Popfly, Microsoft’s Innovative Mashup Programmer, Cancelled

From John Montgomery comes word that Microsoft’s Popfly project has been cancelled. I was bullish on Popfly, predicting that it would become the power-user’s entrance to Silverlight.

I have had a history of wishful thinking about “restoring the bridge between power users and programmers.” I think Microsoft’s Powershell is incredible and it absolutely boggles me that it’s not gotten more traction.

Is it possible that I’m just clinging to an outmoded value? That these potential bridges don’t get traveled because people aren’t interested in the destination?

Kindle DX Review

I love my original Kindle. But the Achilles Heel of the Kindle has always been technical publications. The original Kindle does not have the screen real estate and formatting engine to properly deal with equations, graphs, and, most importantly for programmers, source code. Although it is possible to view some PDF files on the original Kindle, there are likely to be major formatting goofs.

Enter the Kindle DX, which might as well be called “Kindle: Textbook Edition.” The DX features a 9.7-inch diagonal screen (compared to 6-inch on the original and Kindle 2) and can display PDF files natively. It is not nearly as casually portable as the trade-paperback sized Kindles, but it is vastly more totable than a 500-page text.

The eInk display is better for reading than any screen you’ve ever dealt with before. It’s not even in the same league as a laptop display or an iPhone display. My plane-flying routine is to lose myself in reading and I’ve consumed entire novels on the Kindle during East Coast to Hawaii trips. You can’t do that with any other display (even if you had the battery power in your laptop/iPhone etc.).

PDF display is a huge deal because so much technical content comes in PDF form. The Kindle DX displays an entire 8.5″ x 11″ PDF page on a single screen. Note the lack of a qualifier in that sentence. If you have an 8.5 x 11 PDF page to read, the Kindle DX will display the entire page on a single screen. There is no zoom feature for PDF pages.

I hope these photos make the problem apparent:

CACM: Print vs. PDF on Kindle DXFont: Print vs. PDF on Kindle DX

This is a fair comparison: the print version of the Communications of the ACM versus its PDF version (unfortunately not the same page, since I haven’t yet received the print version in the mail!). The print font is about 40% larger than the display font. It’s still readable (for me), but not easily and not for hours on end. Citations, which use an even smaller font, are too small for me to read at arms length.

Now, it’s wonderful that the Kindle DX can display a page layout in its intended form and navigating a multi-column PDF in zoom mode is frustrating, but the option should be there Update: The excellent comment below led me to enable the ‘auto-rotate’ ability of the DX (behind the font button). I had locked my DX in portrait mode; in landscape mode, an 8.5 x 11 PDF is much more readable. One could hope that the Kindle could update this with a firmware update, but Amazon seems in no rush to add features to the Kindle (I think it was 14 months between firmware updates for the original).

Other PDF flaws: you cannot attach notes and highlights to specific ranges within a PDF. Although you can add bookmarks, you cannot title them helpfully. So, for instance, in a technical book, it’s counterproductive to bookmark within the index; I just drop a bookmark on “M” in the index and page from there.

The DX keyboard has another flaw worth mentioning. For space reasons, they have combined the numeric keys with the top row of letters. Which makes sense but for the fact that jumping to a numeric location is the #1 use of the keyboard! In order to enter a numeric location, you have to hold down an ‘alt’ key. Interpreting ‘QPPP’ as location 1,000 is another seemingly-simple feature that one can wish for in an update.


For reading, I prefer the DX to the original. I read quite quickly and although one eventually tunes out the page turns, it’s nicer to have more text on the screen. Also, the DX is significantly better for reading at the gym; on non-PDF files you can zoom the text to a size you can read while on a machine. For portability of technical books, it’s hard to express how much better the DX is. For general portability, the original’s small size makes it essentially invisible, very easy to tuck into a carry-on, etc.

For a technical professional or student, the ability to view PDFs is a great feature, but is obviously already available on a laptop or desktop. Is the Kindle DX a perfect machine for working with PDFs? No. The lack of zoom and annotations are significant flaws.

Personally, I am happy with my purchase, but I am a huge outlier on the “print consumption” graph — shipping costs to Hawaii are very significant, I review a lot of technical books, and I subscribe to a dozen magazines. Unless you’re in the same odd position, I suggest that you stick with a Kindle 2 for now.

Petition to Apple to Create/Improve APIs to Allow Augmented Reality Apps on the iPhone

Apropos my recent discovery of camera heading in iPhone metadata comes word of a petition for Apple to create functions that would allow Augmented Reality development on the iPhone:


Augmented Reality are applications that combine computer-generated imagery (or text) with imagery generated from a camera. Examples include applications that show historic photographs from your current vantage point, identify significant features in the landscape (buildings, mountains, constellations), and games.

iPhone 3Gs Embeds Heading (Not Orientation) in EXIF Metadata

Further to last night’s post and based on some carefully aligned photos, I am a little disappointed to conclude that the EXIF value GPSDirectionRef that the iPhone 3Gs inserts is (only) the camera’s magnetic heading expressed as a rational number and not (as I’d hoped) the camera’s 3D orientation in space.

No sure conclusion on why it’s expressed as a rational and why the denominator varies, but my guess remains that it’s precision.

A review of other EXIF data doesn’t show any other data indicating orientation of the camera (i.e., no tilt or raw accelerometer numbers).

Implications for hackers: You can probably guess what the camera was aimed at by simply casting a cone out from the geocode and the heading (update: Hah! And the orientation! If the camera is in landscape mode, the heading is still the short-side of the rectangle. But note that the iPhone seems to set tiff:orientation EXIF value to ‘1’ regardless of mode.). But I don’t see any way to put Godzilla’s feet on the ground. (Of course, additional EXIF metadata is a simple thing for Apple to add in a firmware update, so as the augmented reality wave rolls in, look for it in updates.)

Pedantic trivia: The ‘heading’ of an object is the direction in which it is pointed. The ‘bearing’ is the position of an object relative to your location. GPSImageDirection clearly ought to be the bearing of the photograph, but is, with the iPhone, the heading (the direction in which the top of the iPhone is pointed).

iPhone 3Gs Encodes Camera Orientation in EXIF (?)

On a hunch I checked the EXIF data attached to my first photos from my new iPhone. The intriguing news is that it has values such as:

GPSImageDirection : 2535/383
GPSImageDirectionRef: T

The GpsImageDirectionRef = “T” appears to be a constant.(Update: signifies True and not Magnetic north.)

Pointing the camera roughly N,E,S,W resulted in:

N : 2535/383 (and then, pointed ‘up’: 19845/113)
E : 27081/81
S : 20749/113
W: 28546/113

My guess is that this is a fraction representing degrees from N. The use of a denominator may indicate accuracy (?). When I first saw it, I thought it could encode tilt, but that doesn’t seem to fit the data (at least to my eye). No other obvious EXIF source of tilt info.

More experimenting in order.

Ob-hack: Augmented reality. Place Godzilla / Cloverfield properly in all photos of Tokyo / New York.

Update: It’s just heading. Cue sad trombone.

DasBlog Links Finally Rewritten

Driven by the mocking of a colleague for never having done the stupid gruntwork of rewriting my old DasBlog-based permalinks, I stole Jon Udell’s python code and ran it over the 8+ year history of Knowing.net. I think it mostly worked, although internal links between posts remain broken, since the DasBlog GUID-based identifiers have been lost.

I suppose I should now write a program to monitor the 404s and do manual fixups as traffic calls for, but I’m not even going to pretend that “I’ll get to that in a couple days.”

I apologize if this caused any spurious “There are 2100 new posts” messages from your newsreader.

[76702,706], signing off…

CompuServe Classic is shutting down. I had no idea they were still extant. I wonder if the CLMFORUM and AIEXPERT message boards are archived…

Coincidentally, tomorrow is the 20th anniversary of my moving to the San Francisco Bay Area to become Product Review Editor at Computer Language and AI Expert magazines (and to get a whizzy “767” administrative CompuServe account).