Member Nodes

ADC's corpus is quite large with a variety of object types represented. Chris said in an email 4/13/16:

There are close to 500,000 objects in total, and we’ve found 163 unique file extensions:

[1] ".csv" ".readme" ".cls" ".NCAR_iss" ".e00"

[6] ".txt" ".pdf" ".xls" ".zip" ".XLS"

[11] ".jpg" ".jpeg" ".dat" ".doc" ".TXT"

[16] NA ".DAT" ".gz" ".asc" ".gif"

[21] ".tar" ".img" ".NCAR_DryTundraSite" ".png" ".I"

[26] ".D" ".N" ".prn" ".tif" ".NCAR_TundraSiteC2"
[31] ".JPG" ".cdf" ".Readme" ".nc" ".html"

[36] ".sav" ".xlsx" ".tgz" ".prj" ".ctl"

[41] ".f" ".htm" ".1st" ".DOC" ".PS"

[46] ".z" ".ppt" ".DIC" ".shp" ".shx"

[51] ".dbf" ".CSV" ".AVI" ".PDF" ".gdat"

[56] ".aux" ".flt" ".xml" ".metadata" ".hdr"

[61] ".rrd" ".xl" ".kml" ".tiff" ".Final"

[66] ".PICT" ".WMF" ".d" ".mdb" ".buoy"

[71] ".scint" ".steele" ".ascii" ".mat" ".raw"

[76] ".GZ" ".WPD" ".log" ".GIF" ".HDR"

[81] ".UP" ".ASC" ".DOWN" ".mpg" ".sh"

[86] ".hex" ".CON" ".COM" ".bz2" ".RDF"

[91] ".EDF" ".bmp" ".001" ".elg" ".LOG"

[96] ".mov" ".xbt" ".jjv" ".PTS" ".WPT"

[101] ".cfg" ".Z" ".sbedat" ".rcmdat" ".btl"

[106] ".ctd" ".trk" ".ZIP" ".eps" ".rdat"

[111] ".ins" ".000" ".btm" ".dta" ".sum"

[116] ".emf" ".psb" ".segy" ".tpf" ".m"

[121] ".docx" ".fig" ".pptx" ".tlx" ".lis"

[126] ".last" ".0001" ".0007" ".0003" ".0008"

[131] ".0009" ".0002" ".0005" ".0004" ".0006"

[136] ".fas" ".qual" ".MOV" ".MTS" ".MPG"

[141] ".GPS" ".ULS" ".kmz" ".rtf" ".TOT"

[146] ".template" ".cnv" ".xmlcon" ".fasta" ".py"

[151] ".crc" ".cpr" ".db" ".sub" ".accdb"

[156] ".download" ".jgw" ".tfw" ".ovr" ".sbn"

[161] ".sbx" ".mxd" ".BMP"

So, it’s going to take some work to figure these out. For any of the non-obvious ones, we’ve used application/octet-stream as the formatId but will need to update system metadata before we start syncing.

It's likely that most of these would be classified as a DATA format (since they're not METADATA or RESOURCE), but some of them just look odd.