[blindlaw] Opinion issued in my FOIA case v TSA; major implications for electronic format

[Sai]

PS Someone asked for a public link. So here you go:

https://plus.google.com/+saizai/posts/cFjYiRdEQqv
https://twitter.com/saizai/status/898707013160534018
Sincerely,
Sai
President, Fiat Fiendum, Inc.


On Sun, May 27, 2018 at 1:03 PM, Sai <sai at fiatfiendum.org> wrote:
> ** Background **
>
> 4 years ago, I sued the TSA under FOIA & Privacy Act for (a) various
> documents related to how it treated me at airports, and its
> investigation thereof, and (b) all of its policy & procedure
> documents. (The requests are now 5 years old.)
>
> TSA filed a motion for summary judgment 2 years ago. I made no
> cross-MSJ, but only a heavily factual objection to the format of
> productions, sufficiency of search, and virtually all withholdings.
>
> The opinion came out a couple days ago. Won some, lost some, and some
> kept open for more proceedings.
>
> Opinion is attached and at
> <https://drive.google.com/open?id=1sPBn2UDHjl305ewboGUH3q0JKUIUeBfq>.
>
>
> ** Major win **
>
> Most notable is the decision that DHS' FOIAXpress and eReview software
> is a per se violation of 5 USC 552(a)(3)(B)'s requirement to produce
> records in the format requested — namely, native format — because it
> irreversibly transforms records into rasterized PDFs. Same goes for
> the government's failure to produce spreadsheets in spreadsheet
> format.
>
> "Increased FOIA-processing costs" are explicitly precluded as an
> acceptable defense.
>
> This has major implications for the availability of native,
> accessible, machine-processable records, and for agencies'
> handicapping their own FOIA offices.
>
> It's possible that TSA may yet defeat my format claims in this request
> — by showing that it'd be too burdensome, in terms of processing costs
> *outside of* FOIA processing.
>
> If they don't fix their software & practices, this'll happen again. To
> prevent that, I intend to follow this up with a permanent injunction
> against the use of either software in FOIA processing — or, indeed,
> *any* software that's incapable of making 508 compliant, native
> electronic, metadata-preserving redactions.
>
>
>
> ** Main takeaways **
>
>
> * A. Requests *
>
> 1. For format, FOIA request must explicitly demand:
> * “native” electronic records
> * “fully digital, non-rasterized, text PDF”
> * “original format”, as normally kept by the agency, i.e. before the
> FOIA process
> * metadata
> * § 508 accessible records, together with a statement of the
> requester’s own specific disability need for § 508 accessible records
> (such as blindness)
> * spreadsheet / database format for spreadsheet / database data
>
> 2. For issue preservation, FOIA request must explicitly be addressed
> to parent agencies.
>
> 3. To get FOIA processing records, asking for “all documents and
> communication[s] related to or responding to the FOIA requests,
> whether internal or external” is enough.
>
> 4. New FOIA request must be made if more incidents happen while the
> request is pending, even pre-search, that would expand its scope.
>
>
> * B. Law *
>
> 1. FOIA offices’ use of noncompliant software — in particular, DHS’
> use of FOIAXpress & eReview — is not, by itself, adequate grounds for
> 552(a)(3)(B) noncompliance as to producing native format records. Nor
> is an interference or burden based on “increased FOIA-processing
> costs”.
>
> 2. Illegible documents must be produced legibly.
>
> 3. Due to 49 USC 46110, District court lacks jurisdiction to determine
> whether TSA acted without authority under 49 USC 114(r)(4) in
> designating SSI.
>
> 4. Neither agency production to a third party, nor the third party’s
> publication, moots agency’s obligation to provide records to
> requester.
>
> 5. (b)(6) withholding of agency officials’ name & official contact
> info not justified unless agency proves “that the release of their
> information would subject them to a real risk of annoyance or
> harassment” or “implicate a substantial privacy interest.”
>
> 6. Evidence that CCTV video was required by agency policy to be made,
> and demanded & agreed to be preserved (within its retention period),
> is not enough to show spoliation or agency bad faith, when the
> agency’s Vaughn only claims that it searched for the video and found
> only one, even though the Vaughn disputed neither the existence of
> many other agency-required videos nor the preservation demand.
> Discovery into this also isn’t allowed.
>
>
> * C. Litigation *
>
> 1. MSJ opposition must explicitly identify what documents are:
> * spreadsheets
> * post-decisional
> * 552(a)(1,2) material.
>
> 2. FOIA office must
> * search offices that the initial search identifies as having been involved
> * state times of search & cut-off dates in its Vaughn
> * document search terms used by itself and by tasked offices
>
> 3. No discovery is allowed for claims made in an agency affidavit
> without a detailed showing of agency bad faith, even if the affidavit
> fails to directly address the problems raised.
>
> 4. 5 C.F.R. § 293.311(a) is not applicable to a (b)(6) withholding on
> non-OPM files.
>
> 5. Problems with the contents of agency records, e.g. 5 U.S.C. §
> 552a(e)(7) (keeping records of First Amendment protected activity),
> can’t be litigated unless it’s pled in the complaint, even if the
> records haven’t yet been produced.
>
>
>
> ** Holdings **
>
> This is only summarizing what the opinion says. I think there are
> several major problems with it, which I intend to raise when I can,
> but I won't get into them here.
>
> If you see any flaws or issues of concern — or if you know anything
> that might be relevant — please contact me privately at <legal at s.ai>
> and let me know.
>
> My intent is to get the maximum possible public interest benefit —
> both by establishing case law and by injunctions — not just one-off
> cures for my particular situation.
>
>
> * A. Format *
>
> 1. Plaintiff may not require Rehab Act § 508 compliant responses,
> because it was not pled in the (pre-production) complaint nor added by
> amendment; their affidavit does not specifically state that they are
> blind; and they may not assert their disabled audience’s rights to
> accessible records. P. 11–14.
>
> 2. Plaintiff may not assert E-FOIA as to TSA’s Vaughn indices & other
> litigation filings. p. 15.
>
> 3. Requests for “digital copy”, or requiring request to be “serviced
> electronically to the maximum extent possible”, are not sufficient to
> invoke E-FOIA for “native” format with embedded metadata. P. 15–16.
>
> 4. TSA production of concatenated, rasterized PDFs does not satisfy
> format request specifying “in an electronic, machine-processable,
> accessible, open, and well- structured format to the maximum extent
> possible”, “individual PDFs per distinct document”, “fully digital
> text PDFs rather than scans or rasterizations”, & “lists and
> structured data as machine-processable spreadsheets”. P. 16–17.
>
> 5. Objection to failure to produce spreadsheets can’t be adjudicated
> due to failure to explicitly state what response documents were
> spreadsheets. Leave granted to state more clearly. P. 17.
>
> 6. Request for documents “in their original electronic format”, “e.g.
> Word, Excel, or electronic PDF”, may be enough to require “native”
> documents. P. 17–18
>
> 7. However, this is not enough to require metadata, and metadata
> possibly unavailable per CREW v DoE, 905 F. Supp. 2d 161, 172 (D.D.C.
> 2012). P. 18
>
> 8. TSA’s use of FOIAXpress by the FOIA office, and eReview by SSI
> office, does not justify 552(a)(3)(B) noncompliance for native
> records. Agency’s requirement to produce format if it “is readily
> reproducible by the agency”, especially when the request is for the
> original format in which the agency kept it (i.e. production, not
> re-production), does not depend on “whether reproducing the file in
> that format would complicate the agency’s FOIA review process”.
> However, TSA might be able to sustain this if it shows “significant
> interference or burden” that is “beyond increased FOIA-processing
> costs”, and if it can show that “administrative costs of that type
> constitute a legally sufficient basis for rejecting a format request”.
> P. 18–23.
>
> 9. Pages that were produced in illegible format (due to extremely poor
> scan resolution) must be produced legibly. P. 23–24.
>
>
> * B. Adequacy of search *
>
> 10. As a contractor, Covenant is not covered by FOIA. p. 28–29
>
> 11. Failure to search DHS Office of Civil Rights and Liberties is OK
> because request was directed to TSA. p. 29
>
> 12. A request for “all documents and communication[s] related to or
> responding to” FOIA requests, “whether internal or external”, “can
> only reasonably be construed to encompass FOIA processing records”, so
> TSA should have searched FOIA office. P. 30
>
> 13. Because correspondence with Pelosi’s office came up in the search,
> as did Office of Chief Counsel and Office of the Executive
> Secretariat, TSA should have also searched the Office of Legislative
> Affairs, OCC, and ExSec. P. 30–31.
>
> 14. Search cut-off date is date search starts. TSA failure to state
> the times of its search and cut-off dates make its Vaughn inadequate
> to support its failure to produce records, when a re-request was made
> for which search was not conducted until well after the last date on
> produced records. P. 32–33.
>
> 15. Requests for records related to a BOS incident, when the agency
> delayed search until after another BOS incident happened, does not
> expand the request to include the new incident. P. 33–34.
>
> 16. A Vaughn index that fails to document the search terms, absent
> enough other documentation, precludes grant of MSJ. P. 34–35. Agency’s
> documentation was insufficient for the searches conducted by other
> tasked offices which didn’t document their search parameters.
>
> 17. Request for “all policies and procedures” is “vastly overbroad”
> and “unreasonably burdensome” because it didn’t allow “a professional
> employee of the agency who [is] familiar with the subject area of the
> request to locate the record with a reasonable amount of effort”.  P.
> 37–39.
>
> 18. TSA re-opening the request post-litigation at its “discretion” did
> not waive its objection to breadth. P. 40–41.
>
>
> * C. Withholdings *
>
> 19. District court lacks jurisdiction, due to 49 USC 46110, to
> determine whether TSA acted without authority under 49 USC 114(r)(4)
> in designating SSI. P. 42–46.
>
> 20. TSA’s production of formerly-designated-SSI records to third party
> (ACLU), which then released them, does not moot its obligation to
> provide Plaintiff with those records — if the designation change
> happened before the search. P. 46–49.
>
> 21. Plaintiff may not take discovery as to a matter affirmed by agency
> affidavit, without first making a detailed showing of agency bad
> faith. P. 52–53.
>
> 22. Factual material may not be withheld unless “the disclosure of
> even purely factual material may so expose the deliberative process
> within an agency that the material is appropriately held privileged”,
> so (b)(5) withholding of information gathered in course of Rehab Act
> investigation is not adequately justified. P. 53–54.
>
> 23. Plaintiff’s “contention that “policies, memoranda of law, and
> similar documents” are “post-decisional” and thus not protected by the
> deliberative process privilege” “is far too amorphous to defeat TSA’s
> motion for summary judgment.” P. 55–56.
>
> 24. Inconsistencies in redacting identical information are not
> sufficient to object to (b)(5). P. 56.
>
> 25. Argument that agency delayed processing of administrative
> complaints is not enough to show obstruction of justice to pierce
> agency attorney-client privilege. P. 56–57.
>
> 26. 5 C.F.R. § 293.311(a) “applies only to information contained in
> “official personnel folder[s],” “performance file system folders,”
> “their automated equivalent records,” and other “personnel record
> files . . . which are under the control of the Office” of Personnel
> Management (“OPM”)”, and therefore is not applicable to contest a
> (b)(6) withholding in other files. P. 59–60.
>
> 27. Plaintiff’s knowledge of (b)(6) withheld information is irrelevant. P. 60.
>
> 28. (b)(6) withholding of “similar” information made available in
> other cases is not sufficient to justify an objection. P. 61.
>
> 29. (b)(6) withholding of parts a record, when other parts release the
> same record without the withholding, is moot. P. 61.
>
> 30. The court has no basis to question the agency’s claim that
> Plaintiff's civil rights complaint, and TSA screeners' incident
> reports, are records “compiled for law enforcement purposes”. P. 62.
>
> 31. (b)(7)(C) withholding of names of local police is assumed, without
> deciding, to “subject them to annoyance or harassment in either their
> official or private lives”, and therefore allowed. P. 62–63..
>
> 32. (b)(6) withholding of the whom someone hoped “enjoyed herself
> today”, or “why another TSA employee was unavailable to provide a
> statement”, “reveals little or nothing about an agency’s own conduct”,
> and is therefore OK. P. 63.
>
> 33. (b)(6) withholding of other TSA officials’ info is not justified
> on the record, because it has not explained “that TSA contract
> employees, the DHS Office of Chief Counsel employee, or the TSA
> Disability Branch employee played “a particular role in an incident”
> or “complain[ed] about a particular incident” such that the release of
> their information would subject them to a real risk of annoyance or
> harassment”; or “why TSA employee names and professional contact
> information contained in the policy documents implicate a substantial
> privacy interest.” P. 64.
>
>
> * D. Segregability *
>
> 34. The Government “bears the burden of justifying nondisclosure”,
> “must show with reasonable specificity why the documents cannot be
> further segregated”, and “must provide a ‘detailed justification’ for
> [withheld records’] non-segregability”, but “segregation is not
> required where the ‘exempt and nonexempt information are inextricably
> intertwined, such that the excision of exempt information would impose
> significant costs on the agency and produce an edited document with
> little informational value”. P. 65.
>
> 35. Plaintiff’s presentation of a previously released SOP, which was
> SSI designated only in part, is not relevant because it is about a
> “defectively overbroad” request, and therefore “Sai fails to identify
> any evidence that the TSA withheld records in whole based on a valid
> FOIA exemption, where a portion of those records were reasonably
> segregable and could have been released without disclosing the exempt
> information. P. 66.
>
> 36. TSA’s Vaughn index is enough to bear the TSA’s burden of proof to
> release reasonably segregable portions, p. 66, even though it gives no
> detail at all about whether there are any segregable portions of
> documents withheld in full.
>
> 37. Re CCTV video, Plaintiff’s statements that
>  “(1) TSA policy requires 15 “[c]amera views” of each passenger during
> screening and requires that the surveillance video be maintained for
> 30 days; (2) [they] submitted a FOIA request seeking the Logan Airport
> surveillance video within that 30-day window; (3) the TSA released
> only one video showing [their] screening at Logan Airport and informed
> him that no other video exists; and (4) “the only possible conclusion
> is that TSA and BOS committed spoliation” …
> “cannot be squared with McCoy’s declaration, submitted under the
> penalty of perjury, which recounts that “BOS searched for responsive
> records including closed circuit television (CCTV)” and located “one
> CCTV video of the incident.”.
>
> Because TSA is entitled to “presumption of good faith”, even though
> “TSA policy may have required additional “views,” and it may have
> required that video records be maintained for thirty days”, “that does
> not mean that the videos Sai sought, in fact, existed at the time
> [they] filed [their] FOIA request, and it certainly does not mean that
> the TSA destroyed some (but not all) of the videos Sai sought in order
> to avoid its obligation of production.”
>
> This is not enough “to give rise to a disputed issue of fact regarding
> an agency’s good faith” because if “a FOIA requester’s reasonable
> belief that other records should have been found were sufficient,
> discovery would likely be the norm, rather than exception, in FOIA
> cases.” P. 66–67. Therefore — although this is on TSA’s MSJ — “Sai has
> failed to rebut the presumption of good faith applicable to the TSA’s
> explication of its search efforts and has failed to show that
> discovery is warranted in this FOIA/Privacy Act action”. P. 68.
>
> 38. 5 U.S.C. § 552a(e)(7) (keeping records of First Amendment
> protected activity) can’t be raised because it wasn’t in the
> pre-production complaint. P. 68.
>
>
>
> ** Conclusion **
>
> * A. For my FOIA requester & litigator kith: *
>
> I hope the takeaways section above is useful. You may want to update
> your request, complaint, & MSJ opposition templates / checklists.
>
>
> * B. For my disabled kith (especially fellow blindies): *
>
> Although I didn't win on Rehab Act grounds per se, I hope the
> electronic format win should lead to more availability of accessible,
> native documents.
>
> It'll take some time, but agencies won't be able to get away with
> producing rasterized documents, when they could've just given you a
> nice Word document like they started with. Down with scans!
>
>
> * C. For those of you on the agency side: *
>
> You should probably start pressing ASAP for updated FOIA processing &
> redaction software. Otherwise, you're going to be facing much higher
> costs in the long run. Maybe send your IT department a copy of p.
> 18–23 of the opinion. ;-)
>
>
> ** Help wanted! **
>
> I need help on litigating this; taking an appeal from final judgment;
> research; drafting / editing; having a devil's advocate for my prep
> work; PR / public communications; document review; etc.
>
> If you could help with any of that, please contact me privately at
> <legal at s.ai> or other methods listed at <https://s.ai/contact>.
>
>
> I've been doing this for the last 4 years, without a salary, despite
> multiple major disabilities, and my own poverty. Because I'm pro se,
> given the Supreme Court's decision in Kay v Ehrler, it's unlikely that
> I will be able to recover even a cent for my time over the last four
> years of litigating this, despite it being primarily public interest.
>
> So if you appreciate my work, please support my 501(c)(3) nonprofit,
> Fiat Fiendum, so that it can finally pay me a salary to do this sort
> of stuff: <https://patreon.com/fiatfiendum>.
>
> Sincerely,
> Sai
> President, Fiat Fiendum, Inc.