Thursday, November 30, 2006

Ready for a new day?

The team I came to join at Microsoft been working hard on a new printing subsystem and a new digital paper format, known as the XML Paper Specification (or XPS), in Windows Vista.

In addition to being built into Vista, XPS support is included for application developers in the newly released .NET Framework 3.0 and for end users of the new 2007 Microsoft Office System (alongside support for saving as PDF). A quick query in your favourite search engine shows a wide range of partners are adopting XPS.

Today is the business launch day for Windows Vista, the 2007 Microsoft Office System, and Microsoft Exchange Server 2007. More information about that little event is available on www.msnewday.com, but more importantly it offers up a nice segue into congratulations to the Digital Documents Platform Team (D2) for Windows Vista and XPS: Well done.

tags: | |

Wednesday, November 29, 2006

Live Writer and Flickr

Want to add pictures direct from flickr into your blog posts without too much messing around?

Grab a copy of flickr4writer, a plug-in for Windows Live Writer that makes adding pictures to blog posts a snitch.

There. No more excuses.

tags: | | |

Tuesday, November 28, 2006

Popular Cameras (part II)

Having pondered this for a bit, of course what you'd really want to know is not which cameras are the most popular (by counting uploaded photos), but which cameras take the better pictures (by counting some 'better picture' metric).

What metric? flickr has a concept of interestingness which I've always been a bit suspicious of. Interestingness is a secret combination of various factors, but since I have a picture with 0 comments, 0 favourite tags and 0 views listed as one of my more interesting photos, the algorithm is either extremely insightful, or not as clever as we'd like. Ideally I'd like to take some combination of the number of favourites plus number of views divided by some factor that combines the number of tags and groups to which the picture belongs [1]. But, and it's a big but, getting the data I want for a better-than-interestingness metric isn't currently possible, so instead here's a look at the number of photo's by different camera models in the top 500 interesting pictures with a taken date over the past twelve months.

flickrinterestingcameras

 [Click for larger image]

Caveats:

  • the unknowns are the largest category by far. They're pictures for which the flickr API didn't return an EXIF tag with model information.
  • the data also doesn't include images for which the flickr user elected not to be included in external queries. So here's the top
  • I've combined Digital Rebel and Digital Rebel XT numbers with 300 and 350 respectively since I assume that's what the flickr stats have done

So here's another view, this time of the top 10 cameras by interesting photos in the top 500 excluding unknowns:

flickrinterestingcamerasII

 [Click for larger image]

 

The top five cameras by number of pictures uploaded (based on data from the flickr cameras page):

  1. Canon EOS Digital Rebel XT / 350D
  2. Nikon D50
  3. Canon EOS 20D
  4. Nikon D70
  5. Canon EOS Digital Rebel / 300D

and here's the top five by number of pictures in the top 500 interesting pictures:

  1. Canon EOS 350D / Digital Rebel XT
  2. Canon EOS 20D
  3. Nikon D50
  4. Nikon D70s
  5. Canon EOS 5D

Conclusions? The number of unknowns is enough to swamp any meaningful comparison, but since there's nothing but SLRs in the top of the list, it's safe to assume that if you want 'interesting' pictures, a dSLR is the way to go (or that people that have dSLRs known how to play the 'interesting' game with their uploads to flickr ;-)

 

 

[1] The rationale here is that adding tags or adding to groups increases the visibility of the picture to the flickr community and so would result in an increase in views and favourites irrespective of how good the picture is.

tags: | | |  | |

Wednesday, November 22, 2006

Popular Cameras

Via Scripting News: flickr is posting stats from the camera's used to take the pictures posted on flickr (scroll down to the graphs under the sponsored slot at the top).

Question: does the fact that the SonyEricsson K750i is the most popular camera phone mean that:

a) people that use flickr seek our good camera phones (and therefore that the 750 is one of the better camera phones?)

b) the pictures taken with other camera phones aren't worth posting to flickr

c) the pictures taken with other camera phones are posted to other photo sharing sites

4) other camera phones don't do the right thing with the EXIF tags in the picture metadata

e) something else

 

Extrapolating a bit, I wonder if there's a correlation between the photo sharing site you use and camera you use (and further, the newspaper you read, political party you last voted for, charities you donate to...)

 

Tags: | | |

Making the News

Congratulations to Global Graphics - recent PR suggest they've been busy! Although I'm sure Peerless, Zoran, Monotype, Software Imaging, Artifex and others have been just as busy...

In contrast: not widely reported in the tech media is Adobe's Mars project. From the Mars FAQ:

The information in a Mars document is organized similar to a PDF document. Mars represents document information by combining standard XML, images, fonts, and color formats within a Zip-based package. Page content is represented in SVG. In Mars, the pages, images, fonts, bookmarks, and other document components appear as separate files within the Zip package. The Mars components such as bookmarks are easier to manipulate by virtue of their XML representation.

Much like a web page which consists of a number of separate files referenced by a “root” HTML file, a Mars document consists of many file packaged together in a Zip file. There is a root "backbone.xml" file that references other files in the document.

Mars looks like a significant development, but so far there's no mention on either PDFZone or PlanetPDF. Given the interest around PDF and XPS I'd have expected at least some comment from people like Karl De Abrew or Don Fluckinger.

What could explain this? My guess is that it's a consequence of most of the news being push driven, with interested parties trying to push their stuff onto the news agenda. A traditional push tool has been the press release. No press release, no news. I didn't see a press release from Adobe about Mars, so maybe that explains it. If so, perhaps blogs aren't influencing the agenda as much as people think.

Tags: | | | | | |

Thursday, November 16, 2006

DRM (& more on sign-up)

The DRM bit: David Berlind over on zdnet points to an interesting article on Microsoft and Rights Management by Matt McKenzie. Both articles are worth a read, because they point to the tension between the increasing importance of open standards and the protection of intellectual property - both your IP and that of other people.

This is complex stuff, and although I'm no expert there are compelling arguments on both sides of the issue. On an intellectual level that makes it a very interesting area of the business of technology to watch. But, in the real world, whatever the long term solution turns out to be, unaware or uneducated consumers are at risk of getting burnt while we get from here to there. It's important that the industry gets this one right, whatever 'right' turns out to be.

The Sign-Up Bit: There's a link in David's article that should point to Matt's article in computer world, but unfortunately right now it's broken (it points to a piece on Zune by Erica Ogg). I was going to place the right link in David's comment stream, but zdnet wanted my home address just to sign up. No thanks. I suppose I'll have to send David an email instead ;-)

Tags: |

Wednesday, November 15, 2006

This feature is P0

What's my highest priority feature for an on-line (or Web 2.0 if you prefer) photo site?

Auto tagging?

Geo tagging?

Open and fair policy on metadata?

Support for video?

Support for HDR?

Support for color management?

Trackbacks?

Slide shows?

Unlimited storage?

Unlimited uploads?

Image manipulation tools?

Fancy multimedia annotations?

Backup and retrieval?

AJAX UI?

Nope, it's none of those.

What I love to see is very quick, very simple, very easy sign-up.

What's the number one reason that friends and family tell me they can't view the private piccy's we post? That's right, signing up for the sharing site we use is too hard. flickr is great if you already have a yahoo! account, but for the non-techie masses things aren't that simple, and the antispam fun and games just make it harder.

Why post about this now? Because Kristopher Tate of zooomr fame has posted a link to a video on how to sign up for a zooomr account using OpenID. Well, it made me smile.

I've recently had friends send links to pictures on Kodak Gallery and Ringo. I couldn't work out why they were using these - especially Ringo that doesn't have the brand awareness of Kodak - until I twigged how simple sign up is. For example, Kodak asks for name, email and a password with a tick box for accepting the T&Cs and another for adding your mobile phone - all on a single page. Yahoo! (e.g. for Yahoo! Photos) asks for first name, second name, preferred content, gender, yahoo ID, password, checkbox for yahoo mail, security question, security question answer, birthday, ZIP/Post code, Alternate email [optional], industry [optional], title [optional], specialization [optional], and the verification code before you can sign in. All that, and then you have to wait for a confirmation email sent to your alternate email address before you can go do anything.

No wonder some people I know have given up on signing up for flickr (now owned by Yahoo!) - which is a real pain for me 'cause flickr is such a great service otherwise.

Come on flickr, any chance of a simple signup like we had in the Old Skool days? Please?

 

Tags: | | |

Monday, November 13, 2006

Automagic tagging - not yet

For a while I've been advocating, and enjoying, the benefits of tagging. Whether it's publicly using tools like flickr, zooomr and del.icio.us, or privately on our piccy collection using Picasa or, across a wider range of documents, with the new tagging capabilities built into Vista. Using free-form tags to add metadata to objects - particularly piccys where the volume of images and content can be large - is quicker than using a more rigid taxonomy, but it's still a time consuming process. That's why I'm always on the look out for an automatic mechanism to tag, or as a first step provide assistance for manual tagging, our piccy collection at home.

For a while a service called Riya looked promising. Riya enables you to tag piccy's and then, using some clever vision-recognition algorithms, provides tags automatically for future images based on the analysis of the images you initially tagged. Unfortunately it appears the focus behind Riya has switched away automatic tagging, last week the team behind Riya launched like.com, an online visual-based fashion shopping site. According to Scoble:

Why not keep working on face detection? Because they learned through user testing that they’d never be able to make it good enough. They found that by focusing on visual image searches they can get a much more satisfied user base.

Another online application that aims to do similar things to Riya is Alipr. Provide an image (or a URI to an image) and Alipr suggests appropriate descriptive keywords from a vocabulary of 332 English language words.

A writeup in MITs Technology Review suggests that on a test of 5,411 images on flickr, Alipr provided at least one keyword that matched a manually assigned flickr tag for that image 98% of the time. A quick test using public images from my flickr stream was less successful.

The limited success of Riya and Alipr suggest that, for some time at least, we'll need to wait on improvements in vision recognition algorithms before we get much help in tagging our image memories. In the meantime though, we can sit back with a certain sense of satisfaction that all the manual effort in tagging images has the added bonus of creating some nice large datasets to help improve those algorithms...

...and with that thought, back to the tag-fest ;-)

Tags: | | |

Friday, November 10, 2006

Photogrammetry

When I was at college studying Photographic Science one of the more interesting subjects in Sid Ray's applied photography module was - the measurement of objects using photographic images (aside, doesn't Sid have just about the best name for someone that writes books about ?).

Well, things have come a long way since my days in the labs at Riding House Street... Photosynth is a technology that takes a large collection of photographs of a particular place and reconstructs a three-dimensional space from which you can explore the images. Microsoft labs have just released a technology preview that includes four sample collections: the Piazza San Marco in Venice; Grassi Lakes in the Canadian Rockies; Piazza San Pietro in Rome; and the studio of artist Gary Faigin in Seattle. It's a great, if sometimes befuddling, way to explore a space. I can't wait to see what my own collection can look like.

Scoble it was the killer app of the Web 2.0 Summit.

Here's a snap of photosynth in action, you can try out the preview for yourself on  (system ).

photosynthphotosynth Hosted on Zooomr

 

I see stuff like this and think what more is to come in this rich world of imaging ;-)

Tags: | |

Tuesday, November 07, 2006

Virtual Earth

local.live.com has been our friend since we moved to the US, the birds eye view is great for figuring out where things are when you're in a new and strange place and the ability to save collections has been useful for building our own geographical list of important places (like the post office, shops, places people have recommended to eat etc.). The coverage for Birds Eye is better in the US than elsewhere, but I met the person whos company is behind the aerial photography in Europe at a school sports day in the UK, and other places are gradually coming on stream.

This week, local.live has added support for 3D views and, as a fan of Google Earth, I had to check it out.

 

Above is the space needle in Seattle from a recent trip we made while below is something similar from local.live.com

seattleMSVirtualEarthseattleMSVirtualEarth Hosted on Zooomr

 

I tried getting views from Gas Works and Rainier, but you can't tilt the viewpoint upwards enough from ground level to get close to a match, while Snoqualmie Falls just doesn't have the imagery (yet?).

It's not as feature rich as Google Earth, but the imagery is great and it's nice to be able to flick back and forth with regular mapping in the same application. Two downers: you'll need to be running Internet Explorer, so no Firefox; and the CTRL-Mouse controls feel wrong headed for my simple hand-eye coordination.

Tags: | | |

Thursday, November 02, 2006

Search on

Thanks to Windows Live Search C R O S S O A K is now searchable! If you want the add search to your own site, go grab the Windows Live Search Box.

 

Big thanks to the Live Search team.

 

[PS Written from a Toshiba M7 tablet running Windows Vista :-)]