FortiusOne GeoCommons Finder! Is Now Public Beta
I’ve recieved an email saying that the great GeoCommons Finder! is now public beta. Many were very interested when it was announced so if you’ve been waiting for an invite to the private beta, you no longer have to wait. Sign up, check out the user manual, and give feedback.
I love Finder! and how it makes searching for data and learning more about the data easy. You got to check it out. This stuff is easy to use if you know nothing about GIS and it provides enough value to GIS professionals who want to get more data and have it documented for their use.
Brilliant!
hrm….the ability to upload tons of data without metadata.
scary thoughts indeed!
I like the format and easy to use look up and export schemes for shapefiles and kml’s. I think it would be pretty cool if they had scale constraints attached to the search and maybe show where the data is using an mashup of some sort.
Guiness…brilliant.
Thanks for positive feedback James -
@Joseph – there are several options for providing metadata with Finder! 1) It takes the elements supplied by the user (or automatically trapped) in the upload process and maps them to 12 core FGDC and ISO 19115 metadata elements. 2) If you have a full metadata specification you can provide a URL link to the specification 3) We are in process for adding the ability to upload a shp.xml with a full metadata specification that gets auto populated to Finder!.
We were not able to get everything in the first release but providing good metadata support is something we’ve worked quite a bit on. It is difficult to strike a balance between most GeoWeb implementations that have no metadata and a full FGDC specification, but we are trying to solicit feedback from both communities to find a workable compromise. There has also been some good work done on this topic by Dr. Goodchild under metadata 2.0 and VGI (volunteered geographic information).
If folks have suggestions on how to better provide flexible metadata support we’d love to hear them.
best,
sean
@Joseph Wallis: My feelings on that subject are clear.
That said, those who take the time to upload data with clear and concise metadata should be encouraged and rewarded by the use of their data and feedback towards its improvement. Given how much data is being produced out there, we need to get tools to those producers to provide things such as metadata.
1st Ive seen of this site – The simplicity of it I like, but as mentioned already – would be better if there was more advanced search capabilities.
For instance filter on country, perhaps add in some user ratings on a few headings such as confidence of datasets, etc.
Good work though.
Sean, I don’t know what you are talking about. I did not see any option to upload metadata with the files on the uploader tool…but you asked for suggestions…
well then here is my suggestion. provide a spot in the shapefile upload the XML file ESRI creates when you create metadata on the shapefile.
ANd if you want to be super cool, provide a way to upload and parse the XML workspace documents you can export with ArcCatalog. That way you won’t force Geodatabase users to spit everything out into a crappy shapefile and their schemas will be preserved.
I tell you, it confounds me that the general consensus out there is that the FGDC standard is something to strive for. How far are people willing to lower the bar? FGDC should be the minimum.
Thanks for the suggestion Joseph. We are working on the XML integration, so we’ll take a look shortly.
While I am fully in support of providing mechanism for supporting existing metadata, I think that setting a bar for contributing geospatial data that requires hundreds of metadata elements will keep GIS out of main stream adoption.
Some may argue that is a good thing, but the alternative is a divided camp of no metadata at all and very complex metadata. A standard is only as good as its adoption rate. The democratization of geospatial technologies has created a new world where the majority of users are not going to adopt a standard that requires hiring a full time person to keep it fufilled (that is what USGS suggests).
I’d argue that standards need to evolve to match technologies, users and adoption – not the other way around.
best,
sean
Personally I think metadata should be no more complex than a soup can label. If that’s all the information people expect to make a decision on what they are willing to put in their mouth then I wouldn’t count on them digging any deeper when it comes to metadata.
Frankly, it gets old hearing the answer always be…lets lower the bar/standards. It’s too hard on or too much time for people.
Then again, since we have been lowering the bar in school. Too many idiots in the work force, means we have to lower standards to match the idiots level of competency.
People need to rise up to the standards, not standards lowered to their level.
KoS
KoS: What Metadata standard do you think should be “standard”? We can’t even get GIS professionals to use them, how can we expect the “public” to do so?
The standards need to evolve to the point where they are useful to everyone including “us”. I’m not sure we are lowering the bar at all, the bar needs to be realistic to what the expectations are of the data. If we have crowdsourcing of hiking trails in a national forrest and I would like to use that with an EIS I’m preparing, I’d like to know basic information about the data (who collected it, when, how, accuracy). I don’t need all that wacky FGDC fields that we can’t even seem to fill out.
KOS: I think you are confusing achievement standards with technology standards. Technology standards are not about setting a bar they are about getting a large group of people do things in a self similar way so we have a standardization in approach and ease of reuse.
This may be opening a can of worms, but the from my perspective the major reason anyone used FGDC in the first place was because it was mandated by Federal law. Not because it was an approach the market embraced and solved the standardization problem.
I’m not saying metadata is not important – in fact the opposite. The point is a standard that requires * hundred of elements to be filled out by hand is archaic in 2008. We should be working on automatically trapping metadata elements and using semantic linking to derive rich context for the data.
Also I believe having a flexible standard that fits the user is a more clever approach than a one size fits all method. This has been used very effectively in library science and across the Web in general with Dublin Core.
I’d pose the question back why does hearing about possible innovations get old?
Good points James.
I understand the importance of metadata, but I have to admit, a lot of the small scale projects I deal with, I dont keep metadata (of any standard). I may keep version historys, and document registers to track the source and life cycle of datasets, but the filling out of detailed metadata = ££$$
I leave this for projects where either the client requires it, or I feel the datasets involved (>large numbers, similar datasets, etc) means that it would be in my favour to keep track of metadata.
Not sure if ill get a hammering for admitting that or not, but I know that majority of GIS shops don’t keep metadata unless its a requirement.
Back to Finder – At least a basic level of metadata will be required IMO, as its information from the Interweb – If anyone can upload to this site, then Im going to be a bit weary of the data unless I have some basic info like what it was captured with, date ranges, etc.
Simon: you are so right about metadata. FGDC itself is such a PITA that we don’t usually even keep it up to data unless we have to do so.
A couple years ago, I was having lunch at the ESRI UC with a GIS analyst who was just promoted to “Senior GIS Analyst” before the conference. I asked him what was the biggest thing for him being a Senior GIS Analyst and he answered that he wouldn’t be responsible for Metadata anymore.
I can totally relate to that person in that Metadata, especially the FGDC, is way too cumbersome for the real world.
Besides the complexity and potential overhead of finding/entering large amounts of Metadata – as James points out there is difficulty in even defining what metadata means for the large sets of dynamic information that is being generated by lightweight services and user-generated content.
What is the metadata for a trail in EveryTrail, or a photo in Flickr? Each has their own set of user information, creation time, etc. but they may not be common sets and there are few mechanisms to know that maybe user “bongo785″ is actually a Park Ranger.
And while it may be easy to shirk this off to simple data will have simple metadata, there is a recent post that brings up very good ideas about user-generated transportation information: Social Reconstruction of Public Transportation Information.
The way we’ve approached it in Mapufacture, where we’ve focused primarily on dynamic and UGC sources is to treat it like any native Web data.
Link backs are beneficial to the source provider and also provide information on categorization of the data for search. Users can optionally follow the link to find out more information from the source provider itself.
Microformats have done well to fill this role – especially hCards. It’s hitting a sweet spot of providing lightweight mechanism to add and markup data inline. FOAF failed because it required maintaining a separate file whereas
rellinks are much simpler to add when building HTML.Well, I could care less what metadata standard we use. It could be one already in use, a new one or a combination of all above. Use one and do the work, not bellyache about how hard or how much work it is.
It’s not hearing about new innovation is old. Rather, it’s hearing, because something is too complex or hard we should not do it or change how we do it. Sometimes it’s a valid agrument, while other times it’s a lame excuse.
I agree things should be more automatic, until then, old school it is.
But in the end, I could be pushing two different “ideas” together or even missing the boat entirely. One of those weeks it would seem.
KoS
Looks like Sean post about our fears….
http://blog.fortiusone.com/2008/07/10/collective-stupidity-the-negative-externalities-of-crowds-and-why-hperlocal-could-be-lame/
Looks like a great app…will folks be able to buy/sell data from the site or will it always be free? Looks like a great opportunity to do so and much better than some of the other sites currently being promoted.
Hi Andrew,
We do not have plans to sell any of the data we are creating but we would like to provide a data mart for third parties that would like to sell data. The hope is to create a network effect for data where the free data creates more reason to use for fee data and vice versa. We figure the larger number of combinations of data we can provide the better of the user will be. Those that want to keep to just free data great and those that want to pay for value added data great. Originally we’d though about providing data from traditional commercial sources, but would be interested in hearing if people are interested in selling their own created data.
best,
sean