Geolode: the motherlode of geospatial data sources
This is a modal window.
The media could not be loaded, either because the server or network failed or because the format is not supported.
Formal Metadata
| Title |
| |
| Title of Series | ||
| Number of Parts | 188 | |
| Author | ||
| License | CC Attribution 3.0 Germany: You are free to use, adapt and copy, distribute and transmit the work or content in adapted or unchanged form for any legal purpose as long as the work is attributed to the author in the manner specified by the author or licensor. | |
| Identifiers | 10.5446/31706 (DOI) | |
| Publisher | ||
| Release Date | ||
| Language | ||
| Producer | ||
| Production Year | 2014 | |
| Production Place | Portland, Oregon, United States of America |
Content Metadata
| Subject Area | ||
| Genre | ||
| Abstract |
| |
| Keywords |
FOSS4G 2014 Portland63 / 188
7
10
14
15
16
23
24
25
28
29
33
37
39
40
43
45
46
48
50
56
64
65
69
72
74
82
89
91
98
102
107
111
114
118
128
131
132
135
138
141
143
147
149
150
157
158
161
164
165
166
173
174
175
179
185
00:00
WebsiteProduct (business)WordOpen sourceWage labourSoftware developerSoftwareCore dumpBitPoint (geometry)Online helpProjective planePresentation of a groupComputer animation
00:50
Level (video gaming)Open sourceSlide ruleSoftwareSource codeVariety (linguistics)DialectDifferent (Kate Ryan album)Latent heatProjective planePlanningHierarchyComputer animation
01:29
Latent heatResultantLevel (video gaming)Physical systemInternetworkingLocal ringMultiplication signStudent's t-testReal numberWordQuicksortTrailPanel paintingComputer animation
03:13
Point (geometry)Computer-assisted translationPresentation of a groupLevel (video gaming)Term (mathematics)ResultantRankingHecke operatorQuicksortCovering spaceMultiplication signSelf-organizationMetropolitan area networkMassView (database)Computer animation
04:30
Group actionFigurate numberShooting methodProduct (business)Library (computing)Different (Kate Ryan album)Musical ensembleBranch (computer science)Meeting/InterviewComputer animation
05:00
Video gameLevel (video gaming)Projective planePlotterBookmark (World Wide Web)WebsiteInterface (computing)QuicksortTrailMetadataChaos (cosmogony)Multiplication signStructural loadProduct (business)Cycle (graph theory)Library (computing)WordWeightModule (mathematics)Computer animationXML
07:21
Covering spaceMaxima and minimaWebsiteHeegaard splittingStructural loadInterface (computing)Different (Kate Ryan album)Uniform resource locatorDemo (music)Source codeDuality (mathematics)Rule of inferenceEnthalpyComputer animation
07:57
WebsiteQuicksortLatent heatDifferent (Kate Ryan album)Descriptive statisticsResultantBookmark (World Wide Web)
08:51
Descriptive statisticsDemo (music)Moment (mathematics)Revision controlElectronic mailing listInterface (computing)InformationWeb browserText editorWebsiteService (economics)AdditionBookmark (World Wide Web)WordUniform resource locatorSource codeXML
10:05
WebsiteBookmark (World Wide Web)Universe (mathematics)Duality (mathematics)Software testingQuicksortDemo (music)Key (cryptography)Computer animation
11:40
Multiplication signDescriptive statisticsWebsiteMessage passingCASE <Informatik>Uniform resource locatorComputer animationXML
12:43
Combinational logicOpen setElement (mathematics)StapeldateiWebsiteDemo (music)Pay televisionLibrary catalogCuboidElectronic mailing listQuery languageSource codeResultantReading (process)Beta function
13:26
Level (video gaming)ResultantInterface (computing)CuboidQuery languageGeometryComputer animationXML
13:52
WebsiteRow (database)Commitment schemeDirection (geometry)Musical ensembleText editorCollaborationismFree groupLink (knot theory)Image resolutionLibrary catalogSource codeStudent's t-testQuicksortTwitterScaling (geometry)Ocean currentProjective planeInformationBitSampling (statistics)Directory serviceTrailAddress spaceWordNeighbourhood (graph theory)Order (biology)Revision controlNumbering schemeDifferent (Kate Ryan album)FeedbackStructural loadState of matterSoftwareElectronic mailing listXMLUML
16:41
Library (computing)WebsiteSource codeWeb pageLink (knot theory)State of matterQuicksortWeb 2.0Standard deviationExistential quantificationElectronic mailing listNumberProjective planeMaxima and minimaComputer configurationObject (grammar)Different (Kate Ryan album)Library catalogResultantLevel (video gaming)Right angleConnected spaceLoginRankingPhysical systemInformationUniverse (mathematics)Error messageGoodness of fitOrientation (vector space)Covering spaceGodResource allocationUniform resource locatorWave packet
20:02
WebsiteDomain nameMultiplicationElectronic mailing listRight angleLevel (video gaming)DialectPoint (geometry)Web pageLink (knot theory)Attribute grammarTouch typingState of matterMatching (graph theory)Boundary value problemCartesian coordinate systemProbability density functionUniverse (mathematics)Self-organizationSpreadsheetProcess (computing)GeometryDomain nameStructural loadXMLUML
22:15
Library catalogDirection (geometry)ImplementationPlastikkarteInternet forumDifferent (Kate Ryan album)Perspective (visual)Chemical equationMultiplication signWebsiteProcess (computing)Uniform resource locatorCategory of beingSelf-organizationPhysical systemCentralizer and normalizerNichtlineares GleichungssystemPay televisionMetadataService (economics)Structural loadComplex (psychology)Variety (linguistics)Musical ensembleSimilarity (geometry)Directed graphRow (database)Computing platformLevel (video gaming)QuicksortVapor barrierLink (knot theory)ACIDOpen setProduct (business)Focus (optics)Point (geometry)Instance (computer science)OpenURLDatabaseMereologyResultantComputer configurationLibrary (computing)Greedy algorithmLimit (category theory)Latent heatLine (geometry)Query languageContent (media)MathematicsAreaChannel capacitySingle-precision floating-point formatSubject indexingWeb portalMobile appEmailAutomatic differentiationServer (computing)Lecture/Conference
Transcript: English(auto-generated)
00:00
All right. Well, I guess I'll start. So if you're in this room, you're here looking for data. And I'm here to introduce a site called geoload.org, which should help you find that data. A little bit of an introduction. My name's Keith. I'm at Cornell University in Ithaca, New York. Anyone else from New York? Sure.
00:21
She's pretending. And I'm also a librarian. So I'm a librarian at Cornell. I'm not a IT developer, although I sort of was pseudo part-time IT, theoretically, at one point. So I do little tech projects on the side. But my main core of my work is as a librarian, so looking for data, helping people use software,
00:41
helping people troubleshoot software, all geospatial related. So here we are at Phosphor-G, which is open source software. But I'm here to talk about data. There was another slide. It disappeared. Because with all this great open source software, you can't do a whole lot until you have data.
01:01
Otherwise, your map looks pretty much like this without the data. So obviously, data. So let's talk about data. There's a whole variety of data out there. It comes from different sources at different levels of governmental hierarchies, regional projects, international NGOs. It comes from kite cameras, quadcopters, paper airplane
01:23
cameras, whatever you have. So there's a lot of data out there, which makes it really hard to find the specific type of data that you're looking for. So if you do a Google search for just plain GIS data,
01:40
no other qualifiers. I just updated this this morning. As of this morning, 63 million. For some reason, last week, it was like 118 million. So I don't know what happened to all that data. It just disappeared off the internet. They did a recount. I don't know what happened. So that's a lot of results. And it's not something you want to browse through. I don't think you can actually get to the end of that list.
02:03
So is this a problem? For some people, maybe. But not necessarily for everyone. Local governments typically have all their own data in their own systems. They don't need to go searching for it. They know where it is. They have it. So that's not really an issue. Experienced researchers already know
02:20
where they get the data they need to process and analyze. So for those folks, it's not so much an issue. But what about the rest of us? And who is the rest of us? Everyone in this room? It's the students who are just learning where to find data. You give them an introductory tutorial. And you provide the data for them. And then all of a sudden, in the real world,
02:40
they're doing their assignments. They have to go find data. And that is very hard the first time. There's also researchers who might be exploring a new topic, who might be new to a particular field and don't know where yet to find that sort of data. There's journalists who might be reporting on a new story. They need to track data down. Map makers, of course.
03:02
Demographers, gerrymanderers. They need that demographic data. Civic activists, whale watchers, birders, treasure hunters, deep sea divers, wilderness explorers, napping cat mappers. I heard you get extra points if you show a cat in your presentation.
03:21
I also heard you get extra points if you use atomic rights big for your presentation. I get two points, I think. Kite Mack Rectification, I can say it. Specialists, farmer's market bikeability analysts. They need the data for where those roads are, the bike routes, the farmer's markets. And they might not know where to find that data. So we want to help them.
03:41
So there's a lot of people looking for data out there. And basically, anyone who's trying to learn more about their world and map it, analyze it spatially, they need geospatial data. So about 63 million results. You do a Google search, it does a relevance ranking. If you add in a term like Portland,
04:00
maybe throw in a term like land cover, you'll get less than 63 million, but still a heck of a lot of results. And it's relevance ranked, whatever that means. But you really expect to find what you want at the very top every time, probably not. So it's sort of a mess. So my question is, who is going to organize this mess? And what I would propose to you, being a librarian,
04:25
the librarian should organize this. Or they can organize it, other people can too. You might be thinking to yourself, librarians? Aren't those the people who work with books and they shush people? This is actually the, I forget the librarian's name, this was modeled after, but the shushing librarian, what's her name?
04:40
Nancy Pearl, action figure shushing librarian. You can buy these on Amazon. So there's this myth that librarians shush people. And maybe we do, we're trying to concentrate, we're trying to find the data. But in reality, librarians, at least in my experience, are a little bit different. And they look a little more like this.
05:02
So it's not just about books anymore. What is the best in life? To shush the noisy, to see them driving home. Mm-hmm. So in reality, librarians strive to conquer the chaos of the world by collecting it and maybe eventually organizing it.
05:21
And I say eventually because this has sort of been a long project. So this is the secret librarian plot revealed here. There are other librarians in the room, so be careful. They will defend me. They have sworn a sacred oath here. So I say eventually because I've sort of been working on this geoloid project for seven years, which is a pretty long time. It's on my back burner.
05:41
I work on it when I have a little bit of time. So seven years ago, 2007, I was using a site called Delicious, and they spelled it like that, punctuated it like that, for better or worse. And I was bookmarking things. And there are a lot of other people bookmarking things. This is great. I can keep track of certain people who are also bookmarking geospatial data, kind of aggregate all that,
06:00
maybe use the API to develop a site where you can filter and browse and search across all these bookmarks. So that sounded great. But Delicious has changed over the years. I'm no longer using it. A few years ago, I had had enough, and I decided to roll my own sort of thing. And eventually now here in 2014 is geoload.org.
06:22
And so what is geoload? So first, I guess I should explain the name. First of all, it's really hard to come up with a good name. Vlad had mentioned leaflet. And it was such a brilliant name, and it hadn't been used. And it was only seven letters. So this is only seven letters. But people are going to misspell it. L-O-A-D. They might not know exactly what it's referring to.
06:42
So the name comes from people are looking for the mother load of geospatial data. That was the idea. If you have a better name, please feel free to suggest something. And it's a minimum viable product. So it's not trying to do everything in the world. It has one goal, and it's going to try to do its best to do that.
07:00
It uses extremely lightweight metadata to categorize and help you find geospatial websites. So I think glorified bookmarks, not ISO XML metadata. So very lightweight. The interface is actually mappless. There's no map.
07:21
It's not like this kind of mappless, but more of a text-only interface, because that's the minimum viable thing. So the idea is that you can search. You can browse. You can filter all these different data sources, primarily by these three things. So the place is the location. The topic, those are roughly ISO topics,
07:41
although I think I split land cover out from whatever else it was bunched together with. And by more general freeform tags, they give more specificity than the topic. So now a little demo here. Of course, I had the website up earlier, so you may have already looked at it. So here's GeoLoad.
08:00
And it's general incarnation. On the left, you can see that there are the different facets of the place, the topic, and tag. I can click on any of these. So if I'm interested in a particular place, so Pakistan, for example, right now there's only six websites in here where you can get data for Pakistan. There's probably others, and I'll be talking about adding things later.
08:23
The other thing you can do is I can say if I'm interested in elevation data, you can combine these sorts of things. So elevation, and then I'm actually interested in elevation data here in the United States. So here I get all these results, and you can filter things down further and further. You can also do freeform searches, which will search those descriptions, as well as the titles
08:41
and everything else. So it's just really a lightweight bookmark, specific for geospatial, though. Oops, that's not it. All right, so that was a demo. You can also edit. So there's an editing interface that up till now it's
09:01
just been me, but we'll talk about that in just a moment. So with editing, obviously you need an editor to go in and add new things. There's also a bookmarklet that if you just drag that up to your browser toolbar, and if you have editing privileges, when you're on a website of geospatial data, you click it. It opens up the editing form, and you can provide a little information about that.
09:21
The bookmarklet will automatically grab the title and the URL. And if you selected text, it will put that in as a description. You can use markdown in the description. So if you wanted bulleted lists or a little more formatting, you can do that as well. There's an autocomplete for the place and the topic and the tags that way. You can select the correct spelling of the country, for example, or use tags that have already been entered.
09:44
There's also, this is a more recent edition of, as I said, it was just me booking marking things for seven years. But to open this up to a wider audience, I thought it'd be important to have history and versioning. So you can see who made the edits, who's mocking things up, who's the person who gets the award for adding
10:01
the most stuff, whatever it might be. So just a quick demo of that. Oh, and I did not log in yet. Oh, this is what's going to kill me trying to type.
10:20
No, this is Dvorak, yes. I was not looking at the keys. So I can just say, I want to add a new thing here, and I could do it that way. Or if I have a new, oh, who has some data, anybody? So let's just say I've found a great site about test GIS data.
10:59
I always type test.
11:01
Let's just pretend this is a website. Oh, look, there's Stanford. Let's do that. Oh, it's a test panel. Oh, OK. Let's just pretend this is a website that has data. Sorry, I apologize for not having a great example here. All I have to do is click on the geologic bookmark, and something happened here.
11:21
That's the fun of the live demo. Let me add something in here then. Let's just take it for granted that it worked. I'm having some sort of a weird issue here. Didn't see me as logged in. Oh, that'll go down.
11:53
I'm just going to do it to my local computer here. So here's a great site, and I wanted to bookmark that. So I go to the bookmarklet.
12:02
It automatically fills in the title and the URL with the description I can put there, whatever I want. The idea with this is that on the first pass, you might just bookmark that. You don't have time to describe it. The idea is that we can review these and come back later to add more details. But in this case, this was Romania. That's not the topic. The place is Romania.
12:22
So I can just select that. And maybe the topic was focusing on, it wasn't, but if it was focusing on climate, I could do that. I could also add editing notes to say that I might need to double check this site about something or other. So once I save that, it's now in there. It's searchable. And I have a duplicate because I just copied that there.
12:41
But let's just ignore that. So you can edit. Now, the data that's in here, it's open data in the sense that it focuses on open data sources. So I'm not looking to catalog things that you have to buy a subscription to or have a membership or anything like that. There might be some sites that have a combination of open data and closed data.
13:01
But if they do have open data, I'm interested in getting it in here. It's also open data in the sense that everything that's in the catalog, you can just download that in batch. There's an API where you can query basically anything that you type into the search box or whenever you end up with a list of results, you can click on the JSON and get a JSON feed of that.
13:24
It's a really simple schema. So maybe just a quick demo of that. So here I've done a search for Africa and have these results. I can look at that in JSON format. It's just really straightforward and simple. So if you want to come up with a map interface or something like that, you can just pull data directly from here, run queries, bring that in, and make nice bounding
13:42
boxes for all the countries or whatever you'd like to do. All right, so back to GEO. Just to clarify a couple things, it's refining geospatial data websites as a whole. So one website has a record. It's not individual data layers.
14:01
There's lots of other projects that are working on deciding how to describe data layers in great detail and then provide that information to harvesters and CSW and all that sort of thing. So this is something completely different than that, working at a different scale. It's lower resolution data, if you will. And as I said, I'm moving from a solo effort
14:21
to a collaborative effort. And I'm kind of new to this, so I'm trying to figure all this out. So I'm interested in finding collaborators. If you have feedback, I'm greatly interested in that. So one question I have for you is, if you'd like to be an editor, if you think it would be fun to add more resources to this
14:40
or really build out the catalog, let me know. There's a Twitter account, GEOload, that you can just tweet to that. If you're not on Twitter, you could also email me at this address. I'll have this at the end as well. So if you don't write it down yet. If you just are interested in suggesting websites, you don't want to be an official editor.
15:00
You don't want any kind of commitment. You don't need a login. You just want to say, look, here's a site. It's not in your thing. Why don't you just put it there? So again, you could just tweet to GEOload, and I'll be monitoring that. And as new things come in, I'll make sure they get added or you could email directly. Questions?
15:21
I've only gone 15 minutes, so let me ask the first question, if I may. It'll be a couple minutes. But the first question, of course, is what's next? Where is this going? And first of all, it needs more documentation. Just this morning, I actually officially put the site not in a preview directory. And I actually thought, oh, maybe I
15:41
should even say, what is this site? Who's behind this? So I just added a little text there. It's sort of like an unmarked car until this morning. It was there. It was driving around. It was more like an ice cream truck, an unmarked ice cream truck going around your neighborhood, giving out free samples of data or something. So obviously, I want to catalog more websites.
16:02
There's just under 700 websites in there currently. But I figure there must be at least 1,500, 2,000 websites that should be in here. So I want to track those down and get things in there. Right now, it's a bit US-centric. I'm from New York state, so there may be a lot of New York state-related data, US, obviously,
16:23
some international data. A lot of it just happens to be things I've searched for in the past for students who are looking for data. And so it's somewhat arbitrary. So I'd like to have a little more of an international scope for this. One idea is to also harvest sources from sprawling lists of links.
16:41
You've seen these before. There's a really great set of links in the University of Oregon library website. Here's a page of state data, and it just goes on and on. For every state, there's a handful of sources. So to basically grab all these URLs, add those in as well would be a quick way to get a lot of things in.
17:03
Also reviewing and annotating the data that's currently in there. Some of it, if it was just pulled from a list of links like that, occasionally those links are out of date. The site might be gone. It might have moved. So correcting those sorts of errors, also adding more annotations provide more detail about what types of data are there.
17:22
But again, not going into too much detail. Those isotopics are great. The tags are things like, if you're looking for snow data, the isotopic might be climate. But if you're looking for snow, that would be a useful tag. So trying to add information that will make it easier for people to find that data if they're searching that site.
17:42
Automated link checking is another thing that should be done. Because as I said, sometimes these things go offline. Sometimes they temporarily go offline. So you don't want to just completely delete it from the catalog. But set up some sort of system in place to monitor that. And if something's gone for over a year, you can assume it's gone forever, probably.
18:01
Also refining the topics and tags. I've been using the isotopics, which is from the ISO. Kim can probably tell me the exact number of the standard. What is it? 1-9-1-1-5? OK. I always get them mixed up. Thank you. So the isotopics are what I'm actually using are sort of more sane names. So it's not like imagery like everbased maps,
18:23
whatever that long term is. So I've simplified those to make them a little more human readable. There might be some weird one-off tags that probably need to get cleaned up. So kind of refining all of that. Right now, the places, and this is where I show my locational bias of being in upstate New York.
18:41
So I might have Tompkins County, which is in New York, which is in the United States. But there's probably other Tompkins counties elsewhere. I think typically I've been adding things in as a place name as Tompkins County, but also New York. I don't think I've been doing United States. So to have a more international scope, it might be worthwhile exploring some linked data that
19:00
makes these connections between the place names. So that when you do a search for New York, at least if you don't include all the results for United States as well, at least making it clear to the user you might want to search at this higher geographic level. Without that becoming too complicated, I still want it to be a minimum type of object project.
19:20
Another thing that might help is website ranking. So some of these sources are pretty obscure and for very specialized needs. Others, like standard elevation data for the world, there's a couple sources that are just places you should go straight to. So to rank those higher so they'll show up first in those lists might be a useful thing.
19:40
How that actually might happen is a good question. And it could be from depending on how much traffic this gets, you could look at the logs of what people have clicked on and try to rank things that way, or just seeing how many different sprawling lists of links on the web is this website included on. That might be a good way. Lots of options. So as I said, I'm also interested in finding some volunteer co-editors.
20:03
So if you are interested in helping with this, please feel free to get in touch with me at Geoload, KGJ2 at Cornell.edu. And at this point, I'll take your questions.
20:29
As you go to the editing process, do you have anything built in to handle duplication? Right now, there's just a really quick check to see if anything at that domain name exists. Because one thing you run into is, for example,
20:41
the University of Oregon website that have lots of lists. It's the same domain, but then they have a page of international data, a page of state data. And sometimes you might not necessarily want to point to the top level of a website, but point to the page that actually has the data. Because I think that would make it easier in the long run rather than sending someone off to a link that takes them to the top level of a site, but then they have to figure
21:01
out where is the data hidden. So that's why I'm not doing exact matching. But that could be made more robust, certainly, as well. How are you defining geodata? Are we talking about stuff that's GIS ready, like it can just automatically go into a mapping application
21:21
and be read, or PDF maps, spreadsheets, that kind of stuff? Yeah, as you get to spreadsheets, and that's where it gets a little question. Obviously, anything like a shapefile, a geotiff, anything like that is clearly geospatial data. But attribute data, tabular data, is a whole other question. I've typically not included those.
21:41
But a lot of that data is almost ready. All it has to do is be joined with the spatial boundaries for it to be spatially useful. To this point, I've not included those. I think if they were included, it would be useful to have another facet that would indicate is it pure spatial data, or is it tabular data, like a lot of data.gov websites have.
22:09
If it's data that's regional data, it's multiple countries, multiple places, how would you have it tagged that way? Would you just say world, or what? Yeah, so I've tried to keep it simple. For example, there is a set of data that covers some of mainly East and maybe Central Africa.
22:24
I just called it Africa to be simple, rather than tagging with every single dozen different countries. And that's another thing that you need to find some balance. Catalogers and libraries are known for attention to detail. And you can really go overboard if you start going down that hole.
22:41
So I'm trying to always, in my mind, think, let's just keep it simple. Because in the end, if you do search for, I didn't have, there weren't that many results for all of Africa. And for individual countries, there might be more websites. But the idea is that you shouldn't be able to necessarily, my goal is not to pinpoint the specific data that you want.
23:03
My goal is to give you a handful that you can then explore. So I think absolute precision is just going to create a level of complexity that is going to slow things down. Did you build any kind of like subscription service,
23:23
like RSS, into the geoload so you could subscribe to a topic, or a tag, or a location, and get updates on that? There's nothing currently, but it seems like that should be pretty easy to do. Because it does record the time, the date modified, and the date added. So if there are new things, it would be fairly simple,
23:42
I think, to set up an RSS feed for a specific query even. How are you addressing sites that are aggregators in and of themselves? For example, a lot of the environmental mapping I do, I get gray data from data basin,
24:02
which is uploaded by researchers who haven't necessarily exposed that data anywhere else. But within data basin, you can find data for literally the entire world. So absolutely every tag would be applicable. It almost feels like you could set up a separate category for that. Or how have you addressed that? Yeah, that's something I've thought about.
24:20
There are a couple entries in here where I started listing all the topics that apply, and it's almost all of them. So at that point, I think it would make sense to have a sort of a kitchen sink category. Maybe not that exact term, but something that basically says, this is a huge variety of data. Because I think it's kind of ridiculous, of the 19 ISO topics, to try to pick out the 17 that apply.
24:42
I mean, you really have to scour the site and check everything. This should not be a time-consuming process. But I think an all-in-one category would possibly make sense. It could also be addressed at the user side. Now, every user may not intuitively think about it this way. But if you don't limit to a topic, you're going to be finding anything,
25:01
whether it has a topic or not, or what the topic is. But that involves user education and trying to give them guidance on how to use this to find things as well. So you could kind of go a couple options there. How do you handle cities open, or just different governmental organizations' open data
25:20
portals that require logins, like civic apps maybe? It's kind of another barrier to actually accessing the data. So do you just index those and say, oh, you're on your own, make a login? Or do you have any other way of handling that? And are you talking about sites that are open data, but you still just have to register? Yeah, that kind of thing. Yeah, I think I would definitely include those. How the individual sites are organized themselves
25:42
and what you have to do there, like even some of the imagery sites, when you download, you have to have an account to actually download your card or whatever it is, or to get the email saying that the thing from the USGS server is ready for you. So I'm not really worried about those concerns in Geoload,
26:03
because just to track that, if they change your policy, you don't have to try to keep up with all that. So here's the site. Go look. Go find. And don't worry about those specific deals. I wonder whether you have compared Geoload to CCAN, the product that comes from the Open Knowledge
26:23
Foundation in the UK. I have not directly compared, but I have thought about harvesting individual links from some of those sites. The reason I ask is it seems to me
26:41
it's becoming a very popular platform and has a lot of the similarity to Geoload. And I guess from my perspective, I don't know where you decide whether the direction that you want to take Geoload in is perhaps quite parallel to what CCAN is doing, and then maybe focus more attention on the curation of the content, which
27:03
are the metadata records, rather than on the tool. Because certainly a lot of the things that I'm hearing as suggestions to Geoload, I know, are already a part of CCAN. Yeah, I think most of the CCAN instances that I'm aware of and have looked at, they're focused on the data layer
27:20
level. But are you suggesting that use the CCAN infrastructure for maintaining this? So exactly that. There are some people who are using it as the metadata system, and some people are bringing the actual data with it. But it can be used in either way. I guess one thing, and I'd certainly
27:42
be open to looking at that. One question I would have is how easy is it to add things to that if you have a loosely affiliated group of people? That'd be a question I would ask, but I don't know the answer. It is possible to use a hierarchical organization
28:01
of allowing people to use it for the US government's implementation. So you can do this kind of moderation thing with having people control their area of work, if that's what you mean. OK, I'll check that out.
28:21
Any other questions? Or if people want to go to lunch? I don't want you to be at the end of the line. Well, thanks a lot for coming. It's been an honor.