Archives

Date
  • 01
  • 02
  • 03
  • 04
  • 05
  • 06
  • 07
  • 08
  • 09
  • 10
  • 11
  • 12
  • 13
  • 14
  • 15
  • 16
  • 17
  • 18
  • 19
  • 20
  • 21
  • 22
  • 23
  • 24
  • 25
  • 26
  • 27
  • 28
  • 29
  • 30
  • 31

Border Travel in an underpants bomber world

I just landed on a flight from Toronto to San Francisco. If you were inside the USA you may not have heard about the various crazy rules applied to travel to the USA, or at least not experienced them. While we were away the rules changed every day, and perhaps every hour.

Toronto was hit the hardest because it has the most flights to the USA of any airport in the world (with a few other Canadian airports not far behind.) Due to the busy border, you clear U.S. customs and immigration through their satellite office in Toronto, so your plane lands you at domestic gates in the USA, making connections far easier.

The USA started insisting on intimate pat-downs on all passengers and complete hand screening of all carry-ons. For a while there was even a regulation that passengers would have to sit in their seats with nothing on their laps (not blankets, not books, not computers) for the last hour of the flight. That got reverted to “pilot’s discretion” and in our case there was no talk of this.

The heavy search requirements brought Toronto’s heavy to-USA traffic to a standstill. Even with extra mounties pitching in, there was now way to get all those people through the terminal, so the CATSA brought in a near-ban on carry-ons. You could only carry on items from a short list. Notable things not on the list (ie. banned) included books, kid’s toys, lenses and various items people bring on not because they need them in flight, but because they are essential to their trip, or are fragile.

After a few days of reduced carry-ons, they got the processing down, as long as you got there 3 hours in advance, sometimes more. A real burden on 1 hour flights to New York, Boston or Washington. Still a burden on my 5 hour flight to SFO, since that was at 7am, meaning getting to the airport at 4am, (1am Pacific Time, about the time I would get to bed.)

The process included the fairly standard x-ray (with agents making various exceptions for people, generally allowing books that could be paged through and even some small knapsacks) with pat down only if you set off the alarm. Then, shortly after you started walking down the row of gates was a 2nd checkpoint. There you got a serious patdown that might remind you of a massage, and a complete hand inspection of everything in your bags. (I suggest they should let you pay extra for a real massage, which also of course detects anything on your body.) Many checks of ID and boarding pass and you are on your way.

There are many disturbing things about the reaction to the underpants bomber but a few stand out.

  • It is certain that the TSA and all other major agencies knew about the risk of somebody strapping explosives to their legs and taking them through the magnetometer. So a plan should have been in place long ago about what to do about it, and how to react at the first public incident.
  • In spite of this the agencies are out running around like chickens with their heads cut off, changing plans every day, no sign of forethought. Are they just testing the public to see what they will tolerate?
  • Lots of talk of thz scanners to see everybody naked. Is this a way to get those accepted, after people complained?
  • For Toronto, and most of the Canadian airports, a bad guy can quite readily drive just 90 minutes and go to another airport like Buffalo and get no special screening! While the public does not like this extra trek, it’s no burden to the terrorist to do this. Only the innocent are punished.
  • You could still smuggle your stuff inside a laptop, or a body cavity or several other places I noticed.
  • Keep this up and people will stop flying, and they will definitely go to airports like Buffalo.
  • As I have suggested before, appointments for security inspections are one answer to the 3 hour early arrival.
  • For me the worst thing was packing lenses in checked bag. I had to improvise protection for them. When such a rule is put in place by surprise over Christmas, you have to expect a lot of people brought stuff that they needed to carry on on the way back, even if they would not plan a new trip today expecting to carry on their fragiles.

With some irony, all this came after a lunch with Peter Watts. If you didn’t hear, Peter was crossing back into Canada at Port Huron/Sarnia and got pull over for exit inspection leaving the USA. Because he wasn’t a complete little sheep, he reports he was beaten up by the border patrol and now is charged with assaulting an officer. I really doubt he did those things, but the most disturbing thing are those who comment on the story saying it’s his fault for not being subservient enough. I understand the reasons for letting police do their jobs, but when you are just inspecting people driving out of the country, with no special reason to believe they are criminals or worthy of above average suspicion or anything but the presumption of innocence we are all owed, then there should be standards, and better defined rights for the subject of the inspections. If a person is not a known threat, why should they not get to ask questions about what is being done to them and their vehicle? Yes, one time in many thousands, an actual nasty criminal might do something odd and need to be set upon with force. It’s one of the risks people take doing an armed policing job. It can happen anywhere, any time. But must the people give up their rights and be complete sheep because of it?

Can’t we have a system where different situations suggest different levels of police control? Where the police, while they may have the power to give you orders and you have to obey without much chance to question, get in trouble if they abuse that power in a non-hostile situation? Where they have a simple way of explaining that they think the situation has escalated, and a way to declare it that we are taught in school to understand? So if the copy says, “I’m escalation — get on the ground now” you have to get on the ground, but the cop has to justify later why he escalated. Simply being a citizen who is mindful of his rights doesn’t seem much grounds for that.

Happy Seasons to all

I’ve been feeling we in the secular, atheist world should still have an official event at the end of the year, since with the Christians and the Jews making merry, it’s a good time to do it. We have New Year’s Eve of course, but so does everybody.

This new holiday, to mark the changing of the Seasons might be called “Seasons.” Of course that is in part so that all the people saying “Seasons Greetings” (without the apostrophe, oddly enough) will now be making our greeting. Another name for it could be “Holidays” but that does have religious roots.

Now the real changing of the seasons is around Dec 21, but it makes far more sense to celebrate a gift holiday on Dec 28-29. That way you can buy gifts for everybody at half price. And we seculars are smart and thrifty.

So Happy Seasons to all.

Video windows that simulate 3-D

I’m waiting for the right price point on a good >24” monitor with a narrow bezel to drop low enough that I can buy 4 or 5 of them to make a panoramic display wall without the gaps being too large.

However, another idea that I think would be very cool would be to exploit the gaps between the monitors to create a simulated set of windows in a wall looking out onto a scene. It’s been done before in lab experiments with single monitors, but not as a large panoramic installation or something long term from what I understand. The value in the multi display approach is that now the gap between displays is a feature rather than a problem, and viewers can see the whole picture by moving. (Video walls must edit out the seams from the picture, removing the wonderful seamlessness of a good panorama.) We restore the seamlessness in the temporal dimension.

To do this, it would be necessary to track the exact location of the eyes of the single viewer. This would only work for one person. From the position of the eyes (in all 3 dimensions) and the monitors the graphics card would then project the panoramic image on the monitors as though they were windows in a wall. As the viewer’s head moved, the image would move the other way. As the viewer approached the wall (to a point) the images would expand and move, and likewise shrink when moving away. Fortunately this sort of real time 3-D projection is just what modern GPUs are good at.

The monitors could be close together, like window panes with bars between them, or further apart like independent windows. Now the size of the bezels is not important.

For extra credit, the panoramic scene could be shot on layers, so it has a foreground and background, and these could be moved independently. To do this is would be necessary to shoot the panorama from spots along a line and both isolate foreground and background (using parallax, focus and hand editing) and also merge the backgrounds from the shots so that the background pixels behind the foreground ones are combined from the left and right shots. This is known as “background subtraction” and there has been quite a lot of work in this area. I’m less certain over what range this would look good. You might want to shoot above and below to get as much of the hidden background as possible in that layer. Of course having several layers is even better.

The next challenge is to very quickly spot the viewer’s head. One easy approach that has been done, at least with single screens, is to give the viewer a special hat or glasses with easily identified coloured dots or LEDs. It would be much nicer if we could do face detection as quickly as possible to identify an unadorned person. Chips that do this for video cameras are becoming common, the key issue is whether the detection can be done with very low latency — I think 10 milliseconds (100hz) would be a likely goal. The use of cameras lets the system work for anybody who walks in the room, and quickly switch among people to give them turns. A camera on the wall plus one above would work easily, two cameras on the left and right sides of the wall should also be able to get position fairly quickly.

Even better would be doing it with one camera. With one camera, one can still get a distance to the subject (with less resolution) by examining changes in the size of features on the head or body. However, that only provides relative distance, for example you can tell if the viewer got 20% closer but not where they started from. You would have to guess that distance, or learn it from other queues (such as a known sized object like the hat) or even have the viewer begin the process by standing on a specific spot. This could also be a good way to initiate the process, especially for a group of people coming to view the illusion. Stand still in the spot for 5 seconds until it beeps or flashes, and then start moving around.

If the face can be detected with high accuracy and quickly, a decent illusion should be possible. I was inspired by this clever simulated 3-D videoconferencing system which simulates 3-D in this way and watches the face of the viewer.

You need high resolution photos for this, as only a subset of the image appears in the “windows” at any given time, particularly when standing away from the windows. It could be possible to let the viewer get reasonably close to the “window” if you have a gigapan style panorama, though a physical barrier (even symbolic) to stop people from getting so close that the illusion breaks would be a good idea.

Twitter clients, only shorten URLs as much as you truly need to and make them readable

I think URL shorteners are are a curse, but thanks to Twitter they are growing vastly in use. If you don’t know, URL shorteners are sites that will generate a compact encoded URL for you to turn a very long link into a short one that’s easier to cut and paste, and in particular these days, one that fits in the 140 character constraint on Twitter.

I understand the attraction, and not just on twitter. Some sites generate hugely long URLs which fold over many lines if put in text files or entered for display in comments and other locations. The result, though, is that you can no longer determine where the link will take you from the URL. This hurts the UI of the web, and makes it possible to fool people into going to attack sites or Rick Astley videos. Because of this, some better twitter clients re-expand the shortened URLs when displaying on a larger screen.

Anyway, here’s an idea for the Twitter clients and URL shorteners, if they must be used. In a tweet, figure out how much room there is to put the compacted URL, and work with a shortener that will let you generate a URL of exactly that length. And if that length has some room, try to put in some elements from the original URL so I can see them. For example, you can probably fit the domain name, especially if you strip off the “www.” from it (in the visible part, not in the real URL.) Try to leave as many things that look like real words, and strip things that look like character encoded binary codes and numbers. Of course, in the end you’ll need something to make the short URL unique, but not that much. Of course, if there already is a URL created for the target, re-use that.

Google just did its own URL shortener. I’m not quite sure what the motives of URL shortener sites are. While sometimes I see redirects that pause at the intermediate site, nobody wants that and so few ever use such sites. The search engines must have started ignoring URL redirect sites when it comes to pagerank long ago. They take donations and run ads on the pages where people create the tiny URLs, but when it comes to ones used on Twitter, these are almost all automatically generated, so the user never sees the site.

Wanted: An IRC Bot to gateway to a twitter backchannel

It’s now becoming common to kludge a conference “backchannel” onto Twitter. I am quite ambivalent about this. I don’t think Twitter works nearly as well as an internal backchannel, even though there are some very nice and fancy twitter clients to help make this look nicer.

But the real problem comes from the public/private confusion. Tweets are (generally) public, and even if tagged by a hashtag to be seen by those tracking an event, they are also seen by your regular followers. This has the following consequences, good and bad.

  • Some people tweet a lot while in a conference. They use it as a backchannel. That’s overwhelming to their followers who are not at the conference, and it fills up the feed.
  • When multiple people do it, it’s almost like a spam. I believe that conferences like using Twitter as backchannel because it causes constant mentions of their conference to be broadcast out into the world.
  • While you can filter out a hashtag in many twitter clients, it’s work to do so, and the general flooding of the feed is annoying to many.
  • People tweeting at a conference are never sure about who they are talking to. Some tweets will clearly be aimed at fellow conference attendees. But many are just repeats of salient lines said on stage, aimed only at the outsiders.
  • While you can use multiple tags and filters to divide up different concurrent sessions of a conference, this doesn’t work well.
  • The interface on Twitter is kludged on, and poor.
  • Twitter’s 140 character limit is a burden on backchannel. Backchannel comments are inherently short, and no fixed limit is needed on them. Sure, sometimes you go longer but never much longer.
  • The Twitter limit forces URLs to be put into URL shorteners, which obscure where they go and are generally a bane of the world.

Dedicated backchannels are better, I think. They don’t reach the outside world unless the outsiders decide to subscribe to them, but I think that’s a plus. I think the right answer is a dedicated, internal-only backchannel, combined with a minimal amount of tweeting to the public (not the meeting audience) for those who want to give their followers some snippets of the conferences their friends are going to. The public tweets may not use a hashtag at all, or a different one from the “official” backchannel as they are not meant for people at the conference.

The most common dedicated backchannel tool is IRC. While IRC has its flaws, it is much better at many things than any of the web applications I have seen for backchannel. It’s faster and has a wide variety of clients available to use with it. While this is rarely done, it is also possible for conferences to put an IRC server on their own LAN so the backchannel is entirely local, and even keeps working when the connection to the outside world gets congested, as is common on conference LANs. I’m not saying IRC is ideal, but until something better comes along, it works. Due to the speed, IRC backchannels tend to be much more rapid fire, with dialog, jokes, questions and answers. Some might view this as a bug, and there are arguments that slowing things down is good, but Twitter is not the way to attain that.

However, we won’t stop those who like to do it via Twitter. As noted, conferences like it because it spams the tweetsphere with mentions of their event.

I would love to see an IRC Bot designed to gateway with the Twitter world. Here are some of the features it might have.  read more »

Why facebook wants you to open up your profile

There is some controversy, including a critique from our team at the EFF of Facebook’s new privacy structure, and their new default and suggested policies that push people to expose more of their profile and data to “everyone.”

I understand why Facebook finds this attractive. “Everyone” means search engines like Google, and also total 3rd party apps like those that sprung up around Twitter.

On Twitter, I tried to have a “protected” profile, open only to friends, but that’s far from the norm there. And it turns out it doesn’t work particularly well. Because twitter is mostly exposed to public view, all sorts of things started appearing to treat twitter as more a micro blogging platform than a way to share short missives with friends. All of these new functions didn’t work on a protected account. With a protected account, you could not even publicly reply to people who did not follow you. Even the Facebook app that imports your tweets to Facebook doesn’t work on protected accounts, though it certainly could.

Worse, many people try to use twitter as a “backchannel” for comments about events like conferences. I think it’s dreadful as a backchannel, and conferences encourage it mostly as a form of spam: when people tweet to one another about the conference, they are also flooding the outside world with constant reminders about the conference. To use the backchannel though, you put in tags and generally this is for the whole world to see, not just your followers. People on twitter want to be seen.

Not so on Facebook and it must be starting to scare them. On Facebook, for all its privacy issues, mainly you are seen by your friends. Well, and all those annoying apps that, just to use them, need to know everything about you. You disclose a lot more to Facebook than you do to Twitter and so it’s scary to see a push to make it more public.

Being public means that search engines will find material, and that’s hugely important commercially, even to a site as successful as Facebook. Most sites in the world are disturbed to learn they get a huge fraction of their traffic from search engines. Facebook is an exception but doesn’t want to be. It wants to get all the traffic it gets now, plus more.

And then there’s the cool 3rd party stuff. Facebook of course has its platform, and that platform has serious privacy issues, but at least Facebook has some control over it, and makes the “apps” (really embedded 3rd party web sites) agree to terms. But you can’t beat the innovation that comes from having less controlled entrepreneurs doing things, and that’s what happens on twitter. Facebook doesn’t want to be left behind.

What’s disturbing about this is the idea that we will see sites starting to feel that abandoning or abusing privacy gives them a competitive edge. We used to always hope that sites would see protecting their users’ privacy as a competitive edge, but the reverse could take place, which would be a disaster.

Is there an answer? It may be to try to build applications in more complex ways that still protect privacy. Though in the end, you can’t do that if search engines are going to spider your secrets in order to do useful things with them; at least not the way search engines work today.

Robocars of the future from 1958 and Google Tech Talk

You may have seen it already but it’s amusing to watch this encoding of a 1958 Disney show on the highway of the future:

This highway features a mixture of human driven cars, robocars and PRT style robocars on private guideways. Much of it is typical of ancient predictions of the future, with an expectation of remarkably cheap and strong materials and portable atomic power, but some of it is on the mark (like urban sprawl.) The roles of mom and dad don’t change in 50 years. It is always humbling to go back into the past and see how futurists have got it wrong, and wonder where you are going wrong in your own predictions. (I’ve learned you should never predict dates for things because while you might have a sense about how long it would take to develop the technology, you can’t as easily predict how markets and governments will react.)

However, I will be prognosticating again next week, giving my Robocar talk in the Google “Tech Talk” series at Google HQ in Mountain View on Dec 14 at 11am. While not generally open to the public, I can bring in guests if they all come in at around 10:30am. Contact me if this is of interest. They will put the talk up on Google Video when done. Of course, if you are a Googler, I hope to see you there.

It’s amusing to note that many of the vision seen in the movie “Minority Report” are found in this Disney video from much earlier.

Make e-Ink tablets an add-on for our phone/PDAs, not stand-alone

It’s over 17 years since I first too a stab at e-Books, and while I was far too early, I must admit I had not predicted I would be that early. The market is now seeing a range of e-Ink based electronic book readers, such as the kindle, and some reasonable adoption. But I don’t have one yet. But I do read e-books on my tiny phone screen. Why?

The phone has the huge advantage that it is always with me. It gives me a book any time I am caught waiting. On a train, in a doctor’s office, there is always a way to catch up on reading. It’s not ideal, and I don’t use it to read at home in bed, but it’s there. The tablets are all large, and for a good reading experience, people like them even larger. This means they are only there when you make deliberate plans to read, and pack them in your bag.

I’m not that thrilled with e-Ink yet, both for its low contrast and the annoying way it has to flash black in order to reset, causing a distracting delay when turning the page. There are ways to help that, but as yet it suffers. e-Ink also can’t readily be used for annotation or interactive operation, so many devices will keep a strip of LCD for things like selecting from menus and the like. Many of the devices also waste a lot of space with a keyboard, and the Kindle includes a cellular radio in order to download books. e-Ink does have a huge advantage in battery life.

What makes sense to me instead would be a sheet (or two sheets, folded) of e-Ink with very little in the way of smarts inside the device. Instead, it would be designed so that a variety of cell phones could dock to the e-Ink sheet and provide the brains. Phones have different form factors, of course, and different connectors though almost all can do USB. (Though annoyingly only as a slave, but this can be kludged around.) It would be necessary to make small plastic holders for the different phone models which can mate to a mount on the book display, ideally connecting the data port at the same time. The tablet of course should be able to connect to a laptop via USB (this time as a slave) but do the same reading actions. The docking can also be, I am reminded by the commenters, done by bluetooth, with interesting consequences.

This has many large advantages:

  • Done right, this tablet is a fair bit cheaper. It has minimal brains inside, and no cell phone. In fact, for most people, it also does not include the cost of a cell phone data service. (I presume with the Kindle the cost of that is split between the unit and the book sales, but either way, you pay for it.)
  • The cell phone provides an interactive LCD screen to use with all the reader’s interactive functions — book buying, annotating etc.
  • The cell phone provides a data connection for downloading books, newspapers and web pages.
  • The cell phone provides a keyboard for the few times you use a keyboard on an e-Book reader
  • When you don’t have your e-Ink tablet, you still have all your books, and can still order books.

The main thing the cell phone doesn’t have is huge battery life. The truth is, however, that cell phones have excellent battery life if they are not turning on their screen or doing complex network apps. We do such activities of course, and they drain our batteries, but we expect that and thus charge regularly and carry more. I’m not too scared at the idea of not being able to read my books with the phone dead.

The tablet could also be used with a laptop, especially a netbook. Laptops can actually run for a very long time if you put them in a power conserving mode, turning off the screen and disks, possibly even suspending the CPU between complex operations.

However, there is no need to run it at all. While I described the tablet as being dumb, it takes very little smarts for it to let you page through a pre-rendered book that was fed to it by the phone or laptop. That can be done with a low power microcontroller. It just would not do any fancy interactive operations without turning on the phone or laptop. And indeed, for the plain reading of a single book, akin to what you can do with the paper version, it would be able to operate on its own.

Of course, the vendors would not want to support every phone. But they could cut a deal to let people use old supported phones (which are in plentiful supply as people recycle phones constantly) with a minimal books-only data plan similar to the plans they have cut for the dedicated devices. In the GSM world, they could offer a special SIM good only for book operations for use in an older phone of the class they do support. And they could also build a custom module that slots perfectly into the tablet with the cell modem, small LCD screen and keyboard for those who still want a stand-alone device.

This approach also allows you to upgrade your tablet and your phone independently.

As noted, I think a folding tablet makes a lot of sense. This is true for two reasons. First, you get more screen real estate in half the width of tablet. Secondly, with two e-Ink panels, you can play some tricks so that you flash-refresh the panel you aren’t reading rather than the one you are finishing. While slightly distracting (depending how it’s done) it means that when you want to switch to the next page, you do it with your eyes, with no delay. You have to push a button when you switch (even going from left to right though it’s not apparently needed) so that the page you have fully finished refreshes while you are reading the next one. This could also be done with timings. Or even with a small camera watching your eyes, though I was trying to make the tablet dumber and this takes CPU and power right now. I can imagine other tricks that would work, such as how you hold the tablet (capacitive detection of your grip, or accelerometer detection of the angle.)

The tablet could also be built so the two pages of e-Ink are on the front and back. In this case it would not fold, though a slipcover would be a good idea. A “flip tablet” would display page 1 to you with page 2 on the back. To read page 2 you would physically flip it over. It would detect that of course, and change page 1 to page 3 when it was on the other side. This would mean the distraction of the flash-refresh would not be visible to you, which is a nice plus.

Cutely, the flip tablet could detect which direction you flip it. So if you flip it counter clockwise, you get the next page. If you flip it clockwise you get the previous page. Changing direction means you might briefly see the flash while you are flipping the unit but the UI seems pretty good to me. For those who don’t like this interface, the unit could still hinge out in the middle to show both pages at once.

Bluetooth connection

Using bluetooth for the connection has a number of interesting consequences. It does use power, and does not allow exchange of power between the tablet and device, but it means you don’t have to physically put the phone on the tablet at all. This may be a pain in some circumstances (needing two hands to do interactive things) but in other circumstances having a remote control to use to flip pages can be a real win.

I have found a very nice way to do e-reading is to have the pages displayed in front of you, at eye height, rather than down low in your hands. In particular, if you can mount your tablet on the top back of an airplane seat, it is much more comfortable than holding a book or tablet in your hands. The main downside is that the overhead light does not shine on the page there, so you need a backlight or LED book light. The ability to do remote control from your phone, in your rested hand would be great. Unfortunately they have the strange idea that they want to ban bluetooth on planes, though it poses no risk. They don’t even like wires.

In the 90s I built a device for reading books on planes where I got a book holder (they do make those) and I rigged it to attach with velcro and hang from the back of the seat in front. In those days it was quite common to have velcro on the top of the seat. Combined with a book light, I found this to be way more comfortable than holding a book in my hands, and I read much more pleasantly. Today you might have to build it so that a plate wedges between the raised table and seatback and a rod sticks out to hold the tablet.

Or on some planes they could support e-books on the screen in the seatback, with the remote control that is in your armrest. Alas, they would indeed need to use bluetooth so your PDA could display the book on that screen. (In general, letting your PDA use the screen in front of you would be very nice. It’s too sucky a resolution for laptops, since it must have been designed years ago in an era of sucky resolution. Today 1650 x 950 displays cost $100.

How to do a distributed Twitter (MSM)

Dave Winer recently made a call for an open source twitter shell, which he suggests be perhaps done with a javascript framework to let any site act like twitter.com. Many people are interested in this sort of suggestion, because while the folks at twitter.com are generally well loved and felt to be good actors, many people fear that no publishing system that becomes important should be controlled by just one company.

For success, such a system would need to be as easy to use and set up as twitter for users, and pretty easy to set up for server operators. One thing it can’t do so easily, alas, is use a simple single namespace the way twitter does. A distributed system probably has to make names be domains, like E-mail addresses. That almost surely means something longer than twitter names and no use of the @name syntax popular in Twitter to refer to users. On the other hand almost everybody already has a domain based ID, ie. their E-mail address. On the other hand most people are afraid to use this ID in public where it might get spam. It’s a shame, but many might well prefer to get a different ID from their E-mail, or of course to use one at twitter, which would now look like user@twitter.com to the outside world instead of @user within twitter.

Naming problems aside, the denizens of the internet are certainly up to building a publish/subscribe based short message multicasting service, which is what twitter is using terms much older than the company. I might propose the name MSM for the techology (Multicast Short Message)  read more »