[Landsat-pds] Decoupling preview html files from Landsat data

[Sundwall, Jed]

Team: Jeff Layton and I are still working on updating the preview index.html files that we create for each Landsat scene. An early mock up of what we’d like to produce can be seen at http://awsopendata.s3.amazonaws.com/testing-landsat/index.html

One idea that keeps occurring to me is that it might make sense to host the preview html files in a separate bucket from s3://landsat-pds.

1. The index.html files feel a little crufty. They’re the one thing in the bucket that doesn’t feel like data, and I feel like they don’t quite fit in. This becomes especially apparent when looking at log data and seeing records of the html files being accessed along with the data. This creates noise when we’re trying to analyze usage of the data, but have to filter out data from crawlers or other casual browsers.

2. I’d like to be able to demonstrate the html interface to the data as an application of Amazon on AWS rather than a core part of it. The goal would be to show off how objects arranged logically on S3 can provide content that is completely separate from the presentation. Hopefully inviting other people to make their own attempts.

3. It would relive the burden of creating the index.html files from Frank’s ingestor. Let’s let just the ingestor ingest.

4. We could create a nice clean URL as an entry point to the browser.

Any thoughts on this? Please chime in.

— Jed.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/landsat-pds/attachments/20150415/91ff1148/attachment.html>