Perhaps it's foolish to point out yet another awesome NYTimes infographic, but I found their recent set of maps describing growth and new architecture in Beijing to be both interesting and beautiful. Their Olympic medal history visualization is also cool. Damn them for being so awesome. All the time. </fawning>
Check this out: a site that has historical maps of US counties. It's really cool stepping through time and seeing how the county borders shift and change. Dead simple site, great data. It's a little more interesting for the older states, like Massachusetts, but California has some interesting changes too.
For those of you without at least a passing interest in Martian cartography, Giovanni Schiaparelli was one of the first astronomer's to map Mars using a halfway decent telescope. He drew exceedingly detailed maps of what he saw, depicting massive, linear trenches he called canali. He firmly believed these were too straight to be formed by any natural process, and that they must have been artificially produced by inhuman minds (perhaps even cool and unsympathetic ones). His maps were the state of the art for about 20 years. BibliOdyssey has a wonderful post showing some of Schiaparelli's maps, which are far more beautiful than I had imagined, having previously only seen crude reproductions in 2-tone print. Wonderful stuff. Via The Map Room
"After Google, What?" was the name of a course at the iSchool, but Paul Smith of EveryBlock has asked the same of Google Maps. His answer? An open source web mapping stack. Check out his cool article on A List Apart, "Take Control of Your Maps". If you haven't checked out EveryBlock, do so. Their maps are beautiful. We use a lot of these technologies in the lab and in the GIIF, with the exception of Mapnik, which, to my shame, I've tried and failed to install. Update Some responses: a thread on geowanking including a response from Google evangelist Pam Fox, and discussion and response from the author on ALA (for some reason the post isn't on Technorati yet...).
It's probably wrong of me to find a UNOSAT map of recent pirate activity off the coast of Somalia kind of hilarious, but they actually used the skull and cross bones to iconify pirate attacks, and did, in fact, use the phrase "mother ship." I'm sort of ambivalent about the map as a whole. The spatial distribution of attacks is interesting (why so many hijackings around Mogadishu and Mudug?) as are the narratives, but the cartography leaves something to be desired. Land features get an inexplicable amount of detail and attention for a map depicting strictly maritime activity, and the iconography is almost meaningless (we get it, skulls and bones mean pirates). The colors create thematic associations fairly well, but seem primarily focused on the narrative callouts, which are arguably of secondary important to the locations of the attacks. Probably the most interesting data graphic is the bar chart at the bottom, depicting a (significant?) drop in absolute pirate attacks correlating with changes in government. Via humanitarian.info and Nick
The folks behind the excellent Oakland Crimespotting are giving a talk today at the iSchool. If you haven't seen the site, it shows crime data from Oakland on a map, with different icons for different types of crime, and allowing you to browse through time using a sliding, expandable window over a bar graph. Pretty sweet use of Flash. Here's the talk info:
Design Futures lecture series sponsored by the Berkeley Center for New Media and the UC Berkeley School of Information TODAY Thursday March 20 5:15-6:30pm 110 South Hall UC Berkeley Mike Migurski and Tom Carden, Stamen Design Visual Urban Data: A Journey Through Oakland Crimespotting A talk about the political, social and technical hiccoughs encountered since the inception of Stamen Design’s Oakland Crimespotting project just over a year ago. The talk will cover the inspirations and influences of the project, and how it relates to Stamen’s recent work in web-based information visualization and mapping. About Stamen Since 2001, Stamen has developed a reputation for beautiful and technologically sophisticated projects in a diverse range of commercial and cultural settings. They work and play with a surprising and growing range of collaborators: news media, financial institutions, artists and architects, car manufacturers, design agencies, museums, technology firms, political action committes, and universities.
Great story on KQED QUEST the other day: tactile street maps for the blind. This is about TMAP, a project that essentially aims to make a street map GIS for the blind, including web-based delivery of images that can be printed on an embossing printer, and phone-based request for mailed maps. You can learn a lot about the technical details in this Google Tech Talk given by Dr. Josh Miele (a Cal alum).
If you're an academic, or even a pseudo-academic like me, you need to stay current on the research in your field. You could just go to the library and grab journals off the shelf (boo), or perhaps steal your advisor's copy of Science every week and and peruse it over an Americano at your coffee shop of choice (yay), OR you could harness the awesome power of the Intertron. Well, you could, if there was some kind of killer app to help you out. As it is, there are only a few apps that range from impotent to mildly threatening, but some of them might be useful.
A feed is a machine readable log of frequently updated content on the Internet (you might have heard of feed formats like Atom or RSS). Say you have a favorite website, like the Kelly Lab Blog. Instead of coming back here every day to see if things have changed, you could subscribe to the site feed (or the comment feed) with your favorite feed reader, and check that instead. That way you could check to see if lots of resources had been updated in one place. Many journals that have their digital act together publish feeds of their content (Nature, Science, TREE, PLoS), so you can always know when new content is available, and/or peruse titles and abstracts. Feeds are a good way to keep up to date, but they might not let you do much cataloguing or recommending. Google Reader (my feed reader of choice) let's you "share" items publicly (and with your GTalk contacts), but doesn't let you specify who you're sharing them with (my shared items are mostly just things I find both amusing and not incriminating, but you could use them for papers). You can also "star" things, which is like bookmarking, and there are tags, but stars lack specificity and I feel like their interface for tags could get unwieldy if you use a lot of them. Other feed readers might be different (Bloglines, Firefox, Thunderbird, NetNewsWire, heck even Outlook 2007).
The premier online bookmarking site isn't a bad tool for keeping track of articles. It won't tell you when resources are updated, but you could use it to save and organize articles you find interesting, by bookmarking and tagging them. Perhaps more importantly, you can recommend papers to other people using del.icio.us by using a tag like for:kueda, and likewise receive recommendations. Used in combo with a feed reader, it could be pretty useful. I actually use del.icio.us a lot for web resources in general. For instance, here's my limited collection of GIS data sources (tagged "gis" and "data"): http://del.icio.us/kueda/gis+data
Now we're entering the realm of specialized sites. Connotea is a sort of social bookmarking site specialized for journal articles, created by the Nature Publishing Group. It's got tagging, like del.icio.us, but it also lets you form explicit groups, and it will auto-extract citation information from the Web versions of articles (they even have a video). You could use Connotea to keep on top of what people are bookmarking by perusing tags (e.g. methane), or by subscribing to a tag's feed. I've found some interesting stuff browsing around the tags, but I haven't benefited from any of the social features like groups, because I don't know anyone else who uses Connotea. It also does some decent citation export in different formats, but it doesn't seem to export in different styles (Chicago vs. APA, for instance). citeUlike is another site very similar to Connotea, but I don't have as much experience with it. Also worth mentioning is Zotero. Zotero is an awesome plugin for Firefox that helps you manage citations. It's a local app without any social functionality (yet), so it's not really for keeping up to date, but it is great for extracting and exporting citations from papers and web pages. I've used it a bunch for writing papers. When it gets some network functionality it will be truly amazing. Ok, that's it for now. Anyone have any other favorite tools for keeping on top of academic literature?
The use of the Web and other networked technologies for disaster response is a hot topic, especially in nations like the US where such technologies are prevalent. Response on the Web has varied from insistence that everyone should use Twitter in disasters, to problems with signal to noise in crowdsourced search and rescue. Since maps and geospatial data play critical roles in disaster response, we spent a lab meeting discussing how web maps and webGIS have contributed. Read on for the collection of links and random thoughts.
Tim walked us through the US Holocaust Memorial Museum's Crisis in Darfur Google Earth layers, which feature photos, stories, and the locations of destroyed villages. The purpose seems to be to inform the population of Google Earth users about the extent of the crisis, and channel them toward the responses recommended by the USHMM.
We also discussed the imagery of Katrina's aftermath that was collected and shown on Google Earth very soon after the storm (and has been suspended at the time of this writing). This NYTimes article suggests the effort began with Google Earth users assembling imagery that NOAA had collected and released, but doesn't mention any data collected at Google behest. Forbes also has a good article. Several mashups, including Scorpionus.com, played a role in the response, and there was a apparently a significant effort by the open source GIS community to collect and distribute imagery, though I haven't seen much media coverage of it or any evidence of its use [please pass it along if you find some].
We've talked about a few maps made in response to the fires on this blog, including maps by
None of us knew whether any of these maps were used by people on the ground, which seemed like one of the purposes of the Google Maps, as they were created while the fires were burning and seemed to be regularly updated with fresh data. I would also argue that one of the most important maps of the fires was the MODIS imagery of the smoke plumes that made it into a lot of the popular news, at least in terms of communicating the magnitude of the disaster. Shasta also showed us maps made by NorthTree Fire International. I also talked briefly about the use of Twitter in disseminating information about the fires. Individuals and the Red Cross were both using Twitter to publish updates about the fires that could have been used in real time response, but it's still unclear how much these sources were actually used.
There have been a few interesting maps made in response to the oil spill in the Bay, including
Web responses to disasters take many forms, and serve several purposes. One interesting trend is the increasing ease of publishing geodata about disasters on the Web. Ten years ago, only governments would have the resources to quickly collect and disseminate imagery and ground-based data. With Katrina we saw a philanthropically motivated private sector achieve similar results based on existing business relationships (Google and imagery collection companies) and existing data infrastructure (Google Earth, Google Maps). Now we are seeing organizations with even fewer technological resources collecting and distributing data using mashups and web mapping tools like Google My Maps. There are even individuals who have made potentially important contributions during disasters (e.g. Nate Ritter during the 2007 SoCal fires, and Kathryn Cramer during Katrina). One issue that I find important in all these mapping efforts is that of trust. How much do you trust data from NOAA, Google, your local public radio station, or some programmer obsessed with microblogging? How useful is a large, crude polygon showing oil slick sightings?
Nick, a classmate of mine, pointed out this interesting use of Google My Maps for disaster reporting by KPBS News in San Diego. KPBS is also maintaining a Twitter stream for updates about the fires, which is the most useful potential use of Twitter I think I've seen. Has anyone seen any fire mapping systems on the web that integrate discussions? Requests for missing people, perhaps requests by evacuees for ground info on their homes?
Y'all might be interested in this. Hell, I might even go. Via the geowanking list,
A quick reminder: WhereCamp is being held June 2-3 at the Yahoo! Sunnyvale campus. There'll be sessions and whatnot. You might also consider it a Geowanking list meet-up of sorts. Description: WhereCamp un-conference. Free. A hugely energetic overlap of diverse interests ranging from newbies, web 2.0 and mobile developers, social place hackers and artists to grad students, geographers, earth scientists and people focused on humanitarian and environmental efforts. Wiki: http://wherecamp.pbwiki.com/WhereCampSF Directions: http://docs.yahoo.com/info/address/ Registered Attendees: http://wherecamp.pbwiki.com/WhereCampSFRegistrants and: http://upcoming.yahoo.com/event/170065/ See you there! -- Joshua Schachter http://del.icio.us/joshua http://joshua.schachter.org/
After I saw this: I created a specialized, simplified, hacked up version on giif, that is pretty useful. the simplest way to see its use is by going here (ken-ichi, i'm leaving the link text == href intentionally): http://firecenter.berkeley.edu/proxy/geoservice.py/geocode/2134%20Allston%20Way,%20Berkeley,%20CA,%2094704 that is the lat/lon of CanCun Burritos. If you prefer your burritos in UTM, you can request by appending ?epsg=26910 to the end: http://firecenter.berkeley.edu/proxy/geoservice.py/geocode/2134%20Allston%20Way,%20Berkeley,%20CA,%2094704?epsg=26910 or any valid EPSG code that GDAL on giifweb will recognize. (if you dont know what EPSG is, you should consider freeing yourself from your dependency on ESRI products.) likewise, reverse geocoding: http://firecenter.berkeley.edu/proxy/geoservice.py/reverse_geocode/-122.2584/37.8675 and it may return you a list of info. depending on wether geonames.org can reverse_geocode your points. again that assumes lat/lon, you can specify an EPSG code by appening ?epsg=26910 to the end of the query. it's useful to have this built into a webGIS. where,perhaps, the user enters an address and you move the map to that lat/lon. the script will generate an example if you browse to: http://firecenter.berkeley.edu/proxy/geoservice.py/example if you enter an address and an EPSG, it should alert the lat/lon, and then try to reverse_geocode and alert that derived place. if you view the source on that page, you can see how to include the necessary javascript: and if you browse to that url, you'll see the lovely javascript that is generated. Since AJAX is used, you'll either need to download the source to somewhere that has GDAL and web.py installed, or be on giifweb. to use the javascript, you need to attach an event to call the geocode() javascript function:
document.getElementById('button').onclick = function(){
geocode(document.getElementById('address').value, 4326);
};
where geocode() generates a url and makes the AJAX request to server to do the geocoding for you. again, the 4326 is the default, optional EPSG, to use other projections, just put that as the 2nd argument. then, when the result returns from the server, it geocode.callback is called with the resulting JSON: so you can define your own callback like: geocode.callback = function(result){ // do something with the result. document.getElementById('x').value = result.x document.getElementById('y').value = result.y }; where result.x and result.y correspond to the x and y properties here it's simpler than it sounds, best to view and read the source of the example. or better yet, just make the python script print itself. in all places in this post where it says firecenter.berkeley.edu, you can substitute giif.cnr.berkeley.edu or giifweb.cnr.berkeley.edu, and it will work as it should. and you will have to do this if you use the javascript/AJAX because of cross-domain issues. sorry i dont know how to format html properly on this blag.
The International Diabetes Federation is sponsoring a worldwide geocaching event to promote diabetes awareness. 20,000 'travel bugs' are moving around the world in support of the UN resolution to promote diabetes awareness, prevention, and care. IDF’s goal to reach 1 billion people with awareness messaging about diabetes care and prevention.
Check out KQED's beautiful website: QUEST
.
Several researchers at Stanford have written some software for visualizing flow maps. There pictures are very pretty, and, I think, good data vis. The map above shows the top ten states providing migrants to NY and CA. Here's their abstract:
Cartographers have long used flow maps to show the movement of objects from one location to another, such as the number of people in a migration, the amount of goods being traded, or the number of packets in a network. The advantage of flow maps is that they reduce visual clutter by merging edges. Most flow maps are drawn by hand and there are few computer algorithms available. We present a method for generating flow maps using hierarchical clustering given a set of nodes, positions, and flow data between the nodes. Our techniques are inspired by graph layout algorithms that minimize edge crossings and distort node positions while maintaining their relative position to one another. We demonstrate our technique by producing flow maps for network traffic, census data, and trade data.
I thought that you might enjoy this ... http://news.mongabay.com/2006/1114-google_earth-act.html
This is on Friday at 3PM down at Y!RB down on University. Here's the Upcoming.org link, and the abstract:
My exploration of spatial and temporal information dissemination started form the observation that a piece of information could have different value at different location and time, and just-in-time information delivery can be essential in achieving information dissemination efficiency. The exploration resulted with a few new information dissemination strategies that gear towards distributed applications that handle location and time sensitive data. A unified theme among the approaches is making space and time the first class citizen in networking and information dissemination. In this talk I will start from discussion of my past work in this direction in mobile and sensor network context, and discuss potential ramifications beyond the networking context (if time permits).
Flash Earth is a really cool in-browser Flash implementation of, well, basically, Google Earth. I am too lazy to post an image right now, but it's very awesome. You can also choose from a list of imagery sources.
Swivel offers free, public and private, data hosting. You can create nice graphs with your data but also search for other people's data. For now this is not entirely geospatial, but when the api is opened up or they update the service, a geospatial component could be added. For now I would seriously question the reliability and accuracy of any data posted there. Below is an example of a graph made with Swivel.
I just found out that there is a Mercury solar eclipse happening today! http://sunearth.gsfc.nasa.gov/eclipse/OH/transit06.html http://sohowww.nascom.nasa.gov/hotshots/ Wow - Mercury solar eclipse and Rumsfeld stepping down - what a great day!