Steve understandably tries to back up his faulty logic with a gem of a post:
So I think he is missing my point – this is not about some new great place to go look at interesting maps. This is about the democratization of GIS tools.
Heat mapping will bring GIS to the masses? Well there you go. I always assumed it would be airplane sightings in Google Earth, but Steve being the visionary he is, goes on record with his thoughts.
19 Comments
I think he called you an “elitist SOB” if I’m not mistaken. You gonna let that slide?
If by you mean me thinking about quality over quantity? Look this isn’t an argument over using classic ArcInfo vs GRASS or MapObjects IMS vs Mapserver.
Garbage in and garbage out. That isn’t to say good analysis might be made using GeoCommons. I’m saying minimum requirement for GeoCommons is so low, we’ll just end up with Geo-spam.
This is an interesting topic. How can you open GIS tools to the unwashed masses (a good thing), without sacraficing quality/honesty/data integrity (a bad thing). Well, GeoCommons is probably a nice mixture of both- it is up to the ‘elitists’ like us to make sure that 1) the bar is raised in terms of how the data is interpreted, and 2) working with the people who make sites like geocommons to include disclaimers / notes on the assumptions of what they are presenting.
We cannot expect every person who uses these services to become experts immediately, or even ever. Yet- Steve makes a good point, in that it is a ‘good thing’ when a biologist can upload some points and see a pattern. We can apply ample amounts of pressure when said biologist submits a paper based on the output from geocommons.
ok now shake hands and make up you two!
dylan
OK, I’m going to say it. James and Steve are two peas in a pod when I’ve seen them at the UC or Dev Summits. I’m guessing GeoCommons has paid them to blog about this and its all about getting traffic to GeoCommons.
K, James and I are friendly with each other but when he is being a short-sighted ESRI fan boy I have no problem calling him on it. All hail the cult of James!
Wow, Steve do us a favor and look up Ad hominem argument. This crap about the “priesthood” and “elitism” doesn’t help your point. The fact is data without metadata is worthless. I view it on the same level as a measurement without a unit. I really don’t know who would trust any worthwhile analysis to unknown data anyhow.
As to the value of heat maps on geocommons, you’ve got to be kidding. If the “small business owner, city worker, or field biologist” cannot invest the ridiculously small amount of time it takes to get into one of several low entry point ($0 – $300) GIS applications that allow controlled interpolation, I have serious reservations about the quality of work done by someone with so little vested interest.
Did I miss the part where everyone learned how to use statistical models? Anyone remember the pre web 1.0 days when we weren’t all armchair specialists?
seriously, does anyone care about metadata except the feds? also, just because it has metadata doesn’t make it more accurate than if it doesn’t. all it means is somebody was bored enough to type it in.
True, metadata doesn’t make the data itself more accurate. By having good metadata, it’s another tool to assess the validity/usablility of the data.
KoS
Metadata will never be updated and complete until it’s done so automatically. I don’t care how easy you want to make it the vast majority won’t bother. Show me a GIS product that does a good job of maintaining and automatically adding metadata? I don’t know of one. GeoCommons done right would automatically reproject & normalize the data for you and append metadata describing the process and parameters used to the output KML file. Appending metadata isn’t technically hard, but no one does it and I’ve never understood why.
Jack..are you kidding me…your saying people dont care about metadata? How do you get intelligent information from an feature dataset, or an image or how do you extract information from a data source. How can i extract a 3D point from an image if i dont have metadata? How can tell you what has changed in an area if i dont have metadata? The problem we have is that we have ‘dud’ formats such as shapefiles, covereage’s, dwg, dgn etc etc that know nothing about data. When these formats were put together they obviously did not think too far into the future of mapping. To do proper mapping you need to be able to persist and track metadata from the moment the source content is collected all the way through the life of that data asset. Until someone solves that problem we will continue to struggle with this problem. Just my 2 cents….
Consider the amount of information that we require to feel comfortable doing something as fundamental as eating food…all we need typically fits on the back of a soup can. Perhaps that is how simple metadata needs to become before it is ubiquitous.
From what I see, making heat maps is pretty simple (I haven’t made one). You aren’t really doing any real statistical interpolation here – you aren’t making a prediction surface of any sort, you are just making some colours with parameters defaulted for you. Kind of like the defaults in SA no? I mean, how many “GIS Professionals” that interpolate stuff in SA really have a clue about what they are doing? I would argue that many do not, and that many could not tell you the difference between a spline, IDW, kriging, etc. Further dumbing down of GIS IMO.
How many of us leverage google maps, yahoo maps, virtual earth, map quest, et et et without truly looking for metadata? Sure, the respective services may source from a second party such as tele atlas or navteq but what do we know about their means of generating data? More often than not the data speaks for itself and metadata is often leveraged to confirm what one already suspects. Metadata certainly has its place but it could be argued that current standards are rather bloated and need to be simplified for both author and consumer and automated as well.
Not sure about you, but I’d never use any of those sources for any NEPA documentation.
Vector is agree with you…have you ever seen the ISO standard for metadata? Its huge…i sometimes wonder if they people sitting around the table defining the standard have ever gone as far as to implement it. Its quite ridiculous really….i would go as fas as to say the OGC has also run into this problem. Its essentially a bunch of guys who like throwing theory around withour actually implementing the standards themselves. Just my 3 cents…
Metadata need not be complex (it depends on the data of course). Asking for metadata is like asking for a recipe for your favorite dish. While the original recipe may have covered 4 pages in a cookbook and described in excruciating detail (similar to the ISO standard) all your really need (and want) can be scribbled down on a index card.
dwg and dgn? Sorry to say those are CAD based formats. Not origionaly intended to for spatial formats. But hey, the spatial world somehow found a way to utilize the data. Funny how most CAD drawings are not accompanied with process lineage after process lineage, but yet I know how valuable utility infrastruture CAD data is to gov agencies.
I think the added value to our field having this open data structure is that when I get asked about accuracy in data. I can 1) show them our data process and legacy info 2) ask them to gmap their home address 3) ask them which data they feel more confident with.
In the end IMHO i would think the majority of people using services such as geocommons understand the source.
side note – I bet most everyone has a disclaimer in their metadata? Know matter how hard we try, we’ll just never acheive 100% confidence.
As a rare breed who is both a GIS practitioner with 20+ years experience and a licensed land surveyor, I see the good and bad of both sides.
And, another thing that I know well is that content is king. Garbage content is worthless, as is great content that nobody can figure out what it is when handed off. Enter metadata, survey-grade and the rest…
Ultimately it always comes down to CAD and GIS being mere tools for getting the job done. As such, it can get in the way, but it can also make the job painless. Metadata has historically been so onerous, with certain fields that, when opened (ATTRDOMV and others) that lead on and on into additional things that demand to be populated, even if the items that go into them are meaningless or irrelevant.
Similarly, with regard to locational data quality, the surveyors have been reluctant to jump into the game and instead just sit on the sidelines and gripe about what crap the GIS contains. I have given many a surveyor a backhander over this.
Just can’t win…
But leading into the future, the more that we focus on content, quality and documentation that lets people make informed decisions, and make the tools transparent (and they are already becoming commoditized, SORRY ESRI…), that is where things are headed.
As Gretzky said, skate to where the puck’s going, not to where it is.
3 Trackbacks
[...] I can pick on James if I want to Posted June 5, 2007 What would blogging and the web be without ad hominem attacks? [...]
[...] Should All Be Surveyors’ | Jun 06th 2007 Partaking in James’ and Steve’s little back-n-forth action in the back channels. I have to rile them with, [...]
[...] think this discussion follows similar lines as the discussion James and I had the other day. Geocommons will not replace Spatial Analyst in the hands of a person with [...]