What I'm proposing is to hold up the final 250 OK until the message has been passed on, then report the result of the forwarding as an SMTP status. If immediate forwarding is not possible, return a 421 Service Not Available, so the sender will retry. If the forwarding returns an error status, return that error status. No need for local message storage or bounce messages.

I've heard Dolby's positional audio, being driven from a game, in the Dolby Labs screening room in San Francisco. It sounds great. You can hear people sneaking up behind you in the game. You can hear someone walking around you. There's a real sense of presence.

That's in a room built, at a cost of millions, as a demo for Dolby's audio technology. The room is on a separate foundation from the rest of the building, with an inner set of vibration isolated walls. The room acoustics are very good; you don't need a microphone when giving a talk there. The walls and ceiling conceal speakers everywhere, and the room with the amps and processors looks like a small server farm.

You're not going to get that in Joe Sixpack's living room. You might get close to it in some high end home theater installations, the ones that look like small movie theaters and are used for no other purpose. It's a niche market.

One of my back-burner ideas is speeding up email forwarding. Most email forwarders (sendmail, etc.) accept emails, put them in a queue, and then later spool them out to the destination. This adds a minute or so of latency. It's done this way for historical reasons. In the early days, the destination mail agent might be down, or the mail transfer might be over some polled protocol like UUCP.

That's dead. Today, if the destination mail agent exists, it's probably up and immediately reachable via a fast connection. So a modern mail fowarder should accept the incoming email via SMTP, and then, while holding the incoming connection open, send the email on to the destination mail agent. Any problems are immediately reported to the sender via SMTP status code.

This not only speeds things up a bit, it eliminates "bounce messages" generated between mail agents. Problem reports come back immediately, as SMTP errors. There's a series of open TCP connections from sender to the receiver's IMAP server. From the IMAP server to the final destination, today you usually have some kind of push notification. So you get the effect of instant messaging, using existing email protocols.

This also eliminates "joe jobs", where impersonation generates vast numbers of bounce messages. The spammer just gets lots of SMTP errors, which never bother anybody else.

Agreed. Anyone know what kind of exploit this is?

Didn't look at the source of a Youtube page, did you? Look for "http://s.ytimg.com/yts/swfbin/player-vflZsDuOu/watch_as3.swf". Videos can also play with "HTML5 video", but there's Flash code there to be executed.

Presumably this attack is via a Flash vulnerability. So why is there no mention of Adobe in the article? Why isn't Adobe being held responsible? Why are there still vulnerabilities in Flash? Who audits that code? Well?

It's good to see this discussion becoming mainstream. Back in 2009, Alan Cox wrote in the Atlantic, "One sometimes wonders, in this era of Market religion, where the skeptics and freethinkers have gone". There's been an assumption in recent decades, since the USSR went down, that capitalism is the only possible system of economic organization. That's starting, cautiously, to be questioned.

In the Great Depression of the 1930s, all sorts of "isms" were proposed. Communism, socialism, technocracy, and others now forgotten had substantial followings. WWII ended those discussions, and the postwar boom made them irrelevant. Now, few people even know that alternatives to capitalism are possible.

The "strong safety net" countries (the Scandinavian countries, some EU countries, and Japan) have done reasonably well. The production side is mostly capitalist, but taxes are higher and the consumption side is partly socialist. This works if international competition is limited to stop the "race to the bottom" in wages. The current Doha round in the WTO is stalled because many countries now want more protectionism.

To be fair, it's a lot easier to write a compiler (or any other program) if you have an existing compiler on hand to help you do it. Writing a compiler using only assembly or machine code is well beyond most undergrads' capacity.

I think this is very close to true -- in particular, many new C++ features are there mainly so that the STL can 'magically' do the right thing in more cases. Mere mortals are not expected to make use of the new features directly; rather they are expected to use the improved STL and benefit from its smarter behavior.

Flying cars work just fine. They're called quadcopters. They're just not for people. Like space travel. The future belongs to robots.

It's just a battery factory. It's unlikely it will employ that many people. Tesla says 6500, but that's probably exaggerated, including the construction phase. The battery factory for the Chevy Volt has only 100 people. It's a big, highly automated plant.

This is a naive article. For a better analysis, see "How Asia Works", which is a comparison of the coastal Asian countries, how they developed, and why. Development requires several phases. One is raising agricultural productivity. There's the heavy-handed approach, which comes in the communist form of collective arms and the capitalist form of big plantations. Then there's the light approach, which involves lots of little services like tractor rental and agricultural agents. (The heavy-handed approach works well only for flat land. Hill operations require too many local decisions.) There's thus a visible relationship between what a country looks like and its Gini coefficient.

The second phase of development is about industrialization. Where investment goes really matters. Market forces do not direct investment towards overall economic growth, but toward short-term profit. The successful "Asian tigers" all had very directed investment controls, and how well countries did relative to each other depends on how well investment was directed.

The book has lots of country-by-country comparisons, both statistical and on the ground. It's worth a read.

Most people could learn to do those things (with greater or lesser degrees of skill) if they cared to devote the time to required do so.

Take 32 of those bits and put them together, now you've got a floating point value that can represent "true" as 1.0, "false" as 0.0, and a few million shades of "maybe" in between those two extremes.

If that's not analog-y enough for you, make it 64 bits and now you can have trillions of shades. And if that's still not enough, add more bits until you've got the resolution you're looking for.

I don't see any significant distinction between analog and digital, since digital logic asymptotically approaches analog as you add bits, and with today's memory sizes there are plenty of bits to go around.

Or perhaps because it's running a radically different kind of algorithm that no human has ever understood or implemented on a digital computer.

This is an important new thing. We've had question-answering programs working against specific data sets since Bobrow's "Baseball" program of the 1960s. We've had a whole range of question-answering specialist systems running in tandem since Yahoo introduced vertical search around 2005. But cross-topic generality has been elusive.

If this is real, it's a major development. Is there anything better than the Tired article available?