relevant 'detrimental to the united states' aspect of the Plenary Powers doctrine. It is much easier to make an argument that we should suspend immigration from a state we are at war with than from a religion, since it is practically impossible to make the argument that Muslims are more detrimental to the united states than Christians or Atheists.

and your fourth article concludes by saying

It does seem reasonably clear that, if a proper challenger could be found, the courts very likely would be open to hear their claim. And it would not be a sure thing that they would lose in that forum.

No it doesn't, or at least not as far as I can read. A summary of that email is

Here is a letter I've drafted outlining our position on trade. This draft assumes that Clinton will ultimately support the TPP and TPA, but we may change the letter dramatically if Clinton does not end up supporting the TPP and TPA (ostensibly because the final agreement doesn't cause a net gain in jobs).

I can find literally 0 constitutional lawyers who agree with this statement. The only people I can find who believe it might be constitutional are non-constitutional laywers quoted by Breitbart. Every other article I looked at (and it was >20) had multiple people stating that a religious test would be unconstitutional, and that the only leg they might have to stand on was that an immigrant not allowed into the US wouldn't have any way to sue the US government to raise the issue in the court system.

No, he said (and this is a direct quote):

Donald J. Trump is calling for a total and complete shutdown of Muslims entering the United States until our country's representatives can figure out what is going on,

He later walked this claim back, but the statement was originally a ban on immigration by anyone of a given religion, not anyone from certain territories. Those claims were later walked back by other republicans, but the original claim made by Trump was for a ban on all Muslim immigration (which is on its face unconstitutional).

Many Americans voted for a candidate who has catered racist supporters and put forth a variety of xenophobic policies and made even more statements in that regard.

As a simple example, our current president campaigned on a wall to keep out "rapist" Mexicans and to prevent anyone who was Muslim from entering the country because they were terrorists. Those are both explicitly xenophobic statements. They "showing a dislike of or prejudice against people from other countries". Still, about 60 million people voted for him, which means that to those 60 million, either his statements were a good thing, or they were minor enough downsides as to be ignored.

Yeah I learned to return values and accept references very quickly, had to entirely refactor my learning project a day or so into it, but that was for the best.

For the Clinton server, I never stated that work email being wiped was intentional.

No, but your graph shows that of the 30K emails that were deleted after the start of the investigation (with bleachbit, as you say), 2800 were work related. The FBI does not claim this, nor do your sources. It is as far as we know, a false claim.

It's a case of Simpson's paradox where the subpopulation does not conform to the aggregated whole.

Its not, you are never combining groups. Its just a bad application of percentages (and perhaps a worse visualization).

The Hilary email server one to start.

Your (I assume your, since your username matches the site) visualization shows a tree, and reads as there being 31800 wiped emails, of which 12200 were later found, and of those 2800 were work related. I'll quote FactCheck.org, one of the sources you list on your site:

There is no evidence to date that work-related emails were intentionally deleted. ... In his July 5 press conference, Comey said “like many e-mail users, Secretary Clinton periodically deleted e-mails or e-mails were purged from the system when devices were changed.”

Your visualization states that the FBI found emails that Clinton wiped. This isn't the case (or at least, the FBI doesn't believe this is the case).

The racial killing one is misleading, although I'll grant you not explicitly false, just heavily misleading. Specifically, your use of percentages is badly phrased and could be seen to push a misleading narrative.

If I hover over raw, I see "6546 blacks killed by blacks". This is a true statement. If I hover over percentages, I see "92.9% of blacks killed by blacks". This is obviously not true, since it would imply that every black person was killed this past year, and 93% of them by other, now dead, black people, a ridiculous assertion. Instead, what you should state is "Of the ~7000 black people killed last year, 92.9% were killed by other black people". That's the true statement.

I don't claim this is false though, because it was obviously just a formatting mistake, which I hope you'll amend now that I've informed you of it.

It seems to me however, that you are trying to include something percentage based to imply that you're normalizing by population. That is, according to the census bureau, there are about 39 million black Americans, and about 224 million white Americans (I rounded both to the nearest million, which was up in both cases).

That means that a black American had a 258/39,000,000 chance of being killed by police, while a white american had a 666/224,000,000 chance.

Or in other words, 6.6 in every million black Americans was killed by police, and 2.97 in every million white Americans was killed by the police.

To put it another way, you're normalizing within groups instead of across groups, and this leads to a practically useless set of numbers that you can throw at people, but don't have any practical value. So even if you do change your formatting issue, it would be better if you showed the across group normalization, instead of what you currently have, which isn't a valuable piece of data.

Edit: For someone who is quick to claim others are making mistakes "its all about the trendline" and "I'm tired of people making the current year being the hottest into something its not", I would expect better work honestly.

So to be fair, I haven't used a lot of C++, but what I have has been unlikeable.

I'm a pythonist most of the time, although I've done a not insignficant amount of work in C, but I like high level languages more, and C++ feels like the worst of all the worlds. It is complex and bloated (templates) and magical and all kinds of things. I think I summed it up one day by saying that, as a newbie to C++ my frustrations with the language were summed up by a 0 argument object constructor having no parenthesis if it was stack allocated, but heap allocation or arguments require parens, because the 0 arg constructor would conflict with a function prototype if it had parens.

There's so much bloat and cruft and oldness that the language has baked in inconsistencies and gotchas and so much stuff (which of the 12 different kinds of pointers should I use) that it is horrible to learn.

Rust on the other hand has complex semantics, but they are well defined, sensible, and logical. Sure I sometimes want to smack the borrow checker, but the reason why isn't "someone made it this way before I was born and now I'm stuck with the results".

(I implore anyone who looks at this to take a look at their other visualizations, all of them push a narrative, and at least three misrepresent information from their sources or explicitly state false information)

Interesting,I absolutely despise C++, and I find rust to be incredibly nice (though admittedly, I'm a neophyte, I've barely had to use lifetimes, so I may be unaware of my own lack of knowledge).

Unplug your phone and play (or record) videos for a few hours. See where the battery is.

You can find the syllabus for Isbell's class and follow along

To be fair, most of Isbell's course (lectures) is also available on Udacity.

If you have to teach the material to yourself, how is your experience improved by being in the class?

There are a couple advantages. One of the most obvious is the lower latency of responses when you have confusion or misunderstanding. In a lecture, you can ask a question and get an answer almost immediately. This is most useful (imo) with algorithms and mathematical concepts, because you can ask, and lecturers are often quick to provide insight, into the interrelationships between algorithms (both in Machine learning and in a more theoretical sense like computability). There are topics that come up a lot, and being able to have instant feedback on those connections allows you to spend less time misunderstanding than not.

That alone is a fairly weak justification, I think the stronger one is feedback in general. Watching lectures only gets you so far. With implementation of algorithms, often your feedback is testable correctness (although my experience in DS&A suggests that most people are capable of constructing incredibly incorrect models for things that perform well on some input, and even on decent autograders), but with things like machine learning algs and intuition about those algorithms, you can't get that. So the feedback that yes, your understanding is correct (even if that feedback is slow) is invaluable. In that regard I think online courses and MOOCs can be good, but MOOCs that don't provide feedback aren't as valuable. I've attended a lot of lectures, and I've ignored a lot of lectures. Listening to someone say something does not mean one has learned it.

I'd also note that, if I recall, the way that Isbell approaches teaching the material, vs. the way the textbook does are very different. Textbooks are (often) references. They provide information on what something is and how it works theoretically, but very often lecturers are able to provide the kinds of things that aren't (and shouldn't?) be in textbooks.

If I'm reading a textbook, its very likely that I want to know how to implement an algorithm, so I care that the algorithm for simulated annealing says that you jump with probability e^(D/T) > Rand[0,1]. Whereas in a lecture, I'm likely much more interested in the idea that simulated annealing is conceptually very similar to throwing a ping-pong ball into a large complex, convex plastic surface and seeing where it lands.

[disclosure: GT CS alum]

Andrew Ng's Coursera class on Machine Learning was the pedagogical highlight of my time at GT, and I did it on my own initiative (and it's free).

This is something I wanted to highlight from your post. I don't think this is surprising, nor do I think it is reasonable to state that a course (or a degree program) is poor quality because it didn't meet the standards of Ng's ML course. That is an exceedingly high bar.

Its something I noticed, because I am a recent grad, so while I was in my mid-level courses, and had recently taken Tech's intro CS course, I was able to watch (Harvard's) CS50 and other courses. But on the other hand, I've seen some very bad online courses. The successful and large online courses are successful and large specifically because they are head and shoulders better than the rest. And there are a lot of decent online courses, so to measure against what are some of the absolute best online courses is to measure against courses that have more resources, more planning, and more feedback than most.

(as an aside, they also have more incentive to be good, but that's a bit tangential to the point that they also have more opportunity to be good).

You type in "Machine Learning" on coursera and you get over 1000 results (not all of which are relevant, but assuming even 10% are), its little wonder that one or two are going to be better than the even the best courses that you'll take during a bachelors or masters, because Coursera offers more Machine Learning courses than most people will take in their Bachelors or Masters.

Combine that with these courses coming prefiltered (you've heard of the Stanford Course, but what about "Applied Text Mining in Python" from UMichigan, which for all I know might be great, but it doesn't come with the hundreds of recommendations that the Ng course does, so I don't know that it will be great) and you have a really great recipe for a bias against the in person courses.

However, I felt like you would only get something out of his lectures if you already knew what you were talking about.

I disagree (having taken the course as an undergraduate and it being my first major exposure to machine learning). Certainly if all you do is attend the lectures, you're going to miss some background knowledge, but that is true of most (if not all) university courses. You're supposed to devote 2-3 hours of outside work for each hour of lecture. Meaning 6-9 hours of studying per week outside of those lectures.

Some of this is doing the projects, although some of it is personal investigation.

There are failings of his course (one of the biggest at this point is that it doesn't do any work with the state of the art now), but I think that the fact that his course caters toward people who are self-driven is not a failing.

The best way to look at what the goal of the course is is by looking at his exams. If they weren't different than you took them, they were intentionally too difficult for the allotted time, leading to low averages and incomplete work by the majority of students.

However, the course allows motivated students to make connections between concepts, with the help of the professor and the coursework. Having someone "leading you" down the right path is very helpful, much moreso than a textbook alone.

I really do think that there is one exam question that sums up Isbell's course perfectly: its the one where you are asked to compare and contrast 4-5 aspects of 4 randomized optimization algorithms (RHC, GA, SA, and MIMIC) and explain situations where you'd use each and why.

The course's goal is to lead to a strong intuition for the algorithms covered (sadly at the partial expense of a theoretical understanding), not everyone puts in the work to develop that understanding, but that's not a failure of the course, necessarily.