Illustration for FOIA 26-32: The office answered its own denial
FOIA 26-32 · Granted in part, denied in part — five records produced across 95 pages, with only personal contact information redacted under MCL 15.243(1)(a); no fee quoted

On February 6, 2026, a requester asked the Delta County Prosecutor's Office a single, narrow question: what records discuss how the office's case-management migration affects its ability to answer FOIA requests? The office received it on Monday, February 9, and answered on Thursday, February 12 — the third business day, well inside the five-day window. It marked the request GRANTED IN PART, DENIED IN PART, named five responsive records, and attached a 95-page release. The release is, almost entirely, the office's own memos to the Board of County Commissioners — documents in which the office describes, at length and in its own words, exactly how the migration affects its ability to answer FOIA requests.

That is the shape of this file. A request about the migration was partly denied on a privacy ground and then answered, in full, by the records the office chose to produce. The grant is the answer to the denial. What the memos say is more candid than anything the office has said in a response letter, because they were written for a different audience — the Board — and were never meant to be read as a FOIA answer. They are now.

How to read this page

Documented facts come from the released records and are dated and quoted. The agency's position is what the office asserted or did. Legal interpretation is analysis of Michigan's Freedom of Information Act, and is labeled as such. Open questions are issues the record does not resolve. Nothing on this page asserts that the partial denial was unlawful, that the office was obligated to create a record it does not maintain, or that anyone acted in bad faith. Where the record supports only a question, only a question is stated.

Summary

The request is captioned “Impact of Migration on FOIA Responses,” and its operative sentence is short: it asks for “any records discussing how the system migration affects the Prosecutor's Office's ability to respond to FOIA requests.” It adds the same courtesy sentence that appears across this archive — “If no such calculation is maintained, please indicate that no responsive record exists. This request does not require creation of a new record” — and it asks for a public-interest fee waiver under MCL 15.234.

The office's answer does two things at once. It grants the request as to five records: two email exchanges and three memoranda, all concerning appeals of earlier FOIA requests (25-278, 25-94, and 25-195). And it denies the request in part, invoking MCL 15.243(1)(a), the exemption for information comingled with material that would be a clearly unwarranted invasion of privacy. The Notice of Denial describes the redacted material precisely: “contact information to include address phone numbers email addresses.” That is a narrow, honest redaction basis, and it is the best-justified denial in this archive. The office did not withhold the substance of anything. It withheld names, addresses, phone numbers, and email addresses, and it produced the rest.

What the rest contains is the story. The three memos were written to the Board in June 2025, September 2025, and December 2025, to defend the office's handling of earlier FOIA appeals. Each of them discusses the migration. Together they establish, in the office's own words, that the move to ProsecutorByKarpel did not carry the old data across cleanly, that the office knew before implementation that it might not, that the old system's data is now gone, that the file structure makes year-based searching impossible by design, and that a county commissioner proposed a cheaper technical fix that the prosecutor declined because “we don't have time.”

Four numbers organize this file. 95 is the number of pages released. 5 is the number of responsive records the office named. 3 is the number of distinct per-file time estimates the release contains — 20 minutes, 14 minutes, and 5 minutes — all describing the same kind of review. And $4.13 is the fractional-increment overcharge that appears twice in the 25-278 cost itemization reproduced inside this release, the same $4.13 that appears in 26-28 and 26-29.

What was requested

The request is dated February 6, 2026, and was transmitted by fax; the Zimbra/IPFone notification in the packet records receipt at 10:22 a.m. Eastern on February 6, and the office's Notice of Denial records a received date of 02/09/2026 — the following Monday. The request asks for one category of records and nothing else:

“Any records discussing how the system migration affects the Prosecutor's Office's ability to respond to FOIA requests.”

Three features of the request matter to what followed. First, it asks for records that discuss the migration, not for the migration itself, and not for a new analysis. It is a request for documents the office already wrote. Second, it carries the no-new-record sentence, so the office had a clean path to say “none exist” if that were true. Third, it asked for a fee waiver under MCL 15.234 — and, unlike almost every other file in this archive, the office did not quote a fee at all. The response is a production, not an invoice.

One small thing about the request, noted for completeness rather than fault. The form's response-time paragraph asks the office to respond “within five (5) business days as required by MCL 15.235(2).” The five-business-day period is in subsection (1); subsection (2) is the extension provision. The citation is the requester's, not the office's, and it had no effect on a response that was timely under either reading.

What the office answered

The February 12 letter states the disposition in a single line — “Your request is GRANTED IN PART, DENIED IN PART” — and then names the five responsive records it located:

  • “Email exchange entitled Memo regarding Appeal of FOIA 25-278”
  • “Memo regarding Appeal of FOIA 25-278”
  • “Memo regarding Appeal of FOIA 25-94”
  • “Memo regarding Appeal of FOIA 25-195”
  • “Email exchange entitled Fwd: Appeal”

The denial basis is stated once, and it is narrow. The letter invokes MCL 15.243(1)(a) — the exemption for “exempt information comingled with non-exempt information” — and says the records “are subject to redaction.” The Notice of Denial form, under reason 3 (Redaction), quotes the same subsection and describes what was redacted as “contact information to include address phone numbers email addresses.” That is the whole of the withholding. There is no claim of privilege, no deliberative-process claim, no law-enforcement claim. The office produced the records and blacked out the contact details.

That is worth stating plainly, because it is unusual in this archive. In file after file, the office has denied categories of records outright or priced them out of reach. Here it did neither. It found five records, produced them, and redacted only personal contact information — which is the one category of redaction that is almost always defensible. On the narrow question of whether this response was properly handled, the answer is largely yes. The trouble, as it usually is in this archive, is in what the produced records say.

The 95-page release

The office named five records, but the release runs to 95 pages, because the three memos carry their attachments: the earlier response packets and appeals they were written to defend. The figure below breaks the release down by what the pages actually are.

Figure 1 · Interactive

Ninety-five pages, and what they are

Select a bar to read what it counts. Most of the release is the office's own memos to the Board and the earlier packets they defend.

Pages released95
Records named5
Memos to the Board3
Per-file time estimates3

Source: page-by-page review of the 95-page responsive-documents release for FOIA 26-32.

Figure 1. The release is mostly the office's own advocacy: three memos to the Board, and the earlier response packets and appeals they were written to defend. The pages that answer the request are the 26 pages of memos.

Documented: The office produced 95 pages across five named records. Twenty-six of those pages are the three memos to the Board, which are the documents that actually discuss the migration. The remaining 69 pages are the earlier packets, appeals, request forms, and boilerplate that the memos attach or reproduce. The release is not padded in any improper sense — a memo that defends an appeal naturally attaches the appeal — but a reader looking for what the office says about the migration should go straight to the memos, and to the two email threads.

What the memos say about the migration

The single most important document in the release is the memo on the appeal of FOIA 25-94, dated June 16, 2025. It is the office's fullest account of the migration anywhere in the public record, and it was written to persuade the Board that the office's handling of an earlier request was reasonable. It reads, in part:

“In 2018, the Delta County Prosecutor's Office began an internal transition to ‘paperless’ data storage. … This system was paired with the case tracking system that had been created and maintained by the Prosecuting Attorneys Association of Michigan (‘PAAM’). … In 2023, PAAM started exploring a new case tracking system for all Michigan Prosecutor's Offices. After much considerations, debate, and input, ProsecutorbyKarpel (‘Karpel’) was chosen. The cost of this new program was borne by PAAM through funding earmarked by the Michigan legislature. … On June 10, 2024, the Delta County Prosecutor's Office ‘went live’ with Karpel.”

So far this is a straightforward account of a software change. What follows is not:

“Through this transition the old data from the previous case tracking system was transitioned to Karpel. However, over time, it has become clear that, despite best intentions, the data did not transition completely, or accurately due to different inputs. This inconsistent and inaccurate transition was a possibility that we were warned of prior to the implementation. In anticipation of this, Prosecutor Wickman ran certain data reports to preserve that information. However, access to the previous case tracking system or its exact data is no longer available.”

Three admissions sit in that paragraph, and each is the office's own. First, the migration did not carry the data across “completely, or accurately.” Second, the office was “warned of this prior to the implementation” — it knew the risk before it took it. Third, and most consequential for anyone trying to use these records, “access to the previous case tracking system or its exact data is no longer available.” The old system is gone. Whatever it held that did not make the crossing is now unrecoverable through the office.

The memo then explains why the office cannot simply search the records it does have by year:

“These files are organized by an individual's name, with each file associated with that person within its own file in the ‘name’ file. … Because the root files are not organized by year (only by name), searching for outcomes by any particular year is not viable.”

That sentence is the technical heart of several earlier denials in this archive. When the office has said it cannot produce outcome data “by year,” this is why: the files are keyed to names, not dates, and the old system that might have indexed them by date is gone. The memo is not hiding that fact. It is stating it, in a document written to justify a denial, and the office has now produced the document.

The memo on the appeal of 25-278, dated December 31, 2025, adds a fourth admission, and it is the one that matters most to the question of ongoing reliability:

“The issue is not with the data being migrated, but with the information from one system not matching with all data fields or areas in the new system. … There are still mistakes and issues being found with the migrated information on nearly a weekly basis due to the age of the case at issue and what information was originally inputted into the old system.”

That is a December 2025 statement that the migrated data is still producing errors “on nearly a weekly basis” — eighteen months after the June 2024 go-live. It is not a description of a completed migration with a few loose ends. It is a description of a system still being repaired in real time.

The commissioner's cheaper fix, and the answer

The release also contains the email thread that the request's first named record refers to — the January 2026 exchange among Prosecutor Wickman, the Board of Commissioners, and the FOIA coordinator about a corrected copy of the 25-278 memo. Most of it is administrative. One exchange is not.

On Saturday, January 3, 2026, at 11:21 a.m., Commissioner Christine Williams wrote to the prosecutor about a feature of ProsecutorByKarpel:

“One of the features of PbK is the ingestion of court form data, however, it is dependent upon how the form is structured. It would be helpful for me to have an example. This feature may cost additional money but might be much less than what was quoted for manual preparation.”

That is a commissioner proposing a technical route — letting the case-management system ingest structured court-form data — that might cost less than the manual preparation the office had been quoting to requesters. On Sunday, January 4, 2026, at 10:00 a.m., the prosecutor replied. The reply is the most revealing paragraph in the release:

“It depends on the Court. The Circuit Court form is a computer generated form, but it scanned into the system. The District Court form is a handwritten form, still scanned into the system. Basic information is manually entered into the new system (i.e. the ultimate sentence, or if a plea is entered, but the terms of the plea agreement or if it is a plea based on a plea offer vs a plea as charged, is not entered. In essence the fact a plea is entered, but provides no context of how that plea came about). Minimal information was entered into the old system and it inconsistently transferred from the old system to the new system due to the two systems not speaking to one another. We have not reentered this information because we don't have time and can be obtained from Court records if absolutely necessary (it is rarely needed).”

That paragraph does several things at once. It confirms that the plea-bargaining detail at the center of earlier requests — whether a plea was based on an offer, what the terms were, how it came about — is not entered in the new system, and was only “minimal” in the old one. It confirms that the transfer was inconsistent “due to the two systems not speaking to one another.” And it states the reason the information has not been restored: “we don't have time.” The commissioner had just suggested a technical route that might cost less than manual preparation; the answer was that the office has not reentered the information because it does not have time, and that the information “can be obtained from Court records if absolutely necessary.”

That is a defensible position for a prosecutor's office managing a caseload. It is also, read as a FOIA answer, an explanation of why the office's own records are incomplete: the detail exists elsewhere (in court records), the office knows it is missing from its system, and it has chosen not to restore it because restoring it would take time the office does not have. The public can now read that choice in the office's own words.

Three estimates for the same work

The release is unusual in that it contains, in a single production, three different estimates of how long it takes to review one file for a FOIA request. All three describe essentially the same task — locating a case file and separating exempt from non-exempt material — and all three are the office's own figures.

How long the office says it takes to review one file
Source documentDatePer-file estimateWhat it was used to price
Memo on the appeal of 25-94June 16, 202520 minutes per file, “conservatively”A review of roughly 11,000 files across the requested period
Response to 25-278December 202510 minutes to locate + 4 minutes to separate = 14 minutes per file4,000 files across four years
Responses to 26-28, 26-29, 26-30February 12, 20263 minutes to locate + 2 minutes to separate = 5 minutes per file800, 500, and other baselines in the statistical cluster

The trend is downward, and steeply. The same office, describing the same kind of file review, estimated 20 minutes per file in June 2025, 14 minutes in December 2025, and 5 minutes in February 2026. Each estimate was attached to a fee. The higher the per-file figure, the higher the fee; the lower the figure, the lower the fee. That is not necessarily improper — an office can get faster at a task, or can scope a task more narrowly — but the three figures cannot all describe the same work at the same level of care, and the release does not reconcile them. A requester who was quoted 20 minutes a file in one packet and 5 minutes a file in another, months apart, is entitled to ask which estimate reflects the actual work.

The 25-94 memo's arithmetic is the most elaborate of the three, and it is worth recording because it shows the office reasoning about its own capacity. From roughly 11,000 files, at 20 minutes each, the memo concludes that “an individual could review three files an hour,” that at $18.31 an hour this yields “approximately 105 files per week,” and that the full review would take “104 weeks, or 1.87 YEARS,” at a cost of “$66,648.40.” That is a striking number, and it is the office's own. It is also the number the office has since revised downward by a factor of four in its later estimates.

The $4.13, twice

The 25-278 cost itemization reproduced inside this release is the same document published with the 25-278 record, and it carries the same defect that appears in 26-28 and 26-29 — but here it appears on two lines, not one.

What the 25-278 itemization charges, and what its own rounding rule produces
LineWork describedMinutesIncrements printedAmount printedRounded down
1Labor to locate (10 min × 4,000 files)40,0002,666.66$16,533.332,666 → $16,529.20
3aSeparating exempt from non-exempt (4 min × 4,000 files)16,0001,066.66$6,613.331,066 → $6,609.20
—Subtotal——$23,146.66$23,138.40

The itemization's own instruction states that “all partial time increments must be rounded down.” Forty thousand minutes divided by fifteen is 2,666.67 increments, which rounds down to 2,666, and 2,666 × $6.20 is $16,529.20, not the $16,533.33 printed. Sixteen thousand minutes divided by fifteen is 1,066.67 increments, which rounds down to 1,066, and 1,066 × $6.20 is $6,609.20, not the $6,613.33 printed. Each line is overcharged by $4.13, and the subtotal is overcharged by $8.26. The appeal that follows this itemization cites the printed subtotal, $23,146.66, and a deposit of $11,573.33 — both of which inherit the error.

This is the third file in this archive in which the same $4.13 appears. In 26-29 it was a single line; in 26-28 it was a single line; here it is two. The amount is trivial. The pattern is not. Whenever the minute total divides into a fraction ending in .6667, the office bills the fraction instead of rounding it down, and the overcharge is always 0.6667 × $6.20, or $4.13. It is a reproducible defect in how the office completes its own form, and it recurs on every itemization where the minutes are not divisible by fifteen.

The timeline

FOIA 26-32, from fax to answer
DateEventSource
February 6, 2026, 10:22 a.m. EasternRequest faxed; IPFone notification records the transmissionZimbra/IPFone notification, page 3
February 9, 2026 (Monday)Office records receiptNotice of Denial, page 6
February 12, 2026 (Thursday)Response letter, Notice of Denial, and 95-page release issuedPages 1–10 of the response; the responsive-documents release

The response is timely. Under MCL 15.235(1), a public body must respond within five business days of receipt. Counting from Monday, February 9, the fifth business day is Monday, February 16; the office answered on Thursday, February 12, the third business day. No extension was taken and none was needed. This page does not suggest any timing violation, because on this record there is none.

What the record shows

The grant answers the denial. The request asked for records discussing how the migration affects the office's ability to answer FOIA requests. The office denied part of the request on a privacy ground and then produced three memos that discuss, at length, exactly how the migration affects its ability to answer FOIA requests. Whatever the partial denial was meant to withhold, it did not withhold the answer.

The migration was known to be risky before it happened. The office's own memo says the inconsistent transition “was a possibility that we were warned of prior to the implementation.” That is not a hindsight judgment from this site. It is the office describing what it knew when it chose to proceed.

The old data is gone. “Access to the previous case tracking system or its exact data is no longer available.” Whatever did not survive the crossing cannot now be recovered from the office, and the office says so.

The file structure defeats year-based search by design. “Because the root files are not organized by year (only by name), searching for outcomes by any particular year is not viable.” That single sentence explains a pattern of denials across this archive better than any response letter does.

A commissioner offered a cheaper technical route; the office said it did not have time. The January 2026 exchange shows a commissioner proposing PbK's court-form ingestion feature as possibly “much less than what was quoted for manual preparation,” and the prosecutor replying that the office has “not reentered this information because we don't have time.”

The per-file estimate has fallen from 20 minutes to 5. Three figures, one production, all the office's own, each attached to a fee. The release does not reconcile them.

The $4.13 is back, twice. The 25-278 itemization reproduced here overcharges two lines by $4.13 each, on the same rounding defect documented in 26-28 and 26-29.

The office answered its own denial

There is a version of this file that is a straightforward win for transparency, and it is worth stating before the criticism. A requester asked a narrow question about a system migration. The office found five records, produced them, redacted only personal contact information, quoted no fee, and answered within three business days. By the standards this archive has documented — denials of categories, fees in the tens of thousands, blank waiver lines, per-file estimates that change with the invoice — that is a good response. It is the kind of response a requester hopes for. The redaction basis is narrow and honest, the timing is clean, and the production is substantial. On process, this file is close to a model.

And yet the substance of what was produced is more damaging to the office than any denial could have been, because it is the office speaking in its own voice, to its own Board, about its own system. The memos were not written to answer this request. They were written to defend earlier appeals. That is precisely why they are candid. In them the office says that the migration did not carry the data across completely or accurately, that it was warned this could happen before it proceeded, that the old system and its exact data are now gone, that the files are organized by name rather than year so year-based search is impossible, that mistakes in the migrated data are still being found on nearly a weekly basis eighteen months later, and that a commissioner's suggestion of a cheaper technical fix was met with the answer that the office does not have time to reenter the information.

None of that is a scandal in the sense of a concealed wrongdoing. It is something more ordinary and, for a public records office, more serious: a candid account of a system that does not reliably hold the records the public is entitled to see, written by the office that operates it, and produced only because a requester asked for records that discuss the migration. The office did not volunteer any of this. It answered a narrow question, and the answer happened to be the truth about its own records.

The deepest problem is the one the commissioner's exchange exposes. The office's case-management system has a feature that could ingest structured court-form data, which might make the plea-bargaining detail at the center of earlier requests searchable and producible. A commissioner raised it. The prosecutor's answer was that the office has not reentered the information because it does not have time, and that the information can be obtained from court records “if absolutely necessary.” That is a reasonable triage decision for a busy prosecutor. But it is also a decision that the office's own records will remain incomplete, that the detail will live only in the courts, and that a member of the public seeking it from the prosecutor's office will be told, in effect, to go elsewhere. The office is entitled to make that choice. The public is entitled to know it made it. This file is how the public finds out.

There is a final irony, and it is the one worth ending on. The request was captioned “Impact of Migration on FOIA Responses.” The office denied part of it and granted the rest. But the granted records are the impact of the migration on FOIA responses, described in detail, by the office, in documents it wrote to defend its FOIA responses. The migration's impact on FOIA, in other words, is that the office now answers FOIA requests about a system that lost data it was warned it might lose, cannot search by year, and is still producing errors weekly — and the proof of that is the very packet the office sent to answer the question. A public body can produce the records and still reveal, in producing them, that the records are the problem. This file is both at once.

What this record does not show

This record does not show that the partial denial was unlawful. The office redacted contact information under MCL 15.243(1)(a), described the redacted material precisely, and produced everything else. That is a narrow and defensible application of the privacy exemption.

It does not show that the migration was mismanaged in a way that harmed any particular case. The memos describe data that did not transfer and errors still being found, but they do not identify a case in which the loss of data changed an outcome, and this page does not infer one.

It does not show that the three per-file estimates are inconsistent in a way that inflated any specific fee. The estimates are the office's own and they trend downward; whether each was appropriate to its task is a judgment the record does not settle.

It does not show that the office was obligated to adopt the commissioner's suggested technical fix, or that doing so would have been cheaper in fact. The commissioner said the feature “may cost additional money but might be much less”; that is a possibility, not a finding.

It does not show what the old case-tracking system contained that did not survive the migration. The office says access to it is gone, and nothing in the packet reconstructs what was lost.

What to ask next

The record suggests three follow-up requests, each narrow enough to be answered without a fee dispute.

The first would ask for the data reports Prosecutor Wickman ran before implementation to preserve information, which the 25-94 memo says she created “in anticipation” of an incomplete transition. Those reports are the office's own snapshot of what the old system held, and they may be the only surviving record of the data that did not cross. A request for them is a request for documents the office says it made.

The second would ask for any written assessment, memo, or communication in which the office was “warned” before implementation that the transition might be inconsistent or inaccurate — the warning the 25-94 memo refers to. Who warned the office, when, and in what document is a question the record raises but does not answer.

The third would ask whether the office has evaluated the ProsecutorByKarpel court-form ingestion feature that Commissioner Williams raised, and for any cost estimate or analysis of it. If the feature was assessed and rejected, the assessment is a record. If it was never assessed, the public learns that the cheaper route the commissioner suggested was not pursued.

All three are catalogued in our internal working list, with the specific wording and the record that supports each one. They are not published here.

Authorities and sources cited on this page

Show the statutes, cases, and sources cited on this page

Michigan Freedom of Information Act, MCL 15.231 et seq. The following provisions are cited above: MCL 15.233 (public body's response obligations); MCL 15.234 (fees, itemization, deposits, and the public-interest waiver); MCL 15.235 (response deadlines, including the five-business-day period in subsection (1) and the extension in subsection (2)); and MCL 15.243(1)(a) (the exemption for information comingled with material whose disclosure would be a clearly unwarranted invasion of privacy, the basis the office invoked for its redactions).

Documents in the packet. The response is ten pages: pages 1–2 the February 12 response letter; pages 3–4 the Zimbra/IPFone fax notification and cover sheet; page 5 the request as received; page 6 the Notice of Denial; pages 7–10 the statutory excerpts, a blank appeal form, and a FOIA Written Policies notice. The responsive-documents release is 95 pages, broken down in Figure 1 above.

Public records relied on for quotations. The June 10, 2024 go-live date, the “did not transition completely, or accurately” admission, the “we were warned of this prior to the implementation” sentence, the “no longer available” sentence, the “organized by name” sentence, and the 20-minute/104-week/$66,648.40 arithmetic are taken from the memo on the appeal of 25-94 (release pages 66–81). The “mistakes and issues … on nearly a weekly basis” sentence and the 10-minute/4-minute per-file estimates are taken from the Response to FOIA 25-278 (release pages 21–37). The January 3–4, 2026 email exchange and the “we don't have time” sentence are taken from the email thread (release pages 1–8). The 3-minute/2-minute estimates are taken from the 26-28 and 26-29 itemizations published elsewhere in this archive. The $16,529.20 and $6,609.20 figures are computed on this page from the itemization's own rate and its own rounding rule, and are labeled as computations, not as statements by the office.

On the redactions. The requester's name, home address, telephone number, email address, and originating fax station identifiers have been removed from every document published with this article, including the request as faxed, the response letter, the Notice of Denial form, the fax notifications, and the reproduced 25-278 request and appeal. No substantive content was removed. The office's own redactions under MCL 15.243(1)(a) — contact information only — are preserved as the office made them.

On the per-file estimates. The three figures — 20, 14, and 5 minutes — were each verified against the page on which they appear in the release. The 20-minute figure appears in the 25-94 memo; the 14-minute figure is the sum of the 10-minute locate and 4-minute separate estimates in the 25-278 response; the 5-minute figure is the sum of the 3-minute and 2-minute estimates in the February 2026 statistical responses. They describe the same category of work at different dates and are reported here as the office stated them, without adjustment.

Documents

These are redacted copies prepared for publication. Requester names, home addresses, personal phone numbers, and email addresses have been removed. Unredacted originals are not hosted on the public web server.

Response Letter (redacted) Two pages: the February 12, 2026 cover letter stating GRANTED IN PART, DENIED IN PART, naming the five responsive records, invoking MCL 15.243(1)(a) for the redactions, and listing the statutory authorities and the FOIA Written Policies notice. Request and Fax Record (redacted) Three pages: the Zimbra/IPFone fax notification, the fax cover sheet recording receipt at 10:22 a.m. Eastern on February 6, 2026, and the request as received — the one-page ‘Impact of Migration on FOIA Responses’ request with the fee-waiver request and the no-new-record sentence. Notice of Denial and Authorities (redacted) Five pages: the Notice of Denial for request 26-32 (received 02/09/2026, notice 02/12/2026, reason 3 redaction under MCL 15.243(1)(a)), two reproductions of the statutory excerpt, a blank FOIA Appeal Form — To Appeal a Denial of Records, and a FOIA Written Policies notice. Email Thread, January 2026 (redacted) Eight pages: the January 2–4, 2026 Zimbra exchange among Prosecutor Wickman, the Board of Commissioners, and the FOIA coordinator about a corrected copy of the 25-278 memo, including Commissioner Williams’s suggestion of the ProsecutorByKarpel court-form ingestion feature and the prosecutor’s reply that the office has ‘not reentered this information because we don’t have time.’ Memo on the Appeal of 25-278, with Attachments (redacted) Thirty-six pages: the December 31, 2025 memo to the Board on the appeal of 25-278, its Attachment A (the 25-278 request and fax record), its Attachment B (the Response to 25-278, including the page admitting mistakes in the migrated data are still found ‘on nearly a weekly basis’), and the Detailed Cost Itemization for 25-278. Appeal of 25-278 as Filed (redacted) Seven pages: the appeal filed December 19, 2025, citing the $23,146.66 fee and $11,573.33 deposit, and arguing the ongoing system-migration problems at length. Memo on the Appeal of 25-94 (redacted) Sixteen pages: the June 16, 2025 memo to the Board on the appeal of 25-94 — the office’s fullest account of the migration, including the June 10, 2024 go-live date, the admission that the data ‘did not transition completely, or accurately,’ that the office ‘was warned of this prior to the implementation,’ that access to the old system ‘is no longer available,’ that the files are organized by name so year-based search is ‘not viable,’ and the 20-minute/104-week/$66,648.40 arithmetic. Memo on the Appeal of 25-195 (redacted) Six pages: the September 15, 2025 memo to the Board on the appeal of 25-195, describing the case-management conversion and the itemization of labor and redaction costs in 15-minute increments rounded down.