Below are the liens for the following NH REX data sets:
    NH-J-REX-2-JUPITER-V1.0
    NH-J-REX-3-JUPITER-V1.0
    NH-X-REX-2-LAUNCH-V1.0
    NH-X-REX-2-PLUTOCRUISE-V1.0
    NH-X-REX-3-LAUNCH-V1.0
    NH-X-REX-3-PLUTOCRUISE-V1.0

o Really need a file containing the tabulated antenna pattern.

o The sequencing file seems to contain insufficient and occasionally 
  inaccurate information, making the data hard to use.

o The tracking data generation is still an open issue with the mission that
  will have to be decided by them as they flyby Pluto.  PDS is happy to 
  archive what is produced, but cannot place requirements on missions
  unilaterally.  We're not above lobbying a bit, though.

o There are several instructions in documentation along the lines of 
  "see documentation for more details."  More specific references should be 
  substituted.  For example, in the  nh-x-rex-3-launch-v1.0  data set :
    DATASET.CAT: lines 140 and 244
    Quick-Start Guide: line 31 in AAREADME.TXT

o Sami notes that some of the DSN references need updating.

o Add the labels in the data/tnf directory to the index.tab file.

o START_TIME for some raw-reduced pairs do not agree.  This needs to be
  investigated.

o Please add a short description in the documentation that describes how 
  the boundaries of the directories were set, if it does not already exist.

file: 'aareadme.txt'
--> The 'quick start' text should be removed from AAREADME.TXT and provided as a separate document.
--> AAREADME.TXT should be a brief overview of what is included in the data set, how the material is organized, and where it came from.  Details about label construction, for example, are better left to a file in the DOCUMENT directory.
--> "Required Reading/Details" lists /DOCUMENT/NH_FOV.* and /DOCUMENT/NH_REX_V###_TI.TXT, both of which have been removed from the data sets.
--> Errata" includes the phrase "are ideally certified with", which should be replaced by "have".
--> Of eleven URLs listed in "Suggested FITS & PDS software" the first and second are "not found", the fourth leads to a redirect, the fifth has nothing obviously related to IDL, the seventh requires a link to PDS/CN (which fails), the ninth is to software which is being superseded and is no longer maintained, and the tenth is "not found".  The previous paragraph begins "The following URLs were current as of 2007 when the early New Horizons data sets were delivered ..." which is of questionable validity for REX.  Only in late 2014 are the first REX data nearing public release; at least an update is in order.  But this also reinforces my concern that these data be described independently of reading software.  A surprise is that the links to PDS resources are the least reliable.
--> The explanation after ^S_TABLE = ("XYZ.FIT", 51) says the table "comprises 97 rows each of width 1080 bytes".  There are 463 rows.
--> Len Tyler may not be the best contact for the archival data; but that is a team decision.  If he stays, his address should be corrected to
--> --> David Packard Building - Room 331 (MC 9515), 350 Serra Mall, Stanford, CA  94305-4020, USA

file: quick-start document
--> "Fortnight" may not be easily interpreted by international partners.
--> Characterization of those who want to read data without using PDS or FITS tools as 'stubborn' is unnecessary.
--> The example label fragment EXTENSION_RAD_TIME_TAGS_TABLE is invalid; PDS has no 8-byte MSB_INTEGER format.  The fragment is from a real label, which is thus non-compliant.

file: 'catalog/dataset.cat'
--> Need to add "DEBOLTETAL2005" as a REFERENCE_KEY_ID in this file.
--> Need to add "TYLERETAL2008" as a REFERENCE_KEY_ID in this file for launch data sets only.
--> Line 180 typo: change "19.April" to "19 April"
--> Line 181 typo: change "has" to "had"
--> Lines 176-183: Determination of the noise floor, the system temperature, and the receiver noise figure needs to be explained in more detail since they form the basis for the quantitative radiometry measurements.  Ordinarily, a receiver's noise figure would be measured in a laboratory before launch; that could be converted to a system temperature once the receiver is connected to an antenna.  The system temperature would vary depending on what is within the antenna's field of view; but a 'best case' value could be considered the optimum performance value and published - such as in REX_SSR.  It would not decrease by a factor of 3 as the spacecraft drifted from Earth to Pluto.  This is not a subject for DATASET.CAT; but it could form the basis for a more comprehensive review of the calibration results in a new DOCUMENT file.
--> Lines 253-256: Add a note explaining the difference between CDH 1 and CDH 2; it could be done here, in REX.CAT, and/or in NHSC.CAT (which might be the most appropriate place, since C&DH is common to all instruments and the subsystem is not mentioned there).
--> Line 303: PDS Standards make no prohibitions on TARGET_NAME values like "RADEC= ..."; it's just never been done.
--> Lines 405-420: Contact information can be omitted from DATASET.CAT; such information more properly belongs in PERSON.CAT.  In this case, the information is not accurate (see A07, above).
--> Lines 473-475: This seems a dubious claim in the case of REX; the statement is undoubtedly a consequence of the template mode of constructing NH instrument documentation.

file: 'catalog/rex.cat'
--> lines 112-113: "... such gravity investigations are beyond the scope of the REX instrument team" is not true (see G01 above).  The remainder of the paragraph may need to be revised depending on NH decisions regarding the tracking data archive.
--> --> G01: REX_SSR (Table 1), the SOC ICD (section 12.1), and REX.CAT identify mass determination of Pluto and Charon as one of the top NH/REX science investigations and objectives.  The NH Data Management and Archive Plan (Appendix A) lists radiometric tracking data as one of 12 example REX product types to be archived with PDS.  REX team member Patzold has confirmed his need for such data to carry out the mass determinations.  The NH archiving team has repeatedly denied responsibility for delivering tracking data as part of the REX archive, saying "the exact disposition of such data is TBD" (current REX.CAT).  There is similar language in the SOC_INST_ICD_REX document (page 106, second paragraph).  Archival tracking data remains a lien against NH, though not necessarily against NH/REX.  Once disposition has been determined, the banners and other warnings in current documentation can be reduced to brief references.
--> Line 130 refers to the on-board REX signal processing card as a "transceiver".  There is a "transceiver" in the on-board instrumentation used by REX; but it is in the telecommunications subsystem, not the REX-specific hardware.  The word "transceiver" also appears in lines 185, 231, and 502, where its use should also be reviewed.
--> A description of the HGA Beam Pattern Calibration begins at line 171.  REX_SSR Figure 13 shows the 3-dimensional pattern and Table VIII gives the gain as 41.9 dBi.  But there is no quantitative information on the beam shape.  A table, giving the gain versus the two offset angles would benefit users, especially those interested in interpreting the details of the radiometry measurements at Pluto and the scans of natural radio sources conducted during the pre-Pluto tests and calibrations.
--> The term "REX mode" appears in line 185.  It is defined in the section Operational Modes, which follows, where it is distinguished from "radiometry mode".  From the documentation, it is not clear that these are different operational modes; they are different uses for the data being collected in a single operational mode.  That is, I/Q samples and radiometry measurements are always obtained simultaneously.  If there are different operational modes, they would be the mode in which the REX signal processor is connected to the HGA and the several modes in which the signal processor can be configured for self tests.  And there is presumably a mode in which the unit is turned off.
--> Line 196 mentions a "calibrated reference" against which I and Q values are calibrated.  I see no other mention of that reference in REX.CAT; and I don't recall seeing mention of it in other documents.  Please describe briefly.
--> Line 199 says that the radiometry samples are calibrated using "a reference temperature calibrated from the noise figure of the New Horizons radio receiver.  This seems circular; but it's also not clear that the noise figure of the receiver is either known or stable (see section "Radiometry Calibration" in DATASET.CAT).
--> line 280 typo: "suits" should be "suites"
--> line 282 typo: Wording anomaly - change to "observe the occultations of the Earth and Sun by Pluto and Charon"
--> Line 352 typo: "1024s" should be "1.024s"
--> Sentence beginning on line 367: Change to "... incrementing ten times within each frame and once between consecutive frames."
--> Line 372 typo: change "is" to "are" and "frame" to "frames"
--> Line 491 typo: The classical heterodyne process is analog; so the use of "heterodyne" in line 483 is correct; two analog signals have been multiplied to obtain a signal at the difference frequency.  I'm not sure the down conversion of 10 MHz digital samples to baseband qualifies.
--> line 502 typo: the 10 MHz samples have 10 bits; at line 517 they have 12 bits.  A note that the 10 bit values are stored in 12-bit registers would reduce confusion.
--> Line 511: The (1/N) normalization in the equation does not appear to have been implemented in the processing software.  The maximum 10-bit twos-complement sample value is |s(i}| <= 512.  The maximum squared value is  s^2(i)<= 262144.  The sum of 10N=10.24*10^6 such values over an ROF should be no more than 2.7*10^12.  After the (1/N) normalization, Pup(k) <= 2621440.  But a typical value (from 20060629_001388/ rex_0013887723_0x7b1_eng_1.fit) is 85462609456, more than 30000 times larger than the expected maximum.  There is no real reason for the normalization; so this equation should be corrected in REX.CAT and a note should be added to ERRATA.TXT that equation (11) in REX_SSR should also not have the (1/N) factor.  Had the normalization been carried out, Pup(k) could have been stored in a 32-bit integer.
--> line 242 typo: change "Dcocument" to "Document"
--> Line 674 typo: What is the meaning of "[SWA]"?
--> If there actually are redundant sets of REX hardware, shouldn't that be mentioned in INSTRUMENT OVERVIEW - FLIGHT ELEMENT?  How does that affect the power, mass, and volume numbers in SPECIFICATIONS and the discussions in "Operational Considerations - REX" and other sections below?
--> Line 715 typo: Change "blind" to "planetary".  All modes except conscan are blind; presumably the DSN uses slowly changing RA-DEC values to track NH.
--> Line 851 typo: change "tbd" to upper case.

file: 'document/nh_met2utc.lbl'
--> The final "NB" in the TABLE DESCRIPTION states "The correlation in this file is extrapolated into the future, relative to the PRODUCT_CREATION_TIME ..."  But the label includes no PRODUCT_CREATION_TIME, so it is impossible to tell when the extrapolation starts.  Add the PRODUCT_CREATION_TIME keyword to the label.

file: 'document/soc_inst_icd*.pdf'
--> typo: Section 12.2.1.1.3, paragraph 3 - change "a integrated" to "an integrated" and change "value of associated" to "value associated"
--> Two lines before Table 12-5 - change "to and/or" to "to specify and/or"
--> Last row in Table 12-5 - Add "for i=I" to the second column
--> typo: Section 12.2.1.2.3, second line - Change "radiometry" to "time tag"
--> typo: Section 1.2.2.4 heading - Change "FIT" to "FITS"
--> typo: Table 12-7 caption - Change "Extension" to "Extensions".  Explain "calibration-relevant".
--> typo: The single sentence in Section 12.2.4.1 doesn't make sense.
--> typo: Section 12.3.1.1, after the bullet - Change "Scale" to "Scales"
--> typo: Section 12.3.1, first bullet - Change ", pairs" to "pairs,"
--> typo: Section 12.3.1, third bullet - Omit "converted to scientific units (see section 12.3.2.3)", which is redundant (repeats the first bullet).
--> Discussion of how time tags are reset is not definitive, though I have confidence that relative time tags are reliable.  There is apparently some uncertainty about initialization of time tags; but once ROF generation is underway, they increment as expected.
--> Section 12.3.1.1 - The bullet statement says ADC output values have 14 bits.  The ASCII figure in REX.CAT shows 12-bit samples coming from the ADC at 10 MHz (but the text implies only 10-bit resolution) and 16-bit I and Q values coming from the 1 kHz Digital Filter.  Why "14" here, a value which does not appear anywhere else?
--> middle of page 115 (AGC) - Explain "Side A or B", neither of which has been mentioned previously in the ICD (or in REX.CAT or DATASET.CAT).  From the ApIDs, it appears that Side A is equivalent to CDH 1 and Side B is CDH 2.  But the different radiometer calibration coefficients suggests that A and B represent different RF and/or processing paths and not just different storage.  Do A and B represent different operational modes (see R04 above)?  Can they run simultaneously?  PAYLOAD_SSR confirms that there are two redundant pieces of REX hardware, but it says nothing about operations.  PAYLOAD_SSR says they can be operated simultaneously to increase SNR; has that been attempted and, if so, during which sequences and what are the results?
--> Section 12.3.1.3 - Add an explanation of how absolute time can be derived from the (relative) Time Tag values - presumably using START_TIME in the labels.

   
 Last updated: 2015-01-08, SBN:T.Barnes


=======================================
Liens general to all New Horizons data sets are listed below:

file: 'aareadme.txt'
--> typo: " to to " in "Refer to to DOCUMENT/DOCINFO.TXT"
--> typo: " by the by " in "and used by the by NH instrument teams."
--> typo: In the 'Generic PDS label details' section, par 1, the text "mission- specific" has an extra space and should read "mission-specific".
--> --> For screen print, see 'frahm_extras/PEPSSI_README.pdf'.
--> Utility of the raw vs. calibrated data products: I was unable to find any description of when a user might benefit from using the raw data products (contained in the nh-x-alice-2... nh-x-leisa-2... directories) rather than using the reduced and calibrated products (contained in the nh-alice-3... and nh-x-leisa-3... directories). My guess is that such instances are likely to be very rare, and users would probably have to work with NH project personnel to do much with the raw products. If that's correct, it would be useful to have a brief statement in the raw data archives saying that for most applications users would be better off using the reduced and calibrated data. It would also be useful to have a brief statement describing when starting with the raw data might be called for (if a summary of those conditions is difficult to compile, perhaps a couple of examples could be given).
--> Filenames and Product IDs section: It would be worth noting that the filename used is an example relevant to Alice. The same filename is used in this file in the Liesa data archive, and perhaps all (?) the others.  It would be helpful to enumerate the instrument IDs, perhaps in another note, or to tailor the example filename used to the instrument that produced the data in the archive being examined (i.e. in a Liesa archive, use a filename like "lsb_0052459065_0x53c_sci_1.fit" rather than an Alice file).
--> Processing Level: Here the Raw and Calibrated data products are associated with something called "CODMAC Level 2" and "CODMAC Level 3", respectively. I have no idea what CODMAC is or why it's relevant here. Maybe explain or delete these associations.
--> Contact information: Alan Stern's address is given as NASA HQ. I think that's incorrect, in which case an updated address and/or an alternative contact should be provided.

file: 'catalog/dataset.cat'
--> typo line 3: capitalize 'H' in "horizons"
--> The Review section in the CONFIDENCE_LEVEL_NOTE should have the TBD date updated after the review.
--> The CITATION_DESC has the data set name in all caps except for only one word (either "Raw" or "Calibrated").  Probably should make it uniform.

file: 'catalog/nh.cat'
--> typo line 90: Should the "11 to 12 Earth days" actually be "12 to 13"?

file: 'catalog/nhsc.cat'
--> In nhsc.cat, the second ASCII art image doesn't seem to be rendering correctly in fixed-space font.
--> typo: line 27.  There is a double space in between two words that should a single space, "requiring  maximum".
--> --> For screen print, see 'frahm_extras/PEnhsc.pdf'.
--> typo line 42: change "PEPPSI" to "PEPSSI".
--> line ~76: The RALPH/LEISA section presents the same erroneous information about the etalon bending the light that I note under comments on the leisa.cat file, and should be fixed.
--> --> leisa.cat lien: The Description section states, incorrectly I think, that "... its etalon acts like a prism to bend different wavelengths of light by different amounts..." Etalons are interference filters that pass particular wavelengths through wile reflecting others. It might be more accurate to say something like, "...its etalon (a filter with a narrow spectral bandpass that varies linearly in one dimension) is bonded to the illuminated side of the IR detector. As a result, each row of detector pixels receives only light of a particular wavelength. Spectral maps are produced by sweeping the FOV of the instrument across a scene, sequentially sampling each point in the scene at each wavelength."
--> typo line 125: The "Long" in "short for Long Range" should be "LOng" with a capital 'O'.
--> typo line 144: The word "Science" is missing from the instrument name "Pluto Energetic Particle Spectrometer Science Investigation (PEPSSI)"
--> line 315: Unable to locate Figure 5.
--> line 323: Unable to locate Table II.

file: 'catalog/ref.cat'
--> Incorrect Reference Year for BELTONETAL2002.  Can not find a 2002 reference which has more than 100 pages.  There is a 2003 reference which exceeds 145 pages.
--> Incomplete Journal Specification for ANDREWSETAL1998.  Should site 'Proceedings SPIE' as the journal.
--> Note that inaccessible JPL documents pose ongoing problems especially when their old references are included in new data by habit or cut & paste.  The REX reviewers should consider whether there is any point in referencing DSN design documents not available to the great majority of end users.
--> Change all occurrences of "PEPPSI" to "PEPSSI".

files: 'data/*/*.fit'
--> Remove ITEMS keywords from any COLUMN that is not actually a vector.
--> Labels missing units of measure should be present wherever they are applicable.

directory: 'document/'
--> The following PDF files are not PDF/A compliant:
--> --> asmar_and_renzetti_1993.pdf
--> --> c2_qe.pdf
--> --> c3_qe.pdf
--> --> flat_field_from_ot2_data.pdf
--> --> morgan_spie.pdf
--> --> pepssi_ssr.pdf
--> --> ralph_align_wrietup03.pdf
--> --> rex_ssr.pdf
--> --> sdc_ssr.pdf
--> --> tnfsis.pdf 

files: 'document/soc_inst_icd*.pdf'
--> A test was conducted on the ICD intro pdf to determine if an acronym was included within the ICD acroabbr pdf file.  The acronym chosen was "CM" and occurs on page 6, section 6.6, line 3 of the ICD intro pdf.  This acronym was not found in the acroabbr pdf file.  

file: 'document/soc_inst_acroabbr.pdf'
--> Within the database acronyms should have a unique meaning to minimize confusion.  In these cases we should pick a single acronym for a single phrase and do a find/replace to get rid of the rest.  Please note that these are found in many documents.
--> --> "ApID" and "APID"
--> --> "G&C", "GN&C", and "GNC"
--> --> "MOPs" and "MOPS"
--> --> Bigger problem for SOC and SOCC - let's not use the same acronym for both "Science Operations Center" and "Solar Occultation Channel".
--> --> In the various documents we see SDC, VSDC, and VB-SDC all to refer to the same instrument - pick one.  I prefer SDC, which is used most of the time (including the literature paper), none of the variants occur on the acronym list.

file: 'errata.txt'
--> The errata.txt files tend to be empty or contain irrelevant notes that are not an errata. Ideally this file should not have to exist at all the time of archiving.  It certainly should not contain anything that is not an actual erratum. 

Other General liens:

o URLs and references to the peccadilloes of specific contemporary software
  packages do not belong in PDS archive labels.  Please expunge them with
  all speed.
   
Last updated: 2015-01-08, SBN:T.Barnes