Let’s begin with a tour of the functionality. When you start the app, it loads some resources while displaying the title screen, and then you have the chance to view an introduction or skip ahead to the map screen.

Because the project centers around the White River near Indianapolis, we only assembled hydrological data for the area around Marion County for the application (I’ll go into more detail later). On the map screen, a prompt appears to inform you that tapping on the map will simulate a rain event. When the map is tapped, the app displays the series of streams, storm drains, and/or sanitary lines that would carry a raindrop from that location to the White River. It also displays the area (known as a drainage basin or watershed) from which other raindrops would follow the same path. Another prompt then appears to let you know that tapping on the raincloud icon allows for selection of storm intensity. As little as a quarter inch of rain can cause sewers to overflow into streams in this area, so when this option is selected, the path displayed will change to reflect this so you can see where you don’t want to go fishing. You can also toggle the display of the 100-year floodplain, which shows you where you can keep your feet dry during a big flood event. In addition to selecting a location on the map, pressing the compass icon locates your device via GPS, and typing in the address bar uses the Google Maps address look-up feature. Tapping on a question mark icon provides some information about pollutants that threaten the path upstream, as reported by the Indiana Department of Environmental Management.

Pressing the “i” icon at the top opens the informational menu. From here, you can learn more about the app, check current weather alerts and conditions, find out how weather differs from climate, get some tips on how to improve water quality, and visit the project website.

Now we can get into some behind the scenes stuff. We wanted to try to reach a broad audience with Raindrop, so we decided to put the time that we had into developing a cross-platform mobile application. These are known as web apps, in contrast to native apps. If we only had an iPhone native app in the App Store, people using Android phones wouldn’t be able to use it, and vice-versa. To handle cross-platform compatibility, we decided to build Raindrop using a framework called jqMobile, which was in a very early stage when we started. It hasn’t quite had an official release yet (it’s in its third beta release at the moment), but has become increasingly robust with each version.

As for the map, you might wonder how we figured out the path that raindrops take to get to the river. Our collaborator at Williams Creek combined information based on digital elevation models, which can be used to derive the boundaries of natural watersheds, with data from the city that indicates where all of the storm drains and sanitary systems are and which areas drain into them.

We then wrote Python scripts to read the scientific data and generate KML geometry files and look-up tables. The application uses a spatial grid look-up to figure out which basin is tapped (so it’s not perfectly accurate, but not too slow either), and then loads the appropriate file with the graphics to display for the path and the basin. It also reads information from another table that has all the details about pollutants.

Along the way, we’ve combined this technology with graphical elements and design guidance provided by Mary’s team, and scientific guidance and content from Butler. The multi-disciplinary process has really embodied the nature of Mary’s City as A Living Laboratory concept. And just as the aim is to lead curious folks from Mary’s eye-catching mirrors and markers along the river to the website and the web app to learn more, hopefully those who discover the project online will follow the raindrop and find their way down to experience the river as well.

When I first visited Japan, I bought an electronic dictionary. This saved me from looking up kanji (the complex characters borrowed from Chinese) by counting strokes and identifying radicals (the root component of a kanji character) to index into the enormous tome that I had been using. The dictionary was much lighter, and had a stylus that could be used to draw kanji. Using this sort of input method, the order that you draw the strokes still matters, but it’s much faster than flipping pages. I used this dictionary for getting around Japan, studying, reading manga, and playing video games. Years later, after the Nintendo DS came out, I upgraded using a cartridge called Kanji Sonomama Rakubiki Jiten. It uses the same stylus input method, but the results are marked up in color and it has a nicer interface, including a history of recently looked-up words, which is extremely useful. One of the other tools that I was using at the time (and still use today) is a plugin for Firefox called Rikai-chan. When enabled, this plugin allows you to hover the mouse over a word and see the definition in a pop-up.

Rikai-chan also has dictionaries for German, French, and Russian, and similar tools may exist for other languages too, so make sure to look for this kind of plugin for the language that you study if you like to read online material in the original language.

My favorite dictionary now is Japanese for the iPhone. You can look up words by typing roman characters with the default Qwerty keyboard, or by enabling the Japanese and Simplified Chinese Handwriting keyboards in the iPhone settings you can enter a word using the Japanese input method (which presents you with possible kanji as you type phonetically), or by drawing kanji. It includes expressions, proverbs, conjunctions, and interjections, and has an impressive catalog of example sentences. As if this isn’t enough, it also has a vocabulary list feature. I like to create a list of new words for each novel, manga, or video game that I’m into, and then study them using the flashcard feature wherever I am when I have a spare minute. The flashcard feature uses an algorithm based on a series of steps.

When you identify a word correctly, it advances to the next step, and eventually into a “Known” category. Usually there are about twenty or so items between the “Unknown” and “Known” categories that you’re working on, but occasionally you see some items that are in the “Known” category as well. I’d highly recommend an app for your language of interest that has this sort of system for reviewing vocabulary.

In fact, I recently discovered Smart.fm, an online tool/community that’s based on a similar system. There’s a video on the home page that describes how their system automatically determines which material you need to be presented with at a given time to have the best chance of remembering it. I actually recall reading an article a while back about a man who was living his life according to this sort of algorithm. One of the other interesting things about Smart.fm is that in addition to the Goals (sets of material to study) curated by Smart.fm, the user community can create their own. So you can find Goals for varying grades of vocabulary for Japanese, Spanish, and other languages, as well as Goals consisting of the capitals of the world or the names of stars and galaxies. For Japanese, the system does a good job quizzing you on the phonetic reading of a word, the meaning of the word, and the kanji. Some of the goals include sentences, which serve as examples and are also presented to quiz your knowledge. From what I’ve read, images can also be incorporated into the material for a Goal, so the system has the capability to support many areas of study. Unfortunately, their main quiz interface is a Flash app and they dropped the iPhone app that they had developed, but I suspect that they’re working on a new HTML5 solution that will work both on their site and on multiple mobile platforms.

On their own, the tools I’ve mentioned really only help review and expand vocabulary. Without a solid foundation in grammar, even Rikai-chan can only help so much. I have a few books and dictionaries for grammar reference, but I haven’t come across any nice apps or plugins to help in this area of language study. My best suggestion here is to take classes to build your foundation, keep your books, and seek out interesting material at your level to practice with. I’m currently trying to figure out if there’s a good way to get manga in Japanese on the iPad. If you can build some friendships with native speakers, that’s another good way to explore new grammar, and maybe even learn a regional dialect.

And of course the ultimate boost for your language skills, once you have a good start, is to immerse yourself by living abroad. You’ll learn more about the culture, customs and lifestyle, which will reinforce your understanding of language and vice-versa. You might even gain some new perspective and learn more about yourself. And you’ll always have an urge to go back someday.

These are the steve tagger’s top 100 most contributed tags. The red ones happen to be a few of those that I’ve entered. You might notice that some of them look a little funny, like lightgray. This is because we use a normalization process to equate words that are essentially the same. For example, consider these tags on The Boat Builders by Winslow Homer:

The Boat Builders, Winslow Homer

A term like “New England” may have been entered with proper capitalization, or with all lowercase or uppercase characters. To treat these entries as the same basic concept, we normalize them to “newengland”. Similarly, some folks might type a term like “seashore” with a space, and other folks without a space. We recognize that sometimes these minor differences make a significant difference in the meaning of a term, so we keep all of the original tags as they were entered. This normalization handles most cases properly (further research into the occurrence rate of special cases continues in the Text Tags and Trust project), and it allows us to display objects that have been tagged with all variants of a term when it is clicked in the tag cloud.

After handling simpler transformations like whitespace removal and lowercasing, the next step we’re interested in taking is lemmatization. You can see a few examples where this can be done in the tag cloud above… the term “rocks” can be lemmatized to “rock”, and “children” to “child”. By adding another normalization routine which takes this extra step using a lemmatizer from the Natural Language Toolkit (which makes use of the WordNet database’s built-in morphy function), we can generate the following tag cloud for this painting:

This may just be my perspective, but I find that the lemmatized version really gives a better sense of representation, uncluttered by redundancy. We’re still doing some fine tuning, and looking into how to handle terms with multiple words. Our partners at the University of Maryland are doing some research to figure out how we might define heuristics to handle multiple word terms (e.g. a rule that says we should lemmatize the second word of an adjective-noun pair), based on the collection of tags that we currently have.

As we study the results, we will definitely be considering scenarios where this sort of normalization fails to recognize nuance and gives misleading results, and how to handle this both in our research and in user interfaces. It’s a tricky problem that is sure to lead to many interesting questions and findings about folksonomic linguistics to discuss in the future.

While previous movies like Sin City have done a great job borrowing static elements of graphic novel design, this is the first film I’ve seen that makes such fine use of dynamic elements. Speed and impact lines are used to convey energy in many of the battle scenes, sound effects are enhanced with fleeting text and graphic representations to give them more impact, and intangible forces are stunningly visualized as they become devastatingly tangible for the characters involved.

And these are all enhancements that pull the audience into the alternate reality of the film when they could have been overdone or done poorly and resulted in the opposite (similar to the uncanny valley effect I wrote about earlier). I’m looking forward to learning how the special effects were thought up and pulled off in the behind-the-scenes goodies.

And speaking of graphic effects, we’ll be showing a variety of animated films at an event called Generations of Animation this weekend in The Toby. The techniques used in the films varies from traditional cell drawing and stop-motion to computer generated imagery. We’ll also have hands-on activities in the art labs and three experts in the field demonstrating their skills.

Comment below to let us know what other movie effects you’ve been impressed by lately.

I can only assume that this is part of the IZOD IndyCar Series event going on at the IMA today. The event is for an announcement that will include information about the future of racing with the IZOD IndyCar series and what the next IndyCar will look like. As the guests enter the Museum, they’ll take a trip down memory lane, er Brickyard.

This is part of Jeppe Hein’s Bench Around the Lake which overlooks the White River. Just to the right, across from that stone pillar out in the river, is a USGS stream gauge. You can see the readings from this gauge on the IMA Dashboard. As of this morning, it’s reading about 14.3 feet, which is 2.3 feet above flood level (according to the National Weather Service). I walked down to the surface of the water, and if the river were about 6 feet higher it would come up to the bottom of the bench.

This is a bench that overlooks the lake, and Andrea Zittel’s Indianapolis Island. You can read Michael and Jessica’s take on the flood over at Give and Take. I went down to the water here, and it would reach the bench with another three feet of depth.

The lake is doing a great job showcasing the design of Kendall Buster’s Stratum Pier. When the water level was at its peak last night it was probably submerging the lowest platform. This morning it appeared to be resting right on top of the water.

The biggest surprise came when I glanced over at Tea Mäkipää’s Eden II. The lake had completely submerged the deck of the shack, which is usually accessible via a few stairs.  The water came almost all the way up to the placard. The relationship between Eden II, climate change, and flooding makes this situation poetic, in a wabi-sabi sort of way.

And finally, here’s a little reminder that the path around the lake is not always circumnavigable. Ahead in the water you can see another of the benches, which overlooks Eden II to the right. What you see here is the lake submerging the path. Further ahead there is an inlet which connects the river to the lake, but obviously I wasn’t able to check out the situation there with my bike in tow.

The Monon & The White River

What better place to start than at the Monon trail White River overlook? This is on the first leg of my ride, as I’m warming up.

The Monon & The Canal Towpath

Not too long after that, I turn off of the Monon and start riding down the Canal Towpath. I usually have to be ready to dodge the dozens of ducks and geese that are sleepily wandering about, but I didn’t see many today.

Broad Ripple on the canal

Broad Ripple is also looking pretty sleepy at this point in the morning. Up past this bridge is the blue bridge that crosses the canal next to College. After crossing at the light there, the towpath turns into a gravelly surface (which had puddles this morning) and continues on the south bank of the canal.

Towpath Canal Crossing

This the view from the bridge where the towpath crosses back over to the north side of the canal. We’re getting close to Butler now.

Canal Near Hinkle Fieldhouse

Here’s the sun coming up over Hinkle Fieldhouse. Great game last night, Bulldogs! I sometimes see athletes running across this bridge and down the towpath… maybe I’ve seen some of the guys on the team and didn’t even know it!

Canal near The Christian Theological Seminary

Here’s another bridge where you can see the Christian Theological Seminary in the distance. I’ve been inside and I can report that they have a very impressive organ.

Canal near 100 Acres

A little later on there’s a wooden bridge that goes underneath the Michigan street bridge. The 100 Acres Virginia B. Fairbanks Art & Nature Park is just to the right of the trail up ahead. At this point, though, I’m headed to the museum, so I turn around at the end of the bridge and head up the ramp to Michigan street.

Michigan Street & The White River

This isn’t technically on my ride, but just a little north of the ramp up to Michigan is the bridge that crosses over the White River. As the river bends to the right in the distance its passing by 100 Acres, forming the northern border.

42nd Street Gate

Okay, this one isn’t a bridge or a waterway, but it’s an important landmark. This is the gate at Michigan and 42nd street, which is a nice entry point for bikers.

Woodstock Bridge

This is a little bridge just inside the gate that’s on a road called Woodstock according to Google Maps. You can see Lilly House to the right and the entryway to The Orchard on the left. The greenhouse is also on the right.

Front Lawn at the IMA

I like to ride around to the front of the museum for this view. And now it’s time to get to work!

Yes, those are our planetary neighbors… hold your hand up at arm’s length and they’re just about three fingers apart. Trust me, it’s a whole different experience seeing this with your own eyes. Which you can, but binoculars might help if you’re in an area that suffers from light pollution.

Enjoy contemplating the universe!

The number of events happening at the IMA on any particular day can range from one to ten, and we typically have anywhere from three to seven ongoing exhibitions as well. We didn’t have a chance of plotting all of this information on the traditional monthly calendar, so we took some time to brainstorm a different approach. This sketch contains some ideas from the middle of that process.

After doing these sketches, a little experimenting, and a lot of pondering, we started over from scratch. This time, rather than considering how to display all of this information, we thought about how the calendar should serve the user. Our main goal is to help the website visitor plan a visit to the museum. Most importantly, then, the visitor should be able to pick a day and see what events are happening. We also thought it was important to convey the distribution of events over a number of weeks to enable a browsing mode of interaction, which would increase awareness of our broad array of activities… a win-win scenario.

The resulting design, which also took some inspiration from those initial sketches,  is what you see on the calendar page. The traditional date picker in the upper right works in tandem with the timeline to present the viewer with information about events happening on any particular day. The timeline, which is really just a fancy date picker, displays an icon for each event, color-coded by the event type, over a period of about four weeks. Clicking on the event types in the legend toggles between showing all types or showing a single type on the timeline. The initial scroll animation to the current day is a cue intended to suggest that you can scroll the timeline (three months into the past and a year into the future). This can be done by dragging the timeline or using the black buttons to the left and right. Some of the wilder ideas for features included having the event icons rain down from above “like Skittles” when the timeline loads, and making the Wheel of Fortune click-click-click sound during the scroll animation, but in the end we decided that these enhancements might be a little over the top. Just as in film, some of the best effects are those that you don’t notice. One of the hidden beauties of the page design is that the primary functionality is preserved even if the calendar timeline can’t be loaded on the page (if Flash is not enabled, for example).

Other ways to find events on the website include the upcoming event listing on the front page, the exhibitions page, genre-specific pages like talks and special events, and the individual event pages themselves, which are accessible via site search. You can even see a list of upcoming events and exhibitions on the mobile page.

So, how has your experience been finding events on the new website? Let us know if you have any comments or suggestions.

Photo by Listen Missy! on Flickr

The term actually refers to a phenomenon experienced with robots designed to look human, in which a person’s emotional response to robot design is increasingly positive as the design becomes more human-like, but only to a point. Designs that are very close to human but fall short tend to have an eerie quality, and these are considered to be in the uncanny valley. Both in robots and animation, the response can be different depending on whether the robot or character is in motion.

Let’s start at the safe end. There are many great examples from Pixar where humans are rendered in a cartoon style, perhaps most notably The Incredibles and Up. The artists at Pixar also excel at portraying human emotion and storytelling via non-human forms.

The Incredibles, Pixar Animation Studios, 2004

To come a few steps closer I’ll take a side step into the video game arena, where we have fine examples of anime-style characters rendered in 3D from Square Enix’s Final Fantasy series. There are many fan sites that have screenshots from the full motion video sequences in these games. Being familiar with anime, these characters don’t seem as strange to me as they might to someone else, but I think that the use of this style does place them in an intermediate zone that isn’t quite at the valley, but perhaps approaches it.

When the Final Fantasy brand hit the big screen with The Spirits Within, Square Pictures had done a fantastic job with the digital models, particularly for Aki, the main character.  At the time, though, we still hadn’t perfected the look of skin, and to do that well also requires significantly more computation. I also think this was about the time when we were just starting to develop face tracking algorithms to capture the facial motions of actors, so the lip syncing also was not as good as it is today. These close-but-not-quite qualities are what give viewers an eerie feeling.

Final Fantasy: The Spirits Within, Square Pictures, 2001

I’ve heard some people say that Beowulf didn’t look right either, but I thought it was amazing. I hadn’t seen this movie until it came out for rent and hadn’t read about it much, so as I was watching I didn’t realize that it was an entirely CG film until a minute or so after the camera was focused on the characters in the first scene. After that point the fact that it was entirely digital never distracted me from the story. This makes me wonder if the effect is reduced if you’ve seen enough CG films.

No matter how many CG films you’ve seen, though, I think that the work done by Weta Digital on Gollum in The Lord of the Rings surpasses the valley. State of the art face tracking, motion capture, and rendering algorithms combined in the hands of that team to integrate a highly compelling digital character with a real environment. Although… I guess Gollum is supposed to be eerie, so maybe he’s not a great example.

Gollum, Lord of the Rings Trilogy, Weta Digital, 2001-2003

Avatar showcases what the state of the art can do today. From what little I’ve read so far, the motion capture system used here afforded James Cameron a real-time virtual rendering of the scene, allowing him to adjust the camera and order new takes as if they were working on a real set. There were also improvements to face tracking and rendering algorithms that might make you wonder if you’re watching real actors in makeup. It’s really amazing how far this craft has come.

Avatar, Lightstorm Entertainment, 2009

What do you think? Have you seen other movies that have (or even exploit) that eerie quality, or are great examples of avoiding or surpassing the valley?