[COG] Welcome!

Mon Jun 4 17:39:56 PDT 2018

Hello Chris (and list),

As a quick introduction, I'm Scott and I'm the Deputy Director at the
Alaska Satellite Facility.  We operate NASA's Distributed Active Archive
Center (DAAC) for synthetic aperture radar (SAR) data, and have a
reasonably large (6PB) archive of SAR data from a variety of sensors.  If
you are familiar with SAR data you know it can be a bit tough for
less-advanced users to climb that learning curve.  We see COGs as a way to
help bridge the divide between a complicated data set and a growing global
user community interested to see what they can do with SAR.  Some of the
particular issues we have been working on or are interested in working on
include,

- How to best organize our data in the cloud to give people easy access?
- How to reduce egress costs via technologies like COG that allow users to
access just the the portion(s) of the files they want and not download the
whole file.  Better yet, we want to enable users to access data directly
(via interfaces like GDAL /vsis3) so they aren't really downloading
anything to their machine, per se.
- What does metadata look like in a COG world?  Can we embed the metadata
we need in our COGs so they are a one stop shop, or do we need to include
geoJSON or some other format of external file that will contain the
metadata?
-  Finally, what tools can we develop to help our users (and other data
providers) access their data in an intuitive and useful manner.   With a
variety of high-cadence data sets available (like ESA's Sentinel
constellations or Planet's doves) can the community come together to
jointly develop advanced tools for time-series analysis, change analysis,
disaster mitigation, etc?

Hopefully some of these questions are already answered and we can learn
from others on the list.  Personally, I see COGs as pushing the data
provider community forward to make data more accessible to a global base of
users who don't have advanced tools, or high-speed Internet, or data
processing capability, but do have real-world problems they are trying to
solve.  My little soap box there.

Looking forward to the discussions,

Scott

On Mon, Jun 4, 2018 at 3:09 PM, Chris Holmes <cholmes at radiant.earth> wrote:

> Hi everyone! Thanks for joining this list to discuss cloud
> optimized geotiffs (COGs). We've already got 68 people on the list. I don't
> anticipate it will be a super active list, but I believe there's a few
> things to discuss to evolve COGs a bit.
>
> I think I know a number of people on the list, but it'd be great if people
> could introduce themselves and share their interest in COG.
>
> * I'm Chris (aka 'cholmes'), and I work 3 days a week at Planet, plus one
> day at Radiant.Earth, and both give me time to help advance COG. At Planet
> all our data is COG, and so we're interested in more tools that can consume
> it, and hopefully save everyone money by using COG to stream our data and
> not having to copy and store so much. Radiant has been supporting the COG
> standard, building tools like cog-map <http://www.cogeo.org/map/> and
> tiles.rdnt.io. Radiant is building a platform for NGO's / developing
> world, and sees it resting on an infrastructure of COG's
>
> *Using the list*
>
> Feel free to post any questions, issues in working with COG's, best
> practices in working with the spec, or ideas of how to make COG's more
> useful. From the BoF at foss4g-na
> <https://medium.com/radiant-earth-insights/cloud-optimized-geotiff-birds-of-a-feather-at-foss4g-na-def26572ae1b>
>  there's a few topics to continue discussion on - feel free to kick those
> off in their own thread, or I'll frame them in a bit.
>
> *Next Steps*
>
> I believe the next step is to create an area in github for us to work. I
> believe a github 'organization
> <https://blog.github.com/2010-06-29-introducing-organizations/>' will
> serve us best, to be able to group a few different projects. We'll port the
> spec on gdal's <https://trac.osgeo.org/gdal/wiki/CloudOptimizedGeoTIFF>
> wiki to its own repo (cog-spec?), and then I'll move the website repo
> <https://github.com/cholmes/www.cogeo.org>, and I think it would make
> sense to move the validator <https://github.com/rouault/cog_validator> in
> to, as it's where the spec gets real. And perhaps do a repo on performance
> testing, with tools and results. And perhaps we can welcome other related
> tooling.
>
> Does anyone have any great ideas for the organization name? cog and cogeo
> are taken. The ideas I had were:
>
> * cogeotiff
> * co-geotiff
> * cog-repo
>
> Let me know if there are more ideas, and I'll send out a poll soon to
> decide. And also ping me if you're up to help create + organize the repos.
>
> best regards,
>
> Chris
>
>
>
>
> _______________________________________________
> COG mailing list
> COG at lists.osgeo.org
> https://lists.osgeo.org/mailman/listinfo/cog
>
>

-- 
Scott Arko
Alaska Satellite Facility Deputy Director
University of Alaska Fairbanks

903 Koyukuk Drive
Fairbanks, AK 99775
Phone: (907) 474 - 5570
email:  saarko at alaska.edu
http://www.asf.alaska.edu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/cog/attachments/20180604/676ffb62/attachment-0001.html>