Andrew Zolnai Blog

Top tips for engaging online communities

aizolnai@gmail.com (Andrew Zolnai) — Tue, 27 Jul 2010 11:06:00 +0000

From the Press Association PR & Communications Newlsetter today:

1. The trick isn't a trick - be useful, or be helpful or fun. These tools are incredibly powerful, but they are not another way to shout at people

2. Listen and respond

3. Be honest and be human

4. When dealing with a crisis: listen to what is being said collectively, formalise that into a list of questions and answers, put it online and point people to it

5. Empower individuals to be individuals within the umbrella of your brand/organisation

6. Try to create a safe space for people to learn that behaviour, an internal wiki or something similar

7. Support people and review your policies regularly

8. Talk to people they will help you

9. Never drink and tweet!

10. Don't be scared

May you negotiate peacefully the intricacies of being online!

in the Golden Pavillion garden, Kyoto Japan, 1967

Trending oilelefant.com, Part IV

aizolnai@gmail.com (Andrew Zolnai) — Mon, 26 Jul 2010 14:37:00 +0000

Slideshare traffic figures show the same trend as web traffic figures posted before: monthly readership (calculated as total reads over months posted) increase from my previous papers, through Interactive Net Mapping business processes, to those on the web and social media.

This reinforces the previously posted message:

social media increased traffic back to oilefant.com over web media alone
web and traditional media are still important in conjunction with the new

Also updated my manuscript Trending Social Media also posted on Slideshare:

Trending Oilelefant

View more presentations from Andrew Zolnai.

The power of context, Part V

aizolnai@gmail.com (Andrew Zolnai) — Fri, 23 Jul 2010 16:50:00 +0000

The tropical storm threatening the Gulf of Mexico (GoM) oilspill cleanup shows up very well in Google Earth (download GE here) by simply using:

EPA's Deepwater Horizon - Gulf Spill Response KML shown previously
Google Earth latest release's legend that comes complete with ocean data

Even though the EPA stopped posting the extent of the slick on 15 July, a daily-updated spill extent KML can be found here, courtesy of NOAA-NESDIS (thanks Peter Giencke). Regardless GE also shows:

shipwreck locations
marine protection areas
endagered species
& other enviro-data

Most importantly today, it shows NOAA's cloud and radar data that can be followed by the hour by simply refreshing the screen - I traced it today from the Florida Panhandle for example.

Isn't this is a mash-up par excellence, bringing together host of related data, some out-of-the-box, and others for free from easily reachable goverment site? And surely anyone can easily remember www.epa.gov/BPspill :-)

Update: not only does NOAA-NESDIS KML mentioned above have data thru 20 July, but this KML animates it directly in Google Earth; compare this recording to my recording of EPA screenshots or Ruth Lang's excellent animated mashup:

The power of context, Part IV

aizolnai@gmail.com (Andrew Zolnai) — Sat, 17 Jul 2010 10:07:00 +0000

Two anecdotes on remote sensing and environmental monitoring highlight some issues in measuring and predicting the current Gulf of Mexico oil rupture.

Last year as a consultant at AGIP KCO to Tony Battle, I spent some time with ice specialists in the north Caspian Sea. It freezes over every winter, and both navigation and oil platforms must be monitored for the safety of the environment as well as the infrastructure. But remote sensing specialists assured me how difficult it is to separate roughness in the water - such as white caps we see on any lake in brisk wind - that may also be caused by ice, oil or loose vegetation. Ground-truthing is a must through vessel observations, buoy measurements etc.

And that had been made patent, when Hurricane Katrina struck the Mississippi delta in the fall of 2005. Below is one of then US Department of the Interior Secretary Gail Norton's pictures from her visit in its aftermath (posted with permission). That aerial view of the river shows a different hue of brown/green working its way downriver shortly after the hurricane landfall. The press immediately suspected an oilslick - nothing like this spring's Macondo oilspill offshore, but a slick nonetheless - and that was not an unfair supposition: many oilwells aren't connected to a pipeline network, either because they're too old and/or they produce too little (stripper wells), or because the marshy environment made pipeline laying hazardous (more so in the past); production is thus collected in oiltanks, which may easily have been dislodged if not ruptured during the hurricane, by water pushed up the Mississippi by the storm surge, or by subsequent flooding from the sudden increase in rainfall.

click on image to see photo gallery

It's those two last events, however, that point to the real cause of the apparent slick observed from the air. David Gisclair of the Louisiana Oil Spill Coordinator's Office (LOSCO) / Office of the Governor told me at the time that ground crews were dispatched immediately. They found that the alleged slick seen from above was in fact marsh grass that had been torn up by the storm upsurge, then washed down in the subsequent flooding... marsh grass is anchored to sustain downstream not upstream currents, much like levees were built to sustain outward pressure from the river not inward pressure from the storm surge! Again, ground-truthing was a must.

[nearby Texas gulf coast a year earlier in the banner of my old website]

Fast forward to a space-shuttle view of the current oilspill in the Gulf of Mexico: a video of screenshots from the US EPA / Coast Guard's Deepwater Horizon Gulf Spill Response KMZ in Google Earth (link at end of video) from 11 May to 15 July:

best viewed on YouTube in full screen mode,
click lower right-hand corner then press F11

Notice as you step through this day-by-day (YouTube helps read the dates to the left):

first the spill spreads offshore, spelling fear of it reaching out of the Gulf into the Atlantic and perhaps the Gulf Stream
then it hugs the eastern shoreline, prompting the deploymeent of protective booms in red featured in the news
and it spreads far west along this shoreline (faint tracks to the west), perhaps a distant effect of Hurricane Alex
to start shrinking (?) toward the end perhaps as an effect of well capping, or offshore cleanup or wave dispersion

This is of course at a very macro scale, by no means the only place to see this as posted earlier, and relies on EPA's science and other agencies such as USGS. And the outcome of these events are just starting to be felt, such as passing the CLEAR Act in the US.

A bit of GIS history

aizolnai@gmail.com (Andrew Zolnai) — Mon, 12 Jul 2010 10:03:00 +0000

As I watch ESRI's 2010 User Conference remotely - green or what?! - via social media and my favorite bloggers like spatialsustain and spatiallyadjusted, this entry from allpointsblog caught my eye:

Not only was Roger Tomlinson a delightful companion to a few Russian meetings in Moscow and Tyumen about five years ago with ESRI (see travelog image atop my blog), but as a Canadian geomatician myself I kind of grandfathered him (little did I know I'd end up near his birthplace soon after). I myself started GIS in the resources sector in 1986, and in those days you sought any help that was far and few between indeed, and you guessed that he was among them (his brief history in Wikipedia and the Toronto Globe and Mail). To see what GIS looked like in 1967, skip to the third video:

Resources issues

Computer mapping

Early GIS

These YouTube videos are "A look at the Canada Land Inventory Geo-Information System, a computer system used to help manage and develop Canadian land". The issues feature in the first video with Tomlison himself appearing at the end, the processes in the second video and the GIS system itself in the third.

These are intriguing videos in that it shows what departmental GIS:

looked like over forty years ago - video terminals, teletype, punchcards, female clerks and male techs & managers etc.
were set out to do right at the begining - to integrate land use, finance and demographic data, thus allowing to query real-world business and geographic scenarios
intended to educate the public of all that is invloved in this, and in plain if long-winded language, complete with Telstar-like space sound effects!

Footnote: that system was almost lost but survived in the Canadian Soil Information System.

The power of context, Part III

aizolnai@gmail.com (Andrew Zolnai) — Thu, 08 Jul 2010 15:25:00 +0000

Ruth Lang created this excellent mashup for the 2010 FIFA World Cup - which coloured all the countries at the beginning and gradually greys them out as they are eliminated - only Spain and the Netherlands are left as at today. This was a labour of love, needing for example some tweeking to work in the Firefox browser, and a testament to SVG.

click on map to go to web app

I happened to be on Geocurrents RSS feed, which alerted me to this post also on the same topic.

click on the image to go to web page

Toward the bottom, however, there's a spreadsheet with some more geodemographic data in spreadseet format keyed by country. Presto! I upload that to Google Fusion tables of which I had some already, and I created another mashup in a matter of minutes.

click on map to go to web app

Needless to say there's hardly a comparison between these two maps, one is a polished product (see my previous post on this topic for another) and the other is quick-and-dirty.

But it does show how cheap and easy it is today for anyone to quickly visualise business statistics on a simple map...
In another example, neither GFTs nor other web apps really work that well: compare this GFT

click on map to go to web app

to my previous posting comparing web apps somewhat like here

click on map to go to web app

and my earlier posting on the distribution of Olympic host cities

click on map to go to web app

Is this also not an example of disruptive technology? One painstakingly creates a map, then another gets a slightly augmented dataset on same, only to instantly create a mashup on more data?

Trending oilelefant.com, Part III

aizolnai@gmail.com (Andrew Zolnai) — Thu, 01 Jul 2010 13:18:00 +0000

The effect of trade shows and publications as well as social media can clearly be seen in the web traffic figures through June 2010.

The short term plot peaks at releases and trade show for both internet-only and social media. Trends differ however in being flat for internet-only but rising markedly for social media.

In the longer term, the raw hits (not separated by medium) also peak at various events. This attests to the effectiveness of getting the word out in trade media as well as electronic media.

Have you seen also Part I and Part II of trending oilelefant.com?

The power of context, Part II

aizolnai@gmail.com (Andrew Zolnai) — Sun, 27 Jun 2010 17:22:00 +0000

In preparation for tomorrow's TEDxOilSpill meet-up in Cambridge UK, let me highlight two among the many, many postings on the Gulf of Mexico oil spill (I also wrote about here and here).

One is Ruth Lang's animated map in SVG using FOSS webware:

click on image to go to website

Another is ESRI's timeline map on their disaster response page:

click on image to go to website

My previous blogpost was a very simple depiction of a local archaeological site as seen through various web media. The two animations above extend the power of context not insignificantly: A picture is worth a thousand words, a map ten thousand and so on for the animations here and the video tomorrow at TEDx... Indeed the patent differences between two maps above surely underscore the power of the assumptions and parameters used in each.

That is why I helped the radio media report on current affairs in the industry, because it is so important that the voice of opinions be as well informed as they can. That is also why I posted that first blog in the first place, to try and help us sort through the sensationalism this incident provided. And why TEDx holds oilspill meet-ups around the globe to help us understand and contribute: crowd-source facts to help high-grade information and improve our knowledge.

Update: Here's a video of NOAA's Gulf Response webmap screenshots from 29 Jun to 15 July, showing the somewhat receding extent of the oil spill - the slight increase coincides with Hurricane Alex, and gives it a somewhat pulsating aspect - this shows how difficult it is to predict its extent over time.

view this on YouTube directly

The power of context...

aizolnai@gmail.com (Andrew Zolnai) — Sat, 26 Jun 2010 10:47:00 +0000

... can be seen in a simple weekend exercise: look up your favourite area on your handy 3D online maps, and you just might get a slew of features, some of it unwanted...

On a recent walk near where I live in Cambridgeshire, I stumbled across this feature on Iram Drove, Willingham: it's an Iron Age settlement or a medieval fort, bisected by a recent path.

Bing Maps Birdseye view shows this best. Although the vintage of the satellite imagery or aerial shots is unknown, the fort ring as well as ridge-and-furrow show well.

Hint: rotate clockwise and anticlockwise to see it better
(if you rotate twice and revert to map mode,
refresh the browser to return to the birds-eye view).

click on image to access web

ArcGIS Explorer Online's map, without adding basemaps in this area, shows in shadow relief a circle next to the pushpin.

click on image and then search for:
iram drove, willingham, cambridgeshire

Search for the same on Google Maps and you'll find a 'teapot spout' on the eastern edge - that is simply a drainage ditch to empty what remains of the moat, in this water-logged region.

click on image and then search for:
iram drove, willingham, cambridgeshire

Note however the a-contextual search Google did: it posted an unrelated link, because the word iram has many, many meanings.

bp oil rupture, Part II

aizolnai@gmail.com (Andrew Zolnai) — Wed, 23 Jun 2010 16:44:00 +0000

Here the latest from bp itself at the current NOC show in London.

video flash player

The go-to websites are the joint response site as well as NOAA's. But the statement that follows adds to Drew Stephen's view from the frontline posted earlier, including the link to bp's own webmap below.

The statement of Marcia K. McNutt before the Committee on Appropriations on May 27, 2010, provided a brief yet in-depth behind-the-scenes view of a key federal agency in this event. Let me highlight three points she made one week after the disaster:

USGS Science: From Response to Recovery

Around 22,000 people are following the effort on Facebook; 4,700 people on Twitter; and 38 videos are available on Youtube with one video having almost 1 million views

Estimating oil volume

This is the first time AVIRIS has been used during a large oil spill, but the technology is widely recognized and has been successfully applied in other instances; for example, to map debris from the September 11 event, to map locations of naturally occurring asbestos, and to map liquid hydrocarbons on Saturn’s moon Titan.

Geospatial Support

The USGS has provided historical images and nautical charts of the Mississippi Delta and Louisiana coast to the FWS, the USCG, the U.S. Navy, and BP.

Imagine what this might have looked like in this fictitious meeting.

Photomontage: mock White House security briefing
Hayward & McNutt look on, Obama rolls up his sleeve

Note: part I is here, and here's an update thanks to the Map Room:

"Ted Weinstein sends along a link to this animated map showing the spread of oil rigs in the Gulf of Mexico since 1942:

The video’s author discusses the maps and data here and here."

Presentations and social networks

aizolnai@gmail.com (Andrew Zolnai) — Mon, 21 Jun 2010 13:22:00 +0000

I posted and saw posted a trio of presentations, so perhaps it's time I related them to you.

Digital Energy Journal / Finding Petroleum published my piece Better Metadta for GIS, following on my posts on that same topic, also posted on LinkedIn and its groups I joined:

click on image to go to link, then go to p. 25

Earlier I was asked to do a short review of two of my after-hours projects:

on historic Fenlands mashups I posted it here on Slideshare:

East anglia fenlands

and on near-real time mapping of ash cloud dispersal, also posted on Slideshare:

Near real time ash cloud

bp oil rupture

aizolnai@gmail.com (Andrew Zolnai) — Thu, 17 Jun 2010 06:38:00 +0000

News swirling around the incident in the Gulf of Mexico (GoM) compels me to draw out some salient facts and gather them here to help the uninitiated (I did this yesterday for my in-laws). This blog title is from the last slide of the video in my last blogpost.

A roughneck almost 35 years ago in the Alberta, western Canada foothills on a service rig, I witnessed its burning by a casing gas kick (sudden increase in pressure outside of the drill pipe), and on another deep gas exploration well it took 12 hrs to trip out (lift 10-15,000 ft. drill pipe to change the drill bit). A geologist at Shell and the Geological Survey of Canada until 25 years ago. Ran a workshop with CBC Radio reporter Des Kilfoil to help improve industrial news reportage. In the petrodata management and GIS service sector ever since, one highlight while at Halliburton was a Y2K project at then BP Amoco in Sunbury UK and witnessing Lord Browne rebrand bp.

No longer linked to any party in this incident, here are comments not criticisms, neither referenced nor documented in detail as that has been done in the media. Let's take these one by one, shall we?

deep offshore GoM drilling is an outer limit frontier
lease permitting had been managed by a single agency
disaster response mixed public and private sectors
economic and political backdrops remain fluid

Drilling at almost a mile water depth raises significant challenges in disaster recovery - ROVs and vidcams notwithstanding, attempts to snuff the blowout were all experimental as reported in the press. And drilling a further couple of miles below a continental margin encounters very high pressures and temperatures that are not yet routine operationally. The only exploration and production analog is offshore Brazil, which also had its share of incidents.

The Mineral Management Service (MMS) managed the offshore petroleum exploration leases, as well as the environmental impact and industrial safety. As reported in the news that can present conflicts of interest, and these functions are now split among different agencies. Also lease auctions garnered enough revenue to almost single-handedly fund the US Department of Interior, and financial success didn't encourage close scrutiny.

Terrorism, hurricanes, tornadoes and earthquakes invoke disaster responses from local through state to federal agencies - coordination has been a lesson in improving inter-agency communication - and GIS has played an increasing role. As mentioned toward the end of my second-last blogpost, this incident however invoked both public and private sectors - coordination has neither been planned nor expected - and the only analog at least in scope, the Exxon Valdez incident was too far removed in time and geographic setting to be applicable.

The political backdrop is also murky. For example the current excitement for shale gas - new domestic reserves of natural gas and close to US markets, but costing twice as much to produce than to sell, and demanding huge amounts of water from aquifers already straining to meet urban demand - underscores the risks both industry and society will take to secure domestic supply. Or just as Outer Continental Shelf and Arctic National Wildlife Refuge moratoria on petroleum exploration were being lifted, drilling is shut down in the GoM. The road ahead is far from being straight...

Update: Oil&Gas Journal announced on 16 July:

The US House Natural Resources Committee July 15 approved by a vote of 27-21 the Consolidated Land, Energy, and Aquatic Resources (CLEAR) Act, HR 3534.

Introduced by Chairman Nick J. Rahall (D-W.Va.) in September 2009, the CLEAR Act addresses the Deepwater Horizon oil rig explosion and spill in the gulf as well as implements reforms in the country’s offshore and onshore oil and gas leasing program.[...]

The Big Easy button

aizolnai@gmail.com (Andrew Zolnai) — Tue, 15 Jun 2010 20:45:00 +0000

Building geoinfo from the ground up is patent in this Gov2.0 presentation on citizen-focussed geoweb at the Greater New Orleans Community Data Center toward sustainability in the Big Easy:

... when you take a user-centered approach, no matter what challenges come your way, you always know where to start: who's the audience, what decisions do they need to make, what data is available to support those decisions.

Also from the ground up is Stack Exchange Area 51: a user community to foster geoknowledge on the web. Its bottom-up pragmatic approach builds a geohelp site as and if, and only if, the community demands - signups will move it from commitment to beta stage, and the content will be shaped by votes on-site.

click on image to access site

The natural resources sector has had openness and the web for well over a decade or so, a lifetime in geo computing - to name only the ones I'm familiar with in alphabetic order: British Geological Survey's Open Geoscience, Natural Resources Canada's Geobase and US Geological Survey's National Oil and Gas Assessment - yet are we not all still seeking big easy button solutions?! (button courtesy staples.com)

So, did you want that with a map on the side? Just released US National Biological Information Infrastructure's GAP map has that!

click on image to access site, then select Land Cover Viewer

Did you want it à la carte? UK Ordnance Survey OpenData has it too!

click on image to access site

If you want them stacked like pancakes, look no further than data.guv.uk Combined Online Information System (COIN):

click on image to access site

And there could be a lot of scope to combine the last two, judging by traffic on the UK government data developpers google group (free login required). Let's see if the Local Data Panel (see blog) include all things geospatial, as say, in the OpenlyLocal hyperlocal sites:

Professor Nigel Shadbolt [...] is chairing a panel of experts to oversee the release of this local data. The Local Public Data Panel will champion the release of local public data and information sharing, accelerate progress in agreeing common standards for data released into the public sphere, and for making local public services better understood and more accessible.

For example searching near Cambridgeshire revealed a blog posting Cambridge City Council items, such as their new online planning system, which was hard to find via the official website.

click on image to access site

The drummer and the dancer

aizolnai@gmail.com (Andrew Zolnai) — Fri, 11 Jun 2010 21:46:00 +0000

Drums closing the FIFA World Cup kick-off celebration brought back my African drumming days in N Texas and S California:

drummers check each other out to stay in time and in tune, and to offer solid backup to the dancers in the center. That creates an energy flow that according to west African tradition pushed energy down from the drum into the ground, then rose up through the dancers feet, spread into the audience whose enthusiasm charged the drummers, and so round&round

Drum camp in San Bernardino Mountains

Project management should work like that too - the back office data providers give the beat of geospatial data crunching, and the front office data interpreters do the dancing GIS interpretation, but agencies and public alike judge the mashups and presentations. This process has been abundantly documented elsewhere in the case of major incidents:

a command center in the Javits Center in the immediate aftermath of the 9/11 collapse of the Twin Towers had a very simple yet effective workflow: data and reports entered the building on one side, GIS pros lined up there updated the maps and fed the data to a line of servers amidship, and a line of plotters along the opposite wall delivered prints to the emergency staff at the opposite side of the hall, ready to take maps to the rescue site

from MTI Report 02-06 on 9/11

a few years later during the devastating fires in southern California, a command post at the foot of the mountains simply took data from the returning firefighting crews, data were input and processed on the spot, and the next crew could fly out with up-to-date data in their handhelds to the field - fires pushed by Santa Anna winds spread so fast that it was crucial that helicopters drop crews or water on the right side of the ridge

ESRI on fire response

The MTI report above was in part responsible for the US National Incident Management System NIMS protocol. This was initially led after the Deepwater Horizon incident by the then GIO Drew Stephens - his website logs in fascinating detail the in&outs of that incident response. All emergency management is difficult, but the Macondo well response mixed local and remote plus private and public concerns, and therein lay the rub. While the Minerals Management Service managed offshore leases very hands-off, incident response on the Gulf coast was very hands-on.

15 Jun update: Drew Stephens restored the post citing concerns, added bp's oilspill webmap, and told me on 13 Jun that data access issues are being worked out...

Until the dust settles, Shakira's Waka Waka lyrics for this FIFA opener may be of help to all parties involved:

You're a good soldier
Choosing your battles
Pick yourself up
And dust yourself off
And back in the saddle

You're on the frontline
Everyone's watching
You know it's serious
We're getting closer
This isn't over

The pressure is on
You feel it
But you've got it all
Believe it

When you fall get up
Oh oh...
And if you fall get up
Oh oh...

full lyrics on the gather site

The stunning beauty of maps, Part II

aizolnai@gmail.com (Andrew Zolnai) — Wed, 09 Jun 2010 19:17:00 +0000

The beauty of maps is topical not only thanks to British Library's show Maginificent Maps. Gary Gale waxed rhapsodic on map as art, and Thierry Gregorius mused on what a map might look like after visiting the same exhibit.

Perhaps even more intriguing are the maps that didn't make it into the British Library show: Smith's original geologic map of Great Britain laid down principles of stratigraphy and geology not widely accepted in 1815. And the (in)famous Red-Line Map separating the UK dominions from the then-new United States, remind us that the CDN acronym for Canada stands for Commonwealth Dominion of North-america.

I myself oscillate between scrolling through utilitarian mashups posted here, and flipping through the pages of An Atlas of Cambridgeshire and Huntigdonshire History (Anglia Polytechnic Press: 75 maps with notes on everything you wanted to know but were afraid to ask about culture and history of East Anglia since Prehistory).

Should we let the data speak for itself, provided it is posted clearly and succintly as in that book? Or should we use the full complement of web mapping tools such as Marten Hogeweg's at the risk of cluttering it up with map bars?

Isn't it ironic that ancient maps filled empty spaces with fanciful illustrations and monikers like here be dragons.

The British Library traces the evolution of mapmaking in its collection: from Christian views of heaven and earth in the Middle Ages, thru political tools of knowledge and posssesion in the Age of Enlightenment and propaganda in the Industrial Revolution, to whimsical portrayals of one's own city today.

Cartography then as GIS now were never easy tasks, especially when it came to projecting spherical data onto a flat map! And marrying the role of surveyor and propagandist then, must not have been any easier than adding GIS expertise to your profession today - I'm a geologist by trade, yet my early exposure to geodata brought me the strangest stories... But one of the stellar examples I just came across (thanks @mattartz) hints at Martian geomorphology.

Now if Carolyn Porco's team mapped in more details the stunning imagery of wetlands and methane lakes on Saturn's moon Titan...

Trending oilelefant.com, Part II

aizolnai@gmail.com (Andrew Zolnai) — Mon, 07 Jun 2010 15:11:00 +0000

Recent web hit statisitics show the power of press releases (PR) in Oilvoice and FindingPetroleum (née Digital Energy Journal) to push traffic to oilelefant.com. The occasion was the new release of a secure trial area - try before you buy - to show on the internet what it will look like on the intranet.

short-term web hits with correlated events

long-term web hits with correlated events

This also shows new media (blogs, RSS etc.) and traditional (PR, internet etc.) work hand-in-hand.

And this is only as preamble to actually meeting prospect and client requirements. As discussed here before, these are very briefly:

in v. 1: ease-of-use and affordability
in v. 2: EDMS in addition to E&P
now also with live trial online
in v. 3: bulk data import/export

Watch this space for further updates on how we work toward this. Hint: we'll continue using petrodata standards such as PPDM as a backbone, and leverage metadata as mentionned earlier.

Gathering clouds over the horizon, Part IV

aizolnai@gmail.com (Andrew Zolnai) — Sun, 23 May 2010 13:19:00 +0000

Slotting straight into my previous post, ESRI just released ArcGIS Explorer (AGX) online and arcgis.com as the new ArcGIS Online (AGO). I cannot, however, see my old AGO postings on the new arcgis.com, neither will the new AGX online consume it - not even posted as a web mapping service (WMS)!

Does arcgis.com consume only ESRI map services?

Until that is fixed we're back to AGX desktop, which does find my WMS at least.
http://www.giscloud.com/wms/0d198f438107df1fa4bd3ecf3b1ed6c6
Enter this under GIS Service, and you will see what I published after hours. We're still left, however, with performance and graphics card issues reported in my previous post... stay tuned!

PS: the Medieval Fenlands items above are now also indexed on UK government data website here.

Gathering clouds over the horizon, Part III

aizolnai@gmail.com (Andrew Zolnai) — Sat, 22 May 2010 11:22:00 +0000

Eyjafjallajoekull volcanic ash blown southeastward caused air traffic disruption last week over the northern British Isles again. I post the North Atlantic section of NOAA's free web mapping services of global cloud and chemical composition:

false colour images give the space-shuttle view of cloud cover
SO2 (sulfur dioxide) accompanies hard-to-see ash clouds (below)
CO (carbon monoxide) is also associated with volcanic eruption

The video shows this week's false colour imagery over the North Atantic (it can be seen on YouTube here).

The quick view below can be seen on a full giscloud map here with all three datasets, LIVE: Step through the previous days layers by turning them on sequentially from the bottom up.

To repeat, these live data feeds give you current and previous coverage against other public vector datasets off the web.

Last week I posted this as ArcGIS Explorer (AGX) document on ArcGIS Online here. The cover snapshot shows the ash plume that ground traffic then. Note however that:

AGX is a desktop plugin and therefore slow on some machines
- my home notebook with old video drivers won't post the WMS services
- issues reported even on Windows 7 also seem to relate to video drivers
it shows WMS, KML and ESRI layer file attributions, but not shape file styling
ArcGIS Online only distributes to ArcGIS, go here to get AGX file it directly

Update: I posted a short paper on this topic

Near real time ash cloud

View more documents from Andrew Zolnai.

Gathering clouds over the horizon, Part II

aizolnai@gmail.com (Andrew Zolnai) — Sat, 15 May 2010 19:44:00 +0000

I posted earlier an example of a simple Google Map reading a list of Olympic cities - this was part of a list of Google and OpenLayer maps I built over the years on my old website - I also posted a nifty display of plate boundaries in Google Earth here. Following on the discussion of new web mappping (Part I of this post), I recreated both on giscloud.com to see how the two implementations compared.

The Olympiads map does not show the drop-down list allowed in Google's Java API - then again I haven't used the giscloud API. Blue dots show the winter games, orange dots the summer games, and in giscloud you get details including weblinks by clicking on the cities.

For the tectonic plates I went to the source USGS Open File report including the volcanoes - lat/long sign conventions and KML files were a challenge along the antimeridian (lines crossing the date line), but a clean-up posts them right on a Google Map overlay.

Here it is as WGS84 vector world map with a Blue Marble backdrop.

Please note that Google maps are all raster maps in the web browser, whereas giscloud.com maps are all vector, with raster tiles if needed. Also Google doesn't claim to be a GIS at all. These are in fact different tools for different purposes.

Have you heard the expression horses for courses - some race courses are better suited for some race horses - meaning various map types are better for different purposes?

UK electoral boundary changes

aizolnai@gmail.com (Andrew Zolnai) — Sat, 08 May 2010 17:42:00 +0000

New UK Ordnance Survey OpenData Bounday Lines released this month, show changes compared with the original public release in April of October 2009 data. Both Electoral Division and Westminster Consituency boundaries are what you see in the BBC News or many other UK election maps on TV or online. Following yesterday's parliamentary and county council election results, one might ask:

were these simple data updates, or were boundaries changed prior to elections
in the latter, seeing how close election results were, was there gerrymandering?

Media maps very effectively portrayed the vote swings leading to a hung parliament. GIS can further be used to dig a little deeper into these boundary changes. I used GIScloud and OpenData to show these differences in two ways illustrated below:

regional map overlays polygons for 2009 (red outline and pink fill) and 2010 (green outline)
polygon centroids for 2009 (large blue) and 2010 (smaller cyan) highlight the changed areas

Got to the full maps to step through the themes (heat map variations of the centroids are another illustration).

Constituency area map, live quickview (full giscloud map)
- zoomed in to East Anglia to show detail

Consituency heat map, live quickview (full giscloud map)

Historic Fenlands Mashup

aizolnai@gmail.com (Andrew Zolnai) — Sat, 01 May 2010 10:11:00 +0000

Here is a mashup on giscloud.com of the geographic history of land cover and surface geology of East Anglia since Domesday based on:

selected land cover from H.C. Darby's publications on the Middle Ages and later
administrative and geographic boundaries from UK Ordnance Survey OpenData
surficial and offshore geology from British Geological Survey web mapping services
and global coastline for a lightweight frame of reference from US NGDC GSHHS

Below is a dynamic quick view of the 1332 Assessment, counties and geology explained here and here - also posted here with Google and Open Sreet Map underlays. In the public domain or with specific rights, data are posted under Creative Commons Attribution - Non-Commercial - Share Alike 2.0 UK: England & Wales Licence. Full metadata are posted here, showing the detailed attribute data.

I share on giscloud.com the free view map with all layers, legend and attibution. Their distinction is posting vector data with GIS functionality on full view maps: Original shape files are uploaded in PostGIS for full vector GIS function, and WMS rasters post in original British National Grid.

Notice on both maps how zooming in and out preserves the features: are you aware of another vector web GIS today?

[08May2010 additions:

giscloud.com put in a lot of work to allow this mashup, originally in stock WGS84, to be posted in the Ordnance Surgey's British National Grid projection - thanks Dino for this
after visiting Cambridge University's S. Oosthuizen, population density data were provided by J. Bowring and added it to the giscloud map and the metadata file - thanks all for this]

Show me the money

aizolnai@gmail.com (Andrew Zolnai) — Thu, 29 Apr 2010 16:02:00 +0000

That was my response to Peter Batty's call for comment on his GITA panel this week:

"show me the money", what business model do crowd-sourcing or other neo-geo's have that generates sustainable income, if: a) data is free (new data.gov), b) software is free (FOSS) and c) services are TBD [in other words] do going concerns really work on neo-geo, or do they use paleo work to bankroll the neo, and if so what is their go-forward plan across Moore's chasm?

While this slots right in with #directionsapb blogpost and my earlier one, watch below for Pete's new insights from his good questions:

Fantastic upcoming panel at GITA!!
[...]
I’m planning for the discussion to be very flexible and interactive, and I’ll take questions from the audience, but I do have a few topics and questions lined up. The following are some candidates:
Can crowdsourcing give you good enough quality? What are its limitations?
What does crowdsourcing do to the notion of “authoritative” data?
Many OGC standards are based on a technical approach that is 10 years old and predates newer web standards and approaches. Do they have a future or do we need to start again, or significantly rework them?
What are the limitations of the more lightweight data sharing standards like KML and GeoRSS?
What are the factors that determine whether a standard becomes widely adopted or not?
Will Google become the default way we find spatial data (or has it already)? Would this be a good thing or a bad thing?
Do you think that the traditional GIS vendors will still be around in 5-10 years?
[...]

Post-medieval Fenlands GIS

aizolnai@gmail.com (Andrew Zolnai) — Sat, 24 Apr 2010 14:17:00 +0000

Let's look at the geographic history of land cover and surface geology of East Anglia after the Civil War , based on Ordnance Survey OpenData and British Geological Survey web mapping services (WMS). My previous posting discussed H.C. Darby's historic & geographic economics of East Anglia Fenlands between the Domesday census and the Civil War.

The 1794-1813 agricultural land cover was derived by assigning a land cover class per parish from Darby's The Draining of the Fens (detailed here) - the map-figure's odd projection precludes its proper registration to the GIS map - this is a gross approximation, but it matches Darby's classification by parishes (posted here). The ~1800 land cover and the 1641 Subsidy classes correlate quite well: this supports Darby's conclusion that the economic regime didn't vary much after the Midlle Ages.

(click on image to enlarge)

Let's now use surficial geology maps, which can be referenced directly - or mashed up - to the GIS map. The 1877 surficial geology of Cambridgeshire in Darby's book is registred to parishes, and can thus be georeferenced to the GIS map. And the current surficial geology from BGS' WMS (detailed here) is posted directly as a GIS layer.

(click on image to enlarge)

The ~1800 land cover and the 1877 surficial geology correlate well, given the approximations of the former. The 1877 and current surficial geology also correlate well, but subtle differences can be seen: silt and peat have similar extents, as do clay in the north central and southeastern Cambridgshire; however the clay belt running SW-NE at centre appears to have shrunk significantly from the 1877 to the current survey. The map-figures being registered to parishes and the WMS at 1:625,000 scale, this bears further examination at much larger scales and on the ground.

GIS is a great way to highlight similarities and differences over time. And suggest further investigation - are the data reliable? are the differences real? are the processes undertood? etc. The great variety of data coverage across time, moreover, points to further research for each period- will this exposé encourage you to find your own? And to help along, this will be posted on the web as a mash-up here.

PS: BBC TV's Mapping the Medieval Mind makes this very topical: "Maps delight, they unsettle, they reveal deep truths, not only where we are, but also who we are". Also the British Library's blog on same.

Happy Earth Day, in a British roundabout way

aizolnai@gmail.com (Andrew Zolnai) — Thu, 22 Apr 2010 16:14:00 +0000

A recent twittersation with @rollohome had me pull out these screen shots from Google Maps - it doesn't appear to give proper driving directions around complex British roundabouts - I thought this was a curious counterpoint for Earth Day, as we go round&round...

Swindon magical roundabout is the most famous example, but this dragged two examples out of my distant memory, when I lived in Walton-on-Thames, Surrey around the new millenium.

Hatton Cross roundabout on the east side of London Heathrow airport - I used to show visitors the hangars storing half a dozen Concordes* at the time, much to my daughter's delight.

Pregnant roundabout going to Garson garden centre near Esher, Surrey - nearby are the Epsom Downs, where I learned a geo-curiosity of the English language, that downs are in fact up (hills).

*: interestingly enough, Bing Maps 3D still shows a Concorde parked outside Heathrow, even though I don't recall seeing them there recently.

Medieval Fenlands GIS

aizolnai@gmail.com (Andrew Zolnai) — Sat, 17 Apr 2010 21:24:00 +0000

The recent release of UK Ordnance Survey OpenData opened the opportunity to post H.C. Darby's data from The Medieval Fenland and The Drainage of the Fens of East Anglia in the eastern UK. And parishes are the geographic unit that remained constant since the Middle Ages.

Thus historic geographic data from the Domesday census commissioned by William the Conqueror in 1087, can be compared in East Anglia with tax assessment data from Lay Subsidies of 1327 - 1332 and that of 1640 - 1641. Darby's magnum opus at King's College, Cambridge UK, was to compile Domesday data for England in 7 volumes. He also studied the Fenlands north of Cambridge as they evolved from medieval swamplands to drained fenlands before and after the Civil War.

At a relatively modest 200 pages, Darby's first book has an astonishing 25 map figures... and his second book has 310 pages and 35 map figures! OpenData now provides free infrastructure data such as counties and parishes, roads and rivers etc. in shape file format. It is therefore easy to derive a local subset for the counties of Norfolk, Cambridgeshire, Linclonshire, and the south end of Yorkshire. As parishes were the constant geographic unit, attributes were simply added to those shapefiles and filled with Darby's classifications. Data and details are available on request, as are metadata. Aso later data will be combined into a webmap.

Maps below illustrate how the economic wealth of East Anglia switched from the northern & southern uplands to the central lowlands after the Norman Conquest, and then changing little until after the Civil War.

(click on image to enlarge)

How does one compare economic wealth since early medival times in spatial and statistical terms? Darby reported two types of datasets: the Domesday census data for 1087, and the tax assessment data for 1327-1332 and 1640-1641 Lay Subsidies. Out of Domesday census, Darby tallies population, meadow extent and ploughteam numbers as economic indicators in an agrarian regime, before comprehensive monetary regimes were instituted - these were normalised on a per 1000 acres base and classified by parish for comparison across the three townships (no data for Norfolk or Suffolk reported, and Cambridgeshire was smaller then separated from Huntingtonshire). During the Middle ages monetary regimes were instituted as feodalism was gradually organised economically and centralised politically. Tax assessments are therefore good economic indicators for the four townhsips where data were reported in Darby's first book, and Cambridgeshire alone in his second book.

The map above classifies plough teams for 1087 and tax assessments for 1327-1332 all per 1000 acres by parish; as those for 1640-1641 are roughly double, the same classes are posted per 500 acres instead. This gives a geospatial reference at about 1000, 650 and 350 years ago! Darby used many more factors such as population, road and river access and commerce, plus governing and religious institutions to fill out the historical geography of the Fenlands. He concluded they were not a backwater as previously thought. They were in fact relatively prosperous first in the drylands and then in the fenlands, with little subsequent change in wealth distribution.