emu

Treat this article with skepticism. I think the journalist has badly misjudged the current consensus in the field when they say:

"There are still a few die-hards who do not believe in hooperons. They suggest that if an ensemble of millisecond pulsars (dead stars that rotate hundreds of times a second) were buried in the Milky Way’s middle, that might do the trick."

The way I hear it, a population of new point sources is a better match for the Fermi data than dark matter --- the excess is "clumpy", indicating point sources, when you would expect a dark matter signal to be diffuse.

This data doesn't include non-salary compensation (e.g., stock grants). For tech companies, that can be substantial.

Standard caveat: I don't speak for my employer.

Using Bazel (aka. Blaze) every day is one of the things that has made me dread ever leaving Google. Fast, reproducible builds are amazing. Once you have used this tool, it is very hard to go back. Personally, I'm thrilled that it has been open sourced.

You are confusing two different things: gradual typing, and type inference.

Type inference infers a static type for every expression in a program. Hindley-Milner is such a type inference algorithm.

Gradual typing allows you to partition your program into typed and untyped parts. The untyped parts need not even be typable under your type system.

Type inference and gradual typing can be combined. See: Siek and Vachharajani: Gradual typing with unification-based inference http://dl.acm.org/citation.cfm?id=1408688

The idea of that paper is that you want to do type inference for the statically typed part of your program in a gradual type system.

The paper reference is: Siek & Taha: Gradual Typing for Functional Languages

http://ecee.colorado.edu/~siek/pubs/pubs/2006/siek06:_gradua...

(Of course, the "functional languages" bit isn't critical here --- the paper defines gradual typing for a lambda calculus, which is the usual vehicle for explaining type systems.)

Basically, you want to partition the program into parts with and without known types. If you know the types, they must be equal. If a type is unknown, it is "equal" (consistent) with anything. The conclusion of the paper is that you need a different, symmetric notion of consistency, that is different from subtyping.

Making Any the root of the type hierarchy doesn't work. You explicitly want to allow things like implicitly passing an Any as an argument where, say, an Int is expected, which would be a down-cast. You don't want to allow implicit down-casts all over the place, because that causes the type hierarchy to collapse --- if you allow implicit down casts, you can always up cast to Any, and then immediately down cast to any other type.

Why? Even if you can eavesdrop on the NFC communication, how does intercepting the challenge/response help you?

Sure they do: https://support.google.com/docs/answer/2375012?hl=en

I might have missed it, but I couldn't find any licensing information for most of these data sets, which troubles me. Personally, I'm hesitant to download or work with much of this data for that reason.

One problem with using an SSN as a identifier is that the number exists for the administration of the Social Security system, so only people who pay into or will receive money from the system are eligible for a number.

Not everyone in the US is eligible for an SSN (e.g., some non-residents living in the US). When I first came to the US as a student, I had difficulty doing things like signing up for a credit card or a cellphone for this reason.

There are other numbers that systems sometimes but do not consistently accept, for instance an ITIN, which is the tax ID number the IRS will issue you if you aren't eligible for a SSN.

If you ask me, the Hangouts actually is a huge improvement over the old Google Talk.

I never sign into the old Talk on my gmail account because I find it very very distracting. The presumption of Talk (or other IM clients) is that if I am online, then I am willing to be interrupted to answer your message immediately. That's absolutely not the case --- if I'm working, I'll answer your message later.

The Hangouts (SMS-like) model is much more useful to me --- you can always feel free send me a message, but I will answer it when I see fit. Hangouts gives a hint as to whether I'm likely to read your message soon, but there's no promise of a synchronous response.

This study appears to ignore non-wage compensation (e.g. stock grants). I know that for me, as a H1B worker who graduated from a US graduate program and working for one of the large companies named, that stock based compensation is a substantial fraction of my total pay.

This is a problem with the blanket term "Intellectual Property". Intellectual property consists of many things, e.g. copyrights, patents, and trademarks.

You can be in favor of some without being in favor of all.

I mostly wanted some better set and map container implementations (hashtables, binary search trees, tries); the built-in map can't be applied to all types, and is quite limited in how configurable it is. Given that we were going to be using these data structures everywhere, we decided Go was a bad choice for our project.

My guess is that better containers aren't in the library because of the lack of generics. I really hope the Go folks get around to adding generics soon; it's a beautiful language in most other respects. It would definitely be one of my first choices for a more systems-y project.

Unfortunately, it seems to be lacking many useful data structures. This bites you hard once you hit the limitations of the built-ins.

I have to agree that the lack of generics is one of the biggest holes in Go.

Common container data structures are one example. You can't write, say, a generic binary search tree in Go, unless you wrap all your data in an interface. While possible, this means you have to downcast all over the place.

The built-in map, while useful, is limited in how far it goes.

Maybe I'm just spoiled by the STL and the Java collections libraries.

It's a little disingenuous for the IE team to "discover" this just now. I'm pretty sure Google has been doing this for years, and it's well-known. (I certainly talked about it as part of a wider discussion about P3P policies with colleagues a year or so ago, and this isn't even my area of expertise.)

I also don't much mind what companies do with tracking cookies --- I recommend using the Vanilla Cookie extension to Chrome to create a whitelist of persistent cookies. It rather nicely avoids the problem.

We're off the original topic now, but I've never understood why the US doesn't enact a dividend imputation scheme, which nicely solves the double taxation problem on dividend income by giving the recipient of a dividend a credit for the corporate tax paid on that dividend. Australia has such a system.

Back up your email! Just because your data is in the cloud doesn't mean you shouldn't make a backup.

I just run getmail which retrieves mail via Gmail's IMAP interface every now and then. If Gmail went away tomorrow, I wouldn't be _that_ sad.

This article is link-bait, but I'll bite.

Simply put, Gmail provides a compelling user experience:

* fast and accurate search. This is key, and I don't know of a single desktop email client that even comes close to providing a decent search feature. Decent here means "answer full-text queries accurately within a second or so". I didn't realize how useful this was until switching to Gmail; now I have it, I can't switch back. I don't even bother organizing my email more than a minimal amount any more, I just search for things when I need them.

* remote access is also key. I regularly access the internet from at least 4 different devices, on at least 3 different operating systems. Sometimes I do so while traveling. Gmail makes this convenient --- all I need is a web browser. IMAP need not apply --- I found IMAP appallingly slow last time I tried it.

Perhaps we need a humor flag on web pages, Twitter. By setting the humor bit you could remove all ambiguity as to whether you're making a joke or a threat.

[this.humor := true]

I'd also recommend a good game of arXiv vs. snarXiv: http://snarxiv.org/vs-arxiv/

As a non high-energy physicist, it's surprisingly hard! I usually do _worse_ than random chance, sometimes substantially so.

The author also has perhaps my favorite definition of a CFG: "The snarXiv is based on a con­text free gram­mar (CFG) — basically a set of rules for computer-generated mad libs."

Well, unfortunately most interesting static analysis problems are undecidable. There can be no "general" solution.

What we have in practice are approximations and heuristics; over time we will develop better approximations for the kinds of code people write in practice. Unfortunately fragility of analysis will always be a problem; if you change your code just a little bit then reasoning may fail.

As someone who has spent years doing research on shape analysis my personal belief is that the dream of a fully automated "find my bugs" static analysis is unrealistic. Some of the problems you must solve to analyze the heap, say, are very hard indeed.

We need to think more about the interactions between language design and verification, rather than hoping that people will build better tools to analyze existing code. Strong typing (e.g., the Hindley-Milner type system) is one example of a sweet spot that demonstrates how language design (type annotation, type declarations) interacts with tools (type inference, type checking). Type systems are some of the most widely used bug finding tools today. Trying to build verification tools without considering language design is always going to be difficult.

I'll make a deal with you. If you don't put flash or animations in your ads then I won't try to block them.

For Verizon and Google: let me remove bloatware from your phones, or I won't buy them. Pity; I was actually going to look into the Galaxy Nexus whenever it comes out.

Seems fair all round to me.

There's certainly no financial incentive for a Ph.D. graduate to remain in academia. That said, there is no financial incentive to do a Ph.D. in the first place. My understanding is that the opportunity cost (salary you didn't earn while doing your Ph.D.) is larger than the total expected extra earnings over your lifetime. My understanding is that by contrast a Masters degree is a good investment.

There are other reasons to stay in academia --- mostly to do with lifestyle. You have a lot more freedom to work on whatever interests you than in many other career paths.

CS isn't a field where a postdoc is a mandatory step on the way to a professorship (unlike, say, physics). Postdocs in CS are not too common, so it's hard to generalize.

Since this is a copy of an email sent to CS students at Stanford, I'm not sure it's intended to be public information. One of the staff members in the CS department emails graduates every year to survey salary data and to ask about how useful the recruiting resources provided by the department are. I'd imagine the numbers accurately reflect the responses received, although as others have mentioned there may be biases in who chose to respond.

Install the GNOME 3 packages using apt (gnome-shell and gnome-panel should get you most of what you need) and run them in fallback mode. The GNOME 3 fallback mode is pretty close to the GNOME 2 experience, although it is missing a couple of features that some people may want (e.g., panel applets). I've been using Oneiric this way for months and I'm pretty happy with it.

It might not kill SMS itself, but it has a decent chance of killing the need for an SMS plan. If most of your text messages are free anyway, then buying a text plan starts to become a poor idea.

I haven't had a text plan on AT&T for many months, mainly because I use Beluga to send texts to my friends. iMessage is similar but much more seamless. I have saved hundreds of dollars by paying for the few text messages I still send/receive as I go.

I'm Australian, age 29, in excellent health, and living in the US. I currently have quite good health insurance. However, if I found myself in a situation in the US where I did not have health insurance, I would try to move back home as soon as I could.

To me, it seems laughably risky that anyone would voluntarily not have health insurance. (I realize there are many reasons why people might involuntarily not have insurance.) Illness can happen at any time to otherwise healthy people with no warning, and if that happens you are royally screwed if you don't have insurance.

Someone I know was diagnosed a month or so ago with stage 4 cancer. They had no symptoms, and it was only found by a diagnostic test administered by a cautious doctor. They now need hundreds of thousands of dollars worth of treatment, although fortunately they are insured. You simply have no way of knowing when your luck will turn.

I also just tried to log in about 6-7 times in a row (clicking the "login" link on the front page, reloading the front page in between attempts), and I repeatedly received the "expired link" page.

It also reliably happens clicking the "next page" link on the bottom of the front page; by the time I'm done reading the front page the next page link usually expires.

Please fix?