Day: March 15, 2011

Flipboard makes news worth flipping through

Rakesh Agrawal2 Comments
Photo spread in Flipboard.
Photo spread in Flipboard.

I wrote earlier about why iPad magazines aren’t selling well. The short version: they weren’t designed to meet the needs of readers, they were designed to meet the dreams of publishers.

Flipboard is designed around the optimal consumer experience. It offers a visually compelling personalized magazine edited by you, your friends and others you trust.

While Twitter has been offering personalized news streams for years, it’s as visually compelling as classified ads: scroll through a long list of  terse text-only snippets with bizarre abbreviations to find something you like.

Flipboard takes content from your Twitter, Facebook, Google Reader and other feeds and renders it automatically with photos, call outs and other design elements that were sadly lost with the transition to the Web and personalized information. It can even generate beautiful photo spreads like the one at right.

I’ve looked at many iPad news apps and Flipboard is the one I come back to. I frequently mark content on my laptop to read later using Flipboard. It is one of the first apps that I check when I get up and the last one that I use before I go to sleep.

Flipboard hits the key elements that are important to the future of news:

  • It’s personalized. The days of one-size-fits-all newspapers are numbered. Prepackaged content from a bunch of people I’ve never met (and who have never met me)  just isn’t compelling. People have diverse interests and differing intensity in those interests. I couldn’t care less about what Lindsay Lohan is up to or what’s going on on Jersey Shore. But I can’t get enough of what’s going on in location-based services. Mass publications are catering to a mass audience that is rapidly disappearing. Magazines do better at this because many are catered to niche audiences, but even they can’t get deep enough for some.
  • It’s hyperpersonal. When I worked in the newspaper business, there was a rule of thumb for what made the front page. The further away it was from us, the more people had to die for it to make A1. With hyperpersonal, things like a friend completing a triathlon, having a baby or getting married can make my own A1. I discovered the power of social and hyperpersonal when I was on a business trip to Dublin. I shot a video of our VP chugging a Guinness and posted it online. Within minutes, that video had circulated all over the offices in Dublin and back in Virginia.
  • It’s social. My interests overlap with the interests of my friends. As a result, I get news that is likely to be of interest to me through Flipboard.  Even the incongruities are helpful. They add a bit of serendipity to news that professional editors seek to provide. Seeing friends’ comments also adds to my understanding.
  • It’s timely. Weeklies and dailies might as well be fortnightlies. With Flipboard, the content is up to the minute.
  • It’s free. People will only pay for very limited types of news. The vast majority of authors will have to find other ways to make money. That’s a challenge. And it’s a challenge that is exacerbated by many talented people who are willing to write for free.

Flipboard has the opportunity to reset the focus on content. One thing that has jumped out at me is how much I enjoyed reading content in tablet form. It wasn’t just the use of touch or the lack of a keyboard. The big reason: the content wasn’t struggling for attention against SEO links and ads for refinancing homes I don’t own.

SEO links have cluttered up many news sites. Content is surrounded by links whose sole purpose is to feed the search engine crawlers. Navigation on many sites is next to impossible. Ironically, this just drives people back to search because it’s much easier to find what you’re looking for there.

Ads, especially remnant ads, have a similar effect. Publishers have chased remnant ads for revenue. (“Even a fraction of a cent is better than nothing.”) As RPMs have gone down, publishers have piled on more ads. In the end, this does a disservice to everyone. It’s the tragedy of the commons. Instead of grass, the common good being destroyed is the attention of readers.

Flipboard is in a rare position to help bring better economics to publishers and a cleaner experience to users. It’s already working with some publishers on specialized content presentations.

As much as I like the existing product, there are a few areas for improvement:

  • Personal prioritization. Even with the limited number of sources I have in Flipboard, I don’t have time to read everything. One of the functions of editors at newspapers and magazines is to help decide what’s important. Flipboard doesn’t yet do this. It would be nice if it would algorithmically determine what were the most important stories and topics to focus my attention on.
  • De-duplication of content. In the current implementation, you can have the same story repeated multiple times. This could be used as a signal in prioritizing content.
  • Aggregation of friends’ comments. Related to the above, when multiple people have shared a story, collapse them into one. And then show the comments from all friends.
  • Better content extraction. When it lays out pages, the snippet that is rendered is often nonsensical. This could be done with site-specific crawling, by offering publishers guidelines for Flipboard optimization or both.
  • Better photo placement. Flipboard will sometimes place horizontal photos into vertical spaces and vice versa, poorly cropping the pictures. At a minimum, photos should use the right aspect ratio. Eventually, I could see Flipboard using techniques like facial recognition to more intelligently crop photos.
  • Publication of personal magazines. It’s so easy to create personal magazines that I want to be able to share with others. For example, I quickly created a Portland food magazine. (This might finally give Twitter lists a raison d’etre.)
  • Filtering out topics. There are times when my friends are all talking about something I’m not interested in and it just floods my stream. An easy way to get rid of these items (e.g. everything related to SXSW) would significantly improve the experience.

Disclosure: I worked for Flipboard founder Mike McCue when I was at Tellme. Before jumping fully into the online world, I worked at washingtonpost.com and launched startribune.com.

EXIF marks the spot: a guide to geotagging pictures

Rakesh Agrawal3 Comments

Geotagging pictures allows you to more easily search your photos based on where you’ve been. In addition to online tools such as flickr and Picasa, desktop apps like iPhoto will take advantage of geotags. (Facebook is a notable exception; it currently doesn’t use geotags.)

You can also create visualizations of your travels like this:

A map of my favorite pictures on flickr.
A map of my favorite pictures on flickr.

Digital photo files contain EXIF headers that store information about the picture. The information that most people are familiar with is the time and date. Each picture can contain hundreds of fields that include minutiae about the cameras settings, including aperture, shutter speed, program modes, etc. Geotags store latitude, longitude and elevation. You can see a sample of EXIF data on flickr. (The geotags are toward the bottom of the page.)

Unfortunately, the actual process of geotagging is still cumbersome. It’s a lot easier than it used to be, but shy of what it should be to make it mainstream.

Here are a few options for geotagging your pictures:

A dedicated GPS that can record GPX tracklogs

A hiking GPS can record your every move. As you walk, hike or run, the GPS unit logs your current location. (As frequently as once per second.) These are recorded on a memory card in a .gpx file. When you return to your computer, you can pull that file and then the photos from your camera’s memory card. Specialized software (such as GeoSetter) then synchronizes the timestamps of the photos with the data from the GPX tracklog and writes the location information into the file.

Don’t worry if the timestamps are off. GeoSetter offers numerous ways to adjust the timestamps so that the time recorded in your pictures lines up with the time in the location data. (To make this easier, I recommend taking a picture of the time display on your GPS at the beginning and end of your trip.)

I’ve used a Garmin eTrex Vista Cx. Any GPS that can write GPX files to a memory card will do the trick.

Advantages: Works really well when outdoors. Precise location, due to GPS accuracy and frequency of updates. The GPS also provides valuable information when you are hiking.

Disadvantages: Additional cost for the GPS unit. Because you aren’t looking at the GPS unit when taking pictures, you may miss errors such as being out of coverage, dead batteries or a full memory card. It’s another thing to carry. Applying the coordinates is a multi-step process. If you go to a distant location, the GPS can take up to 20 minutes to get an initial fix. It may be difficult to get a fix in densely packed urban areas. (In any case, when you head indoors, you’ll have to rely on the last reading.)

Using an Android or iPhone app to record GPX tracklogs

This works pretty much the same as having a dedicated GPS, except that it relies on your cell phone to track your position. The biggest downside is that it will chew through your phone’s battery very quickly. I use Motion-X GPS on the iPhone and My Tracks on Android.

Advantages: Works really well whether indoors or outdoors. Precise location, due to GPS accuracy and frequency of updates. The GPS also provides valuable information when you are hiking. No additional cost. Because the phone can approximate your location in other ways, the time to a GPS fix is much faster.

Disadvantages: Because you aren’t looking at the phone when when taking pictures, you may miss errors such as being out of coverage, dead batteries or a full memory card. Multi-step process. The GPS app will chew through your phone’s battery very quickly.

A geotagging digital camera

Specialized digital cameras can automatically geotag photos. I have a Panasonic DMC-ZS7. It will automatically write the current location into the data file when you take a picture. No lining up separate files or manually geotagging pictures. The Panasonic has a built-in landmarks database. Standing in front of the Statue of Liberty? The screen will show “Statue of Liberty”. In playback mode, the camera will let you browse pictures by location.

Incredible? Yes. Too good to be true? Sadly, also yes. When it works, it’s like magic. When it doesn’t… it just adds incorrect data to the picture.

There are two big problems with Panasonic’s implementation: it takes way too long to get a fix and it doesn’t update when the camera takes a picture. Often when I’ve arrived in a new location, the camera is still showing the old location hours later. The instructions claim that the camera updates location even when it’s off; I haven’t found that to be true.  Even at its best, the location only updates every five minutes. Undoubtedly, the camera’s designers faced challenges trading off the accuracy of the GPS location against battery life. The balance they struck made the GPS feature largely useless.

Advantages: Simplicity. Works OK when outdoors. Because the location is shown on screen, you can determine whether it’s correct and that there are no other issues.

Disadvantages: Location information is often wrong. Very long time to first fix. The GPS uses the camera’s battery. (I didn’t find this to be a huge issue, but you may want to carry a spare battery.) Despite the fact that the camera knows the nearby landmark, it doesn’t write it into the EXIF data in a way flickr and other tools can read. Poor indoor coverage. Additional cost when compared with cameras without GPS capability.

Eye-Fi memory card

Eye-Fi sells a line of memory cards that will geotag locations. The primary purpose of the cards is to automatically upload your pictures to the Internet. But they’ve expanded the capabilities to also geotag the pictures.

Here is how Eye-Fi works: when you take pictures, nearby WiFi networks are recorded. During the upload process, those network locations are used to compute a location using Skyhook’s database of WiFi locations.

Advantages: Simplicity. Works well indoors, especially dense urban areas in the United States.

Disadvantages: The Eye-Fi cards cost substantially more than comparable SD cards (sometimes 10x). It won’t work when you’re in an area without WiFi signals, which rules out EyeFi for geotagging many hikes. The Eye-Fi card’s WiFi capabilities will drain your camera’s battery faster. The locations are added after the fact, so you won’t be aware of any problems when shooting. Geotagging relies on Skyhook’s database of WiFi locations, which can be sparse in foreign locations.

Manually geotagging pictures

For a while, this was the only option. Take pictures that you’ve uploaded and drag them onto a map. This can be as accurate or as inaccurate as you want it to be.

You can take all of your pictures of Venice and drop them onto the city of Venice. Or you can zoom in to just the right piazza and repeat the process for each picture. I once looked into the background of an old picture, found a business name there and did a Google Maps search to put it in the right place. It can be tedious, fun or both. Flickr and Picasa both support manual geotagging.

Advantages: Doesn’t cost anything other than your time. You have precise control over where each picture is placed.

Disadvantages: It can be extremely tedious. Lining up pictures taken outdoors (such as while skiing or hiking) can be difficult.

Using your cellphone’s camera

This is likely the way that most people get into geotagging. If you have an Android phone or iPhone, your camera can do all of the work for you. Based on the same services used for maps and other location services, the phone will write the picture’s coordinates straight into the file. You can also verify that the location is correct by launching a maps app before taking the picture.

It’s so easy that many privacy advocates worry that people are unintentionally revealing their locations when uploading pictures.

Advantages: Simplicity. Can verify information on screen. No additional cost.

Disadvantages: As good as they are, the cameras on phones aren’t as good as regular cameras. This is especially true for pictures needing zoom or taken in low light.

The best solution would be if the camera manufacturers would work with the phone manufacturers to just read the current GPS data when the shutter is pressed. I’d bet that the iPhone gets a microSD card slot before that happens.