Enemy #1 in a large data center is power. When he mentioned the entire rack pulls 1,000W for the density suspected in this thread (~500TB to ~1PB), that piqued my interest. Even if you can't push 2-3PB per rack, the top stat you're crunching forever and always is watt-per-gigabyte. Space is important but if it breaks into the top spot you've done something else wrong along the way.

Enemy #27 for cold(ish) storage might be, say, pressure per sq. in. on your floor. I'd be interested to see a weight comparison of a fully-loaded LTO/cartridge rack and this design.

Perhaps not what you were looking for in an answer, but I'm interested to look over the schematics.

The FISA court of 11 judges is appointed by the Chief Justice of the United States. [2] It absolutely is an Article III court. The ACLU has even acquired its rules [3].

The fact this is the top comment without citation is troubling. Please don't encourage these sort of claims without at least a cursory glance at the documents governing these institutions.

[1] http://www.law.cornell.edu/constitution/articleiii [2] http://www.law.cornell.edu/uscode/text/50/1803 [3] http://www.aclu.org/patriot_foia/2003/court_rules.pdf

I don't think membership consisting of judges who would otherwise preside over Article III courts necessarily makes FISC an Article III court, and judges appointed by the Chief Justice actually suggest the opposite.

There's an article at the Yale Law Review that criticizes FISC for not being a regular Article III court.

It is very easy to get caught up in examining the mechanism itself as a way of stamping out the abuse of the mechanism. As much as I dislike what the court represents, I can't fully qualify its dissolution.

In 50 USC § 1803 (b) it specifically states the Supreme Court has jurisdiction over the decisions. The Title itself is designed for expediency and proximity - something its "parent" court isn't designed for.

I don't think the evidence we have is reliable enough to call someone "wildly misinformed" or "intentionally misleading", though.

Perhaps not. However, my original challenge to rayiner's comment has produced far more meaningful and mostly-cited comments. Could it have been less inflammatory? Absolutely. Would it have produced this discussion if it had been? Uncertain.

The original comment did not add any value to the post. I have never read Orrin Kerr's writings and both of you citing him caused me to do so.

Perhaps citing Kerr in the original comment, as he has well regarded opinions on the subject, would have been additive as I assume rayiner intended to be.

After all, if we don't challenge for citation and additional information, what are we doing here?

That said, I don't think Glidden v. Zdanok unarguably supports the idea that FISC is an Article III court (though Kerr doesn't claim it does), just because it is composed of article III judges. Glidden is actually about the opposite question: whether the judges were article III judges based on whether the Court of Claims and the Court of Customs and Patent Appeals were article III courts. And most of the analysis of Glidden focuses on things like the courts' ability to hear justiciable cases and controversies and to exercise the Article III judicial power. Under these criteria, you cannot call the FISC an Article III court because it can't even hear any cases and controversies, nor can it exercise the essence of article III judicial power (the power to render binding, final judgments with regards to matters affecting life, liberty, and property).

So I would personally call the FISC an Article I court staffed by Article III judges sitting by designation. But I worry it would be an idiosyncratic use of the term. In any case, whatever you call it, it is clearly not empowered to exercise even a substantial amount of the Article III power, which is the substance of my point. It can't put you in jail or take your property.

Obama said yesterday,

"On this telephone program, you've got a federal court with independent federal judges overseeing the entire program," the president continued. "And you've got Congress overseeing the program, not just the intelligence committee and not just the judiciary committee, but all of Congress had available to it before the last reauthorization exactly how this program works."

This is slippery and brings to light a problem with FISC. Its judges are independent, but the process is not. The FISA process (necessarily!) begins with the premise that NSA has a Constitutionally unchecked authority to conduct foreign surveillance.

I think this would be fine, were it not for the fact that NSA seems to be transitioning from a role that was principally involved in nation-state intelligence to a role that is inextricably bound up with law enforcement. You can see the problem when Mueller starts talking about what it would take for FBI people to dig deeper into NSA work product. The FBI shouldn't have access to NSA product to begin with.

Any organization will act to grow its responsibilities and thus its resources. The USA has no real enemies anymore, and yet these agencies keep growing. Organizations will grow until they encounter some sort of limit.

Is the problem that those who would previously have overseen these agencies and curbed the over-extension you cite no longer want that oversight role, or that they are no longer capable of that oversight role? Barrack Obama talked on the phone all his life just like all other Americans do. His predecessors were the same. What kind of President would be able to cut the NSA budget?

But we do have an unmistakable and terrifically serious security problem: proliferation. We spent decades in an arms race with a foreign power that built an arsenal that could end the world several times over. When we won, they disintegrated. Meanwhile, the technology to duplicate atomic weapons technology that we had in the 1950s is improving.

There isn't a lot of foreign surveillance I can imagine having a problem with allocating to the problem of nuclear proliferation.

This would be an excellent priority for our intelligence agencies. Do any of the practices revealed recently seem designed to address that priority? An adversary capable of obtaining, maintaining, and deploying nuclear weapons would probably be capable of secure communications even in the face of PRISM, whether here or overseas. The most likely vector would be something like a submarine or a well-shielded shipping container, but if the plan did require agents in this country I doubt those would be discussing their plans on Facebook or over the phone.

The set of skills required to carry out an atomic attack and the set of skills required to evade surveillance by NSA are disjoint and seem unlikely to get more correlated. But that's just speculation on my part.

My only point here was that "terrorism" isn't the only threat that NSA is tasked with dealing with.

https://www.eff.org/deeplinks/2013/06/public-first-secret-co...

"For the foregoing reasons, EFF's motion is granted in part. The Court holds that FISC Rule 62 does not have the effect of sealing copies of the Opinion in the Government's possession and that the Court has not otherwise prohibited the Government's disclosure of such copies in response to the EFF's FOIA request. This Court expresses no opinion on the other issues presented in the FOIA litigation, including whether the Opinion is ultimately subject to disclosure under FOIA. Such questions are appropriately addressed by the District Court in the FOIA litigation."

Treatable mental illness: depression, anxiety, etc. do not fall into this category. They are not terminal.

Proximity to an instrument of death when a person is contemplating suicide (due to treatable mental illness) is a very real issue. If there is no barrier in place, the rate of change from thought to action is astonishing.

It's also important to note intoxication plays a role in this same sort of escalation.

In these cases you're looking for anything which increases access (proximity) or reduces inhibition (chemicals). Eliminating these gives professionals and loved ones time to help.

Years ago when I was working at a small ISP in Columbia, SC I had lunch with Austin Meyer. He came in to upload the latest release of X-Plane to his FTP servers. We went across the street to get some shrimp and grits.

The man oozes genius. Over lunch he was rambling about some technique he discovered to render scenes at a much higher frame-rate. Something to do with culling and inlining a series of functions responsible for a very particular set of expensive transformations.

His almost hysterical, passionate, hyper, consuming personality made it a completely one-sided encounter. It was my pleasure to just sit and listen.

While we were walking out, I noticed the tail of his plaid shirt had come partially untucked. He had also missed 2 belt loops.

We're reading today about a man who legitimately doesn't have the cycles to spare for belt loops, let alone a patent battle. If for no other reason than to stop distracting people like Austin from doing what they do, we desperately need software patent reform.

Anyway. He probably thought I was a dope, but I still recall that lunch as one of my fondest memories while starting out in the industry.

When we built out our message queue offering at SoftLayer (which uses Cloudant/BigCouch as a data store), latency out to the DB cluster simply never ended up being an issue. Properly leveraging DB features, I/O priority tuning, the caching tier behaving properly - all more often these end up being the culprit in performance issues.

What you need to better understand when considering a cloud-based data store is who's running your cluster. When it comes to guys like Adam, Mike and Robert, you're going to be very hard-pressed to find better (and available to you) talent in the CouchDB world. We certainly appreciate having them around.

The catch is you'll probably want to do this in a friendly country or one you're already a citizen of.