Printable Version of Topic

Posted by: JohnVV Dec 5 2014, 07:29 AM

has anyone else noticed that the pds archives are WORSE than the used to be
the imaging node's ftp is dead
and the NEW and "?? improved ??" pds atlas search III is a bit of ( well something unspeekable that comes out the rear end of -- something-- )
http://pds-imaging.jpl.nasa.gov/search/

there is NO zip archive download
and the wget script --- ??? it you can get it to work ????? gives you a script that is ZERO BITS in size

i do miss the 2 min wait of the light blue "over coloring" of the -- "please wait" of the atlas search II

now the FTP page for the PDS geoscience is still working
ftp://pds-geosciences.wustl.edu/messenger/

but the image node is ........... someplace

now i AM trying to be -- shall we say colorful but not nasty

Posted by: JohnVV Dec 5 2014, 07:38 AM

i hate "auto complete / auto correct "

is CHANGES !!!!!

Posted by: Gerald Dec 5 2014, 09:52 AM

Today a http://pds-geosciences.wustl.edu/missions/msl/ is scheduled to be provided.
I guess, the servers have been busy, and probably not double-buffered. Installation/migration can take some time.

Posted by: Phil Stooke Dec 5 2014, 02:34 PM

You're right, John, it's not as easy to use. But the changes to Map-A-Planet are nothing short of catastrophic. You have to register ( a real nuisance if I want a class full of students to use it), and when I specify a data set, lat-long boundaries and resolution, I don't get to see the image - it can only be obtained via FTP, after waiting an hour or two for an email, which makes iterative modification of the boundaries unbelievably inefficient. I can't believe they would change such a powerful tool into something so clumsy. And yes, I did tell them.

Phil

Posted by: JohnVV Dec 5 2014, 05:10 PM

having a whole class have to register... yuck

now for a OLD protocol like telnet
and using it for horizons ( a email address "123@123.com" works )
that is a bit expected .

right now i am having to use "wget" to grab the messenger RGB 8 band data
ftp'ing it would have been nice

hopefully the issues with ftp are temporary
i do NOT want to have to use http for the HIRES messenger gridded data

Posted by: thare Dec 6 2014, 02:55 AM

Ouch But keep the comments coming - we are listening and if possible we can try to figure out how to roll some changes in. We imagine the original MAP will stick around as long as the server stays healthy.

For some long-time MAP users it is going to be hard to transition to MAP2. We had to make some compromises on the interface to support more robust features. Remember, the over-arching goal for both MAP and MAP2 is to make science-ready data available for download. The fact that the almost 10 year old (second version interface) for MAP is so beloved was a little bit of a shock for me. Hats off to the original USGS designers (who are no longer with us - no they are not dead).

For http://astrocloud.wr.usgs.gov/index.php?view=map2 our goals were to:

1. Allow the ability to add new data sets in hours. The hardest part here is making sure the documentation (metadata) is available and references are in place. We will always strive to have each base well describe (some bases are still a work in progress). Do have a good base and documentation, we will add it.
2. Support much larger file creation and downloads. MAP has a 2GB limit and MAP2 has a soft 20GB recommended limit. This is a limitation of the MAP engine which can't make files larger than 2 GB. MAP2 uses ISIS3 which has no filesize limit. The 20GB recommended creation size is to just help with download volumes. Fortunately, we are also working on a moving the created downloads to a faster network.
3. Support all the map projections available in ISIS 3 (including Orthographic, Transverse Mercator, Lambert, Polar Stereographic, etc.)
4. Nearly all MAP2 files are be available for download as one file (ISIS cube and/or GeoTiff/GeoJpeg2000) instead of always stuck behind an ordering web interface.
5. All output formats are "mapping" ready with proper geospatial headers which include the map projection and cellsize. Even the Jpeg and PNG files have GIS projection auxiliary files. Worldfiles just don't cut it anymore.
6. Because we are using the more modern ISIS 3 and GDAL and our cluster for processing, we can more easily implement new functionality (e.g. hillshades, slope maps, colorizing, blending, filters, even perhaps rendered 3D fly-overs, etc.)

In regards the login system, we know it can be an annoyance. But besides it allowing us to better help users and tally popular use cases, it will eventually allow you to recall previous settings and share settings with others. But feel free to share one login/password with a group or classroom. Other benefits include allowing us to eventually support "domain-specific" jobs for users (e.g. recommended settings for GIS apps or Matlab users) and the same login will also get you access to http://astrocloud.wr.usgs.gov/index.php?view=pow (our single image processing service which runs on our cluster for you, also helps to learn ISIS3). We have considered "guest" accounts for pipeline and anonymous usage but users will still need to provide an email address.

I hope this helps explain some of the decisions we made. Again, feel free to post comments, especially if you find issues or have a data/capability wish list.

If you are missing the interactive part of MAP, then this http://planetarygis.blogspot.com/2014/09/tips-to-interact-with-astros-wms-maps.html might help. Open up the WMS layers in one of a dozen+ applications or create your own OpenLayers map to see the glorious 500GB Kaguya Terrain Camera Ortho mosaic in full-res (~7m/p). Process some images in POW and then add them on top of a downloaded MAP2 base or live planetary WMS layer in QGIS! Pretty cool.

-Trent

Posted by: ugordan Dec 9 2014, 04:38 PM

I don't like the new Atlas III search. While I can see where the idea of streamlining the interface comes from, it is currently pretty broken. For Cassini, I can't figure out a way to specify a filter combination (one of the nice things they added in the previous installment of the search engine) and the interface doesn't seem to work when I search for a specific PDS product filename like "N1757515197_1.IMG".

Posted by: James Sorenson Dec 13 2014, 01:27 PM

I was just browsing the new PDS release for MSL in the notebook, and it appears that all the DRCL and DRCX data products have now been white balanced arbitrarily it looks like. What happened to the original MSSS color corrected images? Why has the color changed?These are awful!

EDIT: So.. after downloading a few IMG's, they still appear to be the same color corrected products. Is this white balancing of the preview images another attempt to make the images more viewable to the public?

Posted by: mcaplinger Dec 13 2014, 05:52 PM

Hasn't the notebook always been this way? I'm sure somebody thought it was a good idea.
As data producers we have no control over how the notebook is implemented, so complain to Wash U.

Posted by: JohnVV Dec 14 2014, 09:41 PM

the search 3 seams to be working a bit better
wget scripts are being made

however
the animations ( exploding close )
that "compz-fusion" gtk eyecandy is "cool" for about 15 min.
( think of Microsoft's " Clippy" the paper clip help bot )


i do need to double check on the voyager pds archive is being used
it was using sending IMG with ATTACHED headers
instead of the IMQ compressed std format