NavList:

   

Reply

Frank,

Thanks for a wonderful essay on smartphones. Some of us old guys have
a hard time keeping up with the latest inovations.

Ken Gebhart
On Jul 20, 2012, at 3:39 PM, Frank Reed wrote:

> Modern "smartphones" are pocket computers. They can do amazing
> things even without being activated as a phone. For example, I have
> an "old" first-generation iPhone from late 2007. It has no phone
> service, but it is fully WiFi capable so I have it as a spare
> Internet access device. If I am upstairs and my phone (a two-year-
> old Android Smarthpone) and laptop are downstairs, and my upstairs
> computer is off, no problem. I pick up the old iPhone, and I can
> almost instantly view any web site, check email, check NavList
> messages, and so on. I can also launch spreadsheets and run various
> astronomy programs that can do nearly anything that I would expect
> to see in a "desktop" computer astronomy application. Solve a
> spherical triangle correctly in all angular ranges? Hell, they do
> that in hardware --it's on the GPS chipset!
>
> Modern smartphones are more than pocket computers because they also
> include a "rich" suite of sensors. And that's an understatement.
>
> Smartphones can listen. If you've never seen "Shazam" in operation,
> it will seem like magic. A song is playing on the radio... you
> decide you want to follow the lyrics... so you tap an icon on your
> smartphone, tap another icon to set it "listening" and in five or
> ten second it comes back with the name of the song, the artist, the
> album artwork, a link to the lyrics, and more. And for many songs,
> if you tap another button, it will go into a "karaoke" mode and
> display the lyrics synched to the song still playing on the radio.
> This isn't science fiction. I do it almost every day.
>
> Smartphones can see. And I don't mean that they're cameras. They
> are indeed cameras, and smartphones have nearly wiped out the
> pocket camera market. But it goes beyond that. They can analyze
> what they "see". This functionality is relatively limited, but if
> you start up an app, for example, "Google Goggles" on your phone
> and point the phone at a piece of artwork (maybe a Renaissance
> painting in an advertisement in a magazine), or the cover of an old
> paperback book, or a logo on a bottle of wine, the app will analyze
> the image and then compare it with Google's vast database of images
> and often locate an exact match. In the case of the painting in the
> advertisement, it will then most likely link you to the Wikipedia
> article for the artist who painted it. And of course, this
> functionality can include translation: photograph a street sign in,
> say, Portuguese, and your phone will translate it for you. This
> basic "vision" functionality has been available for several years
> now. Big improvements are in the pipeline.
>
> Smartphones can measure altitudes and find directions. The earliest
> smartphones included accelerometers in two dimensions that could
> determine the tilt of the phone in two directions. This has obvious
> applications for leveling applications and basic celestial
> navigation. But bear in mind that the accuracy is no better than a
> tenth of a degree AND the device has to be calibrated since the
> accelerometers are aligned with the frame of the smartphone only to
> within a degree or two (in other words, there is an index
> correction, unless it is calibrated manually, of more than a degree
> in most cases). Nonetheless, this has made angular altitude
> measurement with Smartphones a relatively trivial application for
> nearly five years now. Back in the Spring of 2009 rumors began
> circulating that new smartphones would also have electronic
> compasses, and at that point, as I am sure some of you recall, I
> posted on NavList about how this could turn any smartphone into a
> star finder (and there was much "grumbling" on NavList about the
> impossibility of this). Less than a month later, Google Sky Map
> (now known simply as Sky Map since they have made it open source)
> became available and now there are numerous "point and shoot" apps
> for star-finding on smartphones (e.g. "Pocket Universe" on
> iPhones). It's important to recognize that these are only as good
> as the calibration of the compass. Since many smartphone users do
> not know that they have to wave the device around in a peculiar
> fashion to calibrate the magnetic compass (and since those who
> know, often feel embarrassed doing it), the direction capability is
> sometimes wrong by 10 or 20 or even 45 degrees. When calibrated,
> the error is typically less than five degrees. For over a year now,
> many smartphones have also been equipped with an electronic
> gyroscope. This can separate changes in orientation from linear
> accelerations, which is very useful in games but also in star-
> finding apps since it tends to smooth out the turns. I cannot
> emphasize enough that star-finding apps like this are tremendous
> crowd-pleasers. They are a fantastic way to learn the stars, both
> for beginners and experts. For myself, I enjoy placing the device
> on a table with Sky Map running. It lets me "look through the
> Earth" and see what stars are in the nadir at that time, something
> that we are rarely aware of even when we know the night sky well.
>
> Smartphones can determine your position. Long before iPhones and
> Android phones, smartphones have had GPS chipsets and other more
> coarse means of determining the user's position. There are now
> multiple methods of position-determination. GPS positions provide a
> fix within just a few seconds. There's no long bootup wait like in
> old GPS receivers since the network now provides the startup data
> to the device based on its coarse position which is found from
> Internet "IP" address or from the location of the nearest cell
> tower. In addition, huge databases have been created, originally by
> small startups but they have all been eclipsed by the Google beast,
> which provide the physical locations of millions of home WiFi
> networks. Those trucks that compile photography for Google
> Streetview are also sniffing the id's and signal strengths from the
> WiFi in you home and business. This WiFi-based positioning is
> nearly as accurate as GPS and it works especially well in those
> "urban canyons" where GPS is not quite so reliable. And note that
> even a coarse position based on network location is more than
> enough for star-finding and most astronomy applications.
>
> Smartphones create augmented reality. By combining multiple sensors
> and especially the device's camera, it is possible to display data
> directly on the camera's live view of a scene. You aim your phone
> down a street and it labels every restaurant, for example, and
> shows yelp ratings in little bubbles floating over each one. This
> sort of application has been slow to catch on in large part because
> of the embarrassment factor of aiming your phone down a street.
>
> Smartphones are watching you. Among the most recent enhancements
> are phones that watch your eyes and know whether you're reading the
> screen. If you're still reading, the display will not "time out"
> and go dark. Strange but true. Disturbing yet practical. And
> smartphones are already doing facial recognition but strictly on an
> "opt in" basis because of privacy concerns. In other words, the
> technology can identify specific individuals and note their
> presence in photo after photo but you have to tell the phone that
> this person is "Jane Doe" from your contacts list.
>
> A short note on naming: there are at this time two dominant
> operating systems on smartphones and related tablet computers. They
> are iOS and Android. Apple created and developed the iPhone, first
> launched in 2007 and they have since named its operating system
> "iOS" to cover the multiple devices it runs on, especially the
> sizzling hot iPad. Google almost simultaneously launched the
> Android operating system to steal as much as they could from
> Apple's projects without going to jail. There is a good analogy
> here with the Mac versus Windows in the 80s and 90s. Since Apple
> has not licensed iOS and probably never will, the market of devices
> which run iOS is quite small but beautifully integrated and
> efficient. It's "iPhones" and "iPads" with the "iPod Touch" as a
> minor shrub in the Apple orchard. From the outset, Google has
> attempted to make friends and influence corporate partners by
> selling Android as the "open" alternative to Apple's walled
> orchard. This has also partially protected Google from massive
> lawsuits since Apple has gone after the device manufacturers who
> use the Android operating system first. This approach has led to a
> proliferation of devices and serious fragmentation of the market
> (since there are numerous major versions and minor variants of
> Android). In addition, all of these devices have unique names
> despite the fact that they are all running Android. So for example
> there are Heros, and Nexuses, and Galaxies, and even the Nook and
> the latest Kindle tablets run Android. There is also a line of
> phones from Motorola called Droids. This is where things can get
> confusing. All Droids run Android, but by no means all Android
> devices are Droids. For people who are familiar with the Apple side
> of the market, this is especially surprising. Incidentally, the
> name "Android" required no special licensing since it is a long-
> standing generic science fiction term. But Motorola had to license
> the name "Droid" from a very wealthy man. Can you guess who? Hint:
> "These aren't the droids we're looking for."
>
> Finally, as far as Gary's report of less than 5 minutes of arc
> error in an LOP from a smartphone app, that's either good luck or
> it's an app that is cheating (perhaps even in a way that the
> programmer did not realize). There's no magic here --beyond the
> amazing magic that you get all of these features in a handheld
> device... oh, and it's a phone, too... :). The device cannot be
> better than an ordinary bubble sextant as far as altitude
> measurements go. Without calibration for index error, you can
> expect errors in LOPs of one or two degrees. With calibration and
> assuming no magnification and sighting along the edge of the
> device, you could expect no better than about half a degree
> accuracy on a single sight. This could be improved exactly the way
> that traditional bubble sextants improved accuracy by averaging
> over time. And of course, this would be an easy addition to a
> smartphone app. If there's calibration for index correction, and if
> averaging is done, and if there's a nice clean edge to sight along
> (no guarantee given the multitude of Android devices), I think you
> could expect standard deviation errors in sights on the order of
> five to ten minutes of arc. Of course, if we throw in coarse
> positioning from a user-input DR or from the network, the device
> could also usually guess what stars you're looking at and you
> wouldn't even need to know basic star patterns to get a position.
> Useful? Probably not in any real practical sense. Fun? Well, for a
> certain personality type. Myself included. :) Educational?
> Absolutely, if you ask students to figure out what the device is
> doing and relate it back to historical navigation, there's a
> thousand lessons here. And that's really the great thing about all
> of this: any "kid" with a smartphone, ignoring obvious financial
> barriers for the moment, can download these software apps and learn
> star-finding and navigation topics and a thousand other things...
> There's tremendous opportunity for education.
>
> -FER
>
>
> ----------------------------------------------------------------
> NavList message boards and member settings: www.fer3.com/NavList
> Members may optionally receive posts by email.
> To cancel email delivery, send a message to NoMail[at]fer3.com
> ----------------------------------------------------------------
>

   

Reply