MapRoom - a multi-user spatial data management tool
Hey all. So Brent and I have continued thinking about this whole spatial data management system we've all been discussing. Here are two mock-ups of what we think such a site might look lile. These are, of course, super rough right now, lacking some features and design considerations our first draft will have, but we'd love to get people's comments and thoughts from the get go. Main Page First page you see, with a simple search bar and a Google map to limit your search spatially. You'll be searching on tags and metadata we derive from the files initially, maybe other sources of metadata later. The Categories tab will be an alternate, categorical view of the data, based on hierarchical categories like data type, source, theme, etc.
Dataset View This is the view of a single piece of data, or file. Shows all the metadata, tags organized by popularity (or most searched on?), and big bright download button. Oh and imagine an "Add a tag" text box there beneath the tags. WMS/WFS/WCS functionality is sort of something we might like later on down the road, but not immediately.
The upload view will look very similar. Either you're remote user and you upload a file or you're an admin and you tell the application that you've put some data you'd like to add in a designated holding directory. The app will read the file, try to figure out as much of the metadata as possible, and then the uploader will have to fill in the required fields that the app can't figure out. It'll look a lot like the dataset view, except those fields will be editable.
Reader Comments (5)
The font for the headers etc. is the always stylish , and the content is all in Verdana. Gill Sans isn't a standard web font, so we probably wouldn't use it as widely on the real site, but it might be ok for the logo or anything that we could render as an image. Unless, of course, we wanted to use something cooky like
I really like this design, Ken-ichi. Here are four points I can think of to consider as you continue to implement it.
1. In addition to the blog Abe mentioned above, the GISC has a Map Room project of its own: . So sadly I really think another name/logo is required.
2. The home page with the spatially-delimited map extents of datasets is . What happens when the user zooms in closer than the full extent of a dataset (i.e. fully inside of it) - can they still "see" it and select it? It could get messy if we start serving a number of California state-wide datasets, for example. But maybe there could be an option to show all datasets that intersect the current view extent?
3. In the dataset view, I think you're wasting valuable space showing the full Spatial Reference System text - my suggestion would be to display just projection and datum by default, assuming that in most [not , I'm sure] cases those two parameters are unambiguous. If there's any doubt, I think you'll provide a way for the user to download the entire metadata file (without necessarily getting the entire shapefile/raster dataset/etc.
4. Finally (for now), there may be different of a dataset, and also, not all metadata are created equal. I think we're likely to serve some datasets that are "works in progress," e.g. an IfSAR-derived DEM that may experience future processing to improve its "bare-earth" terrain interpretation, or a historic air photo that's been rectified with low precision and could be improved later. Any versions that go on our server probably need to remain there for archival or validation purposes, but we want new users to default to the new versions of any duplicated datasets. This may not be a problem in real-life, just something to think about.
By the way, did I tell you how I think this design is? Great work.