VGI - Meh

March 13th, 2008 · 29 Comments · GIS

Look I believe in the concept, but I’m so sick and tired of these freaking terms. Personally I think folks use the term VGI to segregate themselves from the community at large. I mean look at this list of participants at a workshop late last year.

Does it bother anyone else that a workshop meeting on social cartography, limited to 40 persons is elitist? This infatuation with VGI is becoming unhealthy to the direction we all want GIS and Geography to move. Why waste efforts down a side street when the main street geospatial crowd is working so hard to get more open. This VGI fad is just more of that “University” mentality that drags down innovation.

Update Fantomplanet says:

You take VGI for what it is. Someone has to sort that crap out half the time……. I asked Goodchild how his concept isn’t any different that the stuff the wiki-blog crowd is trying to sort out. “It isn’t.

Well if Wikipedia is a goal of the concept, then VGI is a non-starter. The world needs less, not more gatekeepers and the whole Wiki concept is based on editors deciding what is right and wrong, not the community. Just because you can edit something doesn’t mean that it is collaborative. Wikipedia is flawed and cult-like and if that is the direction of Geography, count me out.

Tags:

29 responses so far ↓

1 David Davis // Mar 13, 2008 at 9:39 am

Wow James, too much coffee this morning.

I tend to agree with your overall statement. There seems to be much elitistism these days. OSGeo, OpenStreetMap, VGI, etc all come off as “we know better, so just wait until we tell you what you should think and do it”. I’ve tuned all out.
2 Lefty // Mar 13, 2008 at 9:41 am

James’ checklist for the Dev Summit:

1. Piss off ESRI
2. Piss off influentials

Looking great James.
3 Arnold // Mar 13, 2008 at 9:44 am

OK, I’ve been following this VGI thing since the blogs first started talking about it. I like the concept, but thyere seems to be much talking and little action. I do my own part at my city so I guess I’m helping change the world without the help of UCSB.

Horray for me.
4 Anon // Mar 13, 2008 at 9:45 am

If you were invited James, would you still think it is a cult?
5 James Fee // Mar 13, 2008 at 9:55 am

@Anon: I wouldn’t go even if asked. Workshops really don’t interest me and I’d rather spend time with my son than talking about concepts that will never be implemented.
6 Lenny // Mar 13, 2008 at 9:58 am

Why does everything have to be social networks? The majority of the world has no clue what that means. We should be working through existing organizations, not creating new ones.
7 Kirk // Mar 13, 2008 at 10:18 am

I think organizations like FEMA need to be prodded into accomodating VGI, or crowdsourcing or whatever you want to call it. Sometimes it takes an academic imprimatur to do this. I posted on it here:
http://ambergis.wordpress.com/2008/03/13/vgi-for-flood-plains-around-myhouse/
8 Adam // Mar 13, 2008 at 10:25 am

Wait…what is VGI?

Oh ok. Can’t we just call it sharing and move on? No? Alright.

I’ll have to take a deep dive into VGI later this afternoon. Right now I’m trying to upgrade to Web 2.5 (beta)….

-Adam
9 Link // Mar 13, 2008 at 10:38 am

@Kirk: You really want this “organization” working with FEMA? I’m not sure this is the way to go about it.
10 Matt Perry // Mar 13, 2008 at 10:49 am

Personally I like the concept but find the “University mentality” and the term VGI to be ridiculously… academic. Where are the results?

Santa Barbara hosted the VGI conference, UCSB is home to Goodchild and the self proclaimed birthplace of VGI. Yet when I look at open street map data for this area, 99.9 % of the data is either imported from TIGER or edited by a single person (me).

Apparently they’ve been too busy thinking about VGI to actually volunteer any geographic information.
11 Lefty // Mar 13, 2008 at 11:04 am

@Matt Perry: Your last sentence sums up what I’ve been thinking. Well put.
12 Kirk // Mar 13, 2008 at 11:21 am

@Link:
I think academia has a role in researching what motivates people to volunteer information. Is it altruism, enlightened self-interest or what? If we had a better understanding then when FEMA is re-organized, which I think is inevitable, policies can be adopted that encourage voluntary behavior. Look at all the activity surrounding Katrina maybe just a few small things like requiring Rest APIs, or adoption of Common Alerting Protocol would go a long way.
http://www.ucgis.org/katrina/inc_katrina_details.asp?ID=16

Or, I guess we could leave it to the private sector and let Google figure out how to generate ad revenue from disasters (e.g. ads for wind damage repair services).
13 Phil // Mar 13, 2008 at 11:39 am

@Kirk: While I agree that FEMA floodplain maps are full of errors and inconsistencies, they are one of the last things you would want to crowdsource/wikify/massively multieditor online map or whatever.

First of all, everyone would edit their home out of the floodplain to avoid the mandatory purchase requirement. Then we would be back to before 1968, when the National Flood Insurance Protection Act signed, and people would build freely in the floodplain only to encounter massive devastation when a flood would inevitably occur.

Secondly, a Professional Engineer is required to create the hydrologic/hydraulic model that is used to determine base flood elevations (i.e. during a 1%-annual-chance flood, the stream’s water will reach an elevation of X feet).

If you are proposing to alter the floodplain based on the existing base flood elevations and use updated topo, what would the requirement be? Personal GPS? TotalStation? 2′ contours from lidar? FEMA currently has guidelines for topo during the remapping process (http://www.fema.gov/library/viewRecord.do?id=2206)

I am not sure what problems exist in your area, but throughout the U.S. many Flood Insurance Rate Maps were created using USGS topo maps from the 50’s, 60’s, and 70’s. These were very good for their time, but do not have near the accuracy that we now desire (and have available through technology such as lidar).

Also, FEMA currently has procedures if your home has mistakenly been placed in a floodplain. First you have to determine the base flood elevation for your property. Next you have a surveyor determine the elevation of your house at the lowest adjacent grade. If the house is higher than the flood level, apply for a LOMA (letter of map amendment). This is an official letter that states that your structures are out of the floodplain and you are not required to purchase flood insurance.

Finally, I want to say Kirk that I am not trying to attack you or call you out personally — I am fully aware and appreciative of all the help you’ve provided to the GIS community over the years. I just know that flood insurance is a very weird subject, and only after working with it for a brief while did I begin to understand the frustrating magnitude of mapping the U.S. at a scale of 1:6000* with little funding for new H&H studies.

(note: urban areas are mapped at a 1:6000 scale. Rural areas are mapped at 1:12,000 or 1:24,000)
14 Kirk // Mar 13, 2008 at 12:03 pm

@Phil -
I’m not suggesting that crowdsourced data replace surveys. Our house was in a flood plain and we filed for a LOMA. We told FEMA that we saw no reason to hire a surveyor when two nearby neighbors had done so (at $800 each as I recall) and it could clearly be seen (on the map) that we are higher elevation.

FEMA replied saying we must hire a surveyor.

Instead, we wrote our congressman complaining . I’m not sure what tools or maps our congressman used, but somehow he convinced FEMA that we should get a LOMA without a survey.

Now if there were a website that coordinated protests, then homeowners could have collectively bargained for an engineer to come survey the entire area. Instead, individual homeowners paid around $800 each for a survey. If FEMA had a policy saying that surveys must be placed at a RestFul site where I could discover and mash them up, I could have filed for a LOMA without a survey. FEMA could have clearly seen, based on my neighbors survey, that I was not in a flood zone.

I’m not picking on you personally, but I sense that a lot of surveyors see VGI/crowdsourcing as a threat, when it is not.
15 Fantom Planet // Mar 13, 2008 at 12:07 pm

Geez? I just say that a guy visited, we discussed some stuff, and then this happens?

So much for ill-intended consequences.
16 anon // Mar 13, 2008 at 12:44 pm

guess that will teach you to blog, eh?
17 Phil // Mar 13, 2008 at 12:49 pm

Kirk: I wholeheartedly agree that a website to coordinate protests would be of great worth to the public. I misunderstood your initial comment — my bad.

Also, sorry for derailing the conversation to flood maps. Back to our regularly scheduled program!
18 Lefty // Mar 13, 2008 at 12:52 pm

@Kirk Do we really need VGI to accomplish what you are saying there? I’m thinking more middlemen isn’t the problem.
19 SeanG // Mar 13, 2008 at 1:27 pm

I confess I don’t know much about VGI and crowdsourcing, no more than the average blog reader. From what I understand, I’m not sure I buy the concept that the crowd creates the spatial data. I think a better responsibility for the community is to describe the place, to fill in the attributes of the location.
20 Kirk // Mar 13, 2008 at 2:11 pm

@SeanG, Lefty: If I can beat the FEMA dead horse one more time, I do think it offers a use case illustrating the value of VGI …

My neighbors both had volunteered to give me their surveys, but could not locate the papers. In this case volunteering data does not involve data creation, but rather offering private data for public use. There needs to be an easy way for homeowners to manage the spatial data that they own, maybe using the MyHouse concept described on my blog.

If FEMA held on to surveys submitted with all LOMA applications, they could present a website where applicants could later opt to volunteer those surveys to the public. Other applicants could then browse those surveys and use them to justify LOMA approval.

I don’t think the mappers updating the flood maps are using surveys from LOMA applicants, but seems like they should.
21 SteveC // Mar 13, 2008 at 3:20 pm

geowankers layed in to the VGI meeting months ago, see the thread starting here:

http://lists.burri.to/pipermail/geowanking/2008-January/004722.html

@ Matt Perry ‘Apparently they’ve been too busy thinking about VGI to actually volunteer any geographic information.’ - I couldn’t agree more. That’s why we just concentrate on the maps at OSM.
22 Brendan Hemens // Mar 13, 2008 at 7:00 pm

James’ comment on Wikipedia, less gatekeepers not more, really got me thinking. I really enjoyed Wikipedia at first as a resource, and especially the fact that mistakes, bones of contention, etc., in many ways exposed reality as I understand it - perfection in ideas doesn’t exist. Wikipedia revealed debate and disagreement as well as consensus. More and more frequently, the articles I read are marked up with the criticisms of would-be censors, e.g., “This section does not meet content standards for…This section contains instructional material, Wikipedia is not a how-to…”, etc., etc. It’s a short hop to outright restriction on content.
This conference hints at the first attempts to critique (”What methods can be used to validate…”) The question makes sense, but the destination is all too clear.
Wikipedia articles and ‘volunteered’ spatial data are models. Error is simply inescapable, and for those who really seek professionalism, it is actually a valuable addition to the data itself. Openly contributed error isn’t in a blackbox - it’s where everyone can see it.
It’s like the English language - once it was (and thankfully, for many remains) a vibrant and dynamic playground for creative minds; sadly, it also engendered the grammarian.
23 Waddy // Mar 13, 2008 at 7:53 pm

We tend to get so tied up in trying to make these ideas perfect that we never achieve anything. By the time they get around to accomlishing something, there will be something bigger and better around.

James, Let them waste their days while we get on with with the job of being innovative.
24 Paul // Mar 13, 2008 at 10:52 pm

James … I’m totally with you with on the spirit and pragmatism of this post; however I have to take issue with your last statement on “university” mentality that drags down innovation.

Jackie D came out of the Harvard Map Lab (Nick Christman wrote a treasure of a book on LCGS ) — it was the entire “university” thing that set Jack off on his ESRI quest….think Tufte and data visualization - it’s just that it’s the old guard. They deserve - or have earned the right - to relish in their accomplishments. But they are not part of the future (which, I suspect, is the undercurrent of your post). They are trying to “find their place” in history - leave a legacy that ties to the new functional ‘open-geography’

I have no problem if Jack, Goodchild, et al want to lend their name / sponsor public initiatives; that is what brings awareness and funding. Unfortunately - too often either the wrong thing gets funded or a good project gets too many “generals” and not enough soliders (hello Census and FEMA).

The old guard isn’t vested in ‘open technologies’ - they didn’t invent it.

The future is all about LIDAR, that kid who started Facebook, geo-rss, and mobile-voice activated location-aware content (?)— which, in some respects, may rely on VGI (…that’s so NOT where I thought I was going with this).
25 Waddy // Mar 14, 2008 at 2:45 am

Point taken Paul. But there is no undercurrent with my statement. My intention is to highlight that too often academia slows down progression with analysis. The world is full of very intelligent people who have thought through many of the issues we have considered. I agree with James on the fact that if this VGI team wanted to go down the open source route they would have opened up this issue as many have before them to the whole Geospatial communitly. There is technology out to do this. It appears to me that the VGI forum is designed to caress a few egos.
26 SteveC // Mar 14, 2008 at 2:49 am

‘It appears to me that the VGI forum is designed to caress a few egos.’

I don’t know about the academic sphere that may be continuing to propagate the VGI meme, but you have a very strange idea of what actually happened at the event.
27 Dave Smith // Mar 14, 2008 at 8:35 am

I would wholeheartedly agree that the FEMA mapping needs to be overhauled. It’s my understanding that it was the intent that the last big bundled contract was to begin this process, but instead the primary focus has just been digitizing paper maps. Garbage in = garbage out.

As far as VGI and crowdsourcing, it doesn’t necessarily need gatekeepers - gatekeepers stifle production. But it does need some ground rules, because again, it’s very easy to contaminate a dataset with bad inputs and/or continually keep propagating bad inputs. Data needs to be vetted and assessed periodically, needs to have a lineage, source, metadata - if someone intentionally or otherwise unintentionally submits a whole heap of bad data, it needs to be able to be isolated and corrected, lest one end up chasing around error.

But again, that vetting process can also be community driven. No need for any elite gatekeepers, but there is need for some domain expertise to make sure the effort doesn’t miss the mark.
28 Pmarc // Mar 14, 2008 at 11:45 am

@ David Davis:

What’s the issue with OSGeo? I felt it was a good initiative. PVT if you like.

@Dave Smith:

I guess that if anything, Wikipedia has shown us that no community vetting process will occur. Most likely, some truth commissioner will be appointed and will act on absolutely arbitrary basis on his appointed realm.
29 Ninja // Mar 31, 2008 at 3:51 am

The question makes sense, but the destination is all too clear.