Criminal trials have virtually disappeared in many federal courtrooms. According to a recent U.S. Sentencing Commission report, “[i]n recent years, 97 percent of federal defendants convicted of a felony or Class A misdemeanor offense are adjudicated guilty based on a guilty plea rather than on a verdict at a trial.” But sentencing has seemed to remain visible. In open court, federal judges traditionally explained to victims, the defendant, and the community the sentences they imposed. Now sentencing’s openness is in jeopardy, as federal prosecutors and defense counsel seek to conceal or disguise defendants’ cooperation with prosecutors or law enforcement and how that cooperation decreases their sentences. Why do they do so? Because, they say, defendants known or believed to have cooperated face violence in federal prison.
Cooperation affects a remarkable number of federal sentences. For fiscal year 2017, the U.S. Sentencing Commission reported that federal prosecutors sought cooperation reduced sentences for 7,128 defendants — 10.8 percent of all federal criminal defendants sentenced. In some districts the percentage is much higher — in Maine, for example, it is 28.9 percent; in Pennsylvania Eastern (which includes Philadelphia), it is 34.8 percent. That is a significant part of the federal sentencing process to disguise or conceal from the public, from victims who want to understand the sentence, and from other defense lawyers who seek proportional sentencing for their clients. And those numbers do not include cases in which a judge is asked to reward cooperation that was insufficient to produce a prosecution motion. Indeed, if cooperation can be successfully disguised, the public will be unable to ascertain whether a federal judge’s explanation for any sentence is forthright and complete. Potentially all federal sentences and their rationales will seem veiled.
Federal prosecutors have long used cooperators to help prove crimes. While the cooperation continues, it needs to be secret in order to avoid alerting wrongdoers who are still being investigated but not yet charged. So judges seal from public view cooperation that has not been completed. The need for that type of secrecy generally ends when the cooperation ends, in most cases at the time the cooperator is sentenced.
When mandatory guidelines and mandatory minimum sentences arrived in the late 20th century, federal defendants gained a new incentive to cooperate: Only a prosecution motion based upon a defendant’s “substantial assistance” (a “5K1.1” motion to those in the know) could generate a sentence below the guideline range or below an otherwise mandatory sentencing floor created by statute. The Supreme Court made the guidelines advisory in 2005, but cooperation remains the only way to get a sentence below the statutory floor. Plus, cooperation produces a lower guideline range that operates as a powerful anchor on the sentencing judge.
Sadly, however, 21st-century federal prisons have become dangerous for cooperators — “snitches get stitches,” as the saying goes. Here, chronologically, is some background on the current situation.
In 2001, the United States Judicial Conference (the federal judiciary’s policy-setting body) made the statement-of-reasons portion of a judge’s sentencing document publicly unavailable after the Bureau of Prisons (BOP) reported “an emerging problem of inmates pressuring other inmates for copies of their presentence reports and statements of reasons to learn if they are informants.” In 2002, BOP prohibited prisoners from possessing presentence reports and statements of reasons in their cells, allowing them to consult those documents only in a secure area.
Other paper documents in a criminal case file could reveal cooperation, but they were buried in the courthouse archives and functionally obscure. Then along came electronic case filing to replace paper documents and dockets. At first, the Judicial Conference declined to extend electronic filing to criminal cases (2001). After a pilot project with no apparent ill effects, however, the Conference in 2003–04 approved making criminal case files accessible over the internet. By 2006, according to a notice to federal judges, a new website called www.whosarat.com was outing those who cooperated and revealing identities of undercover agents. The website used many electronic documents downloaded from the courts’ electronic case files system. As a result, Judicial Conference committees recommended “that judges consider sealing documents or hearing transcripts” in cases involving “sensitive information” or yielding “incorrect inferences.” That same year, the Justice Department asked that electronic access to all plea agreements be eliminated, but after seeking public comment, a Conference Committee in 2008 recommended against the change. It did ask each court to adopt a local policy to protect cooperating information while recognizing “the need to preserve legitimate public access to court files.”
Following a major Fordham Law School conference in 2010 on cooperation confidentiality, the Judicial Conference reported to Congress that no single best practice had emerged on how to handle plea agreements revealing cooperation, and therefore no single uniform national rule was proposed.
In 2014, three conference committees jointly asked judges to stop using cooperators’ names in opinions and orders.
In a 2015 high-profile Texas case, United States v. McCraney, federal prosecutors were pressed to justify their request that cooperation documents remain sealed despite a journalist’s request to see them. Prosecutors called as witnesses a supervisory intelligence officer for the Bureau of Prisons, a special investigator at a Federal Correctional Complex, and an assistant United States attorney, possessing a combined 64 years of experience with cooperating defendants. Their testimony was explosive. For several hours these officials described to a federal judge “specific cases in which disclosure of the identity of an informant, and even disclosure that an inmate had relayed any information at all to prison officials, resulted in retaliation in the form of realistic and believable threats of violence, actual violence, and death.”
That same year, the Federal Judicial Center surveyed federal district judges, United States Attorneys’ Offices, federal defenders, court-appointed defense counsel, and chief probation and pretrial services officers about harm to cooperators. The survey, published in 2016, revealed that cooperating defendants “were most likely to be harmed or threatened when in some type of custody,” that “court documents or court proceedings [were frequently] the source for identifying cooperators,” that concerns about harm or its threat “affected the willingness” of defendants and others to cooperate, and that it was “a significant problem.”
Following the survey, the conference’s Committee on Court Administration and Case Management (CACM) determined that further action, including formal rulemaking, was necessary to protect cooperators. But the federal rulemaking process is time-consuming. So while asking the Criminal Rules Committee to adopt rules to address the issue, CACM made an interim recommendation: Create sealed supplements in all cases — whether cooperation occurred or not — so that an outsider could not determine from the docket whether a defendant had cooperated. It said that “the harms to individuals and the administration of criminal justice in this instance are so significant and ubiquitous that immediate and effective action should be taken to halt the malevolent use of court documents in perpetuating these harms . . . .” Some courts adopted that interim proposal, thereby (they hoped) disguising cooperation.
But the proposal was controversial. So the director of the federal courts’ Administrative Office appointed the Task Force on Protecting Cooperators, “charged with taking a broad look at this issue and recommending actions by all entities concerned, including the Bureau of Prisons, the Criminal Division of the Department of Justice, the U.S. Marshals Service, and United States Attorneys.” The task force first met in November 2016, and the director predicted a final report by the fall of 2017, but given the difficulty of the assignment, the final report was not issued until August of 2018. As of this writing, it has not been released publicly.
Meanwhile, in response to increasing requests from prisoners and third parties for court documents — including transcripts — to reveal whether a prisoner has cooperated,federal prosecutors and defense counsel are asking many federal judges to hide cooperation, including how the judge used it in calculating the sentence; to discuss the subject in chambers, not the courtroom, or at least out of public hearing in a whispered conference at sidebar; to partially seal and redact sentencing transcripts; and to disguise the process so that others cannot determine whether a defendant cooperated. Moreover, they have moved to seal any documents that might reveal cooperation — such as cooperation agreements, motions to depart under 5K1.1, and sentencing memoranda — and to keep them sealed for the duration of the sentence. Defendants want their cooperation secret for their safety; prosecutors want it secret so that defendants are not afraid to cooperate. Many judges comply with these requests, and some districts have adopted standing orders to preserve cooperation secrecy. But nationally, and sometimes even within individual districts, it is a patchwork quilt.
In summary, today’s federal sentencing landscape includes courts where the courtroom is physically closed for any cooperation discussion; courts where the courtroom is not closed but any cooperation discussion occurs out of public hearing in chambers or at a private sidebar (some judges hold a pro forma sidebar even where there is no cooperation so that observers cannot infer cooperation from the sidebar); courts where everything is done in open court without sidebars; courts where the lawyers submit cooperation details under seal but the judge announces the sentencing rationale in open court; courts where transcripts of some or all of the above are sealed; courts where virtually nothing is sealed; courts where docket entries are structured so that outsiders cannot determine whether a defendant has cooperated; and other variations. Pity the journalist or citizen who seeks to know with certainty what happened at a particular federal sentencing.
That is where things stand.
In its unpublished interim report some months ago, the Task Force on Protecting Cooperators focused its first recommendations on corrective activity by the Bureau of Prisons, because that is where the problem of violence primarily lies; as a result, the director of the federal courts’ Administrative Office made a number of proposals to BOP in a letter that is not yet public. But a committee report to the Judicial Conference in September 2018 reveals that the recommendations “most significantly would: (a) require that certain case documents sent by courts to requesting inmates be delivered only to the warden, who will provide a secure area in which inmates can review them; (b) expand the list of case documents that will be treated as contraband within BOP facilities; and (c) introduce disciplinary measures for inmates who force other inmates to produce case documents.” In its August 2018 unpublished final report, the task force suggested alterations to electronic case filing procedures that would create new electronic “folders” for documents related to pleas and sentencings. Remote access to those folders would generally be unavailable to laypeople; instead, a courthouse visit would be necessary to see them, hopefully discouraging some malevolent scrutiny and, even then, access would be denied to particular documents a judge had sealed.
This would mark a return to some of the earlier practical obscurity of court files and would be somewhat analogous to the treatment of social security and immigration civil cases under Federal Rule of Civil Procedure 5.2(c). Lawyers, however, would have remote access to all the plea and sentencing folders so that they could see how defendants are treated in similar cases and make proportionality arguments for their own clients. But the task force is not recommending changes to the varying courtroom practices federal judges now employ in hiding or disguising cooperation and its effect on sentencing, and the Federal Rules Committees have rejected the request to alter the Criminal Rules.
Notwithstanding task force recommendations, protecting cooperators will be a difficult challenge for BOP with its limited resources. Making plea and sentencing documents unavailable remotely will make it harder, but not impossible, for prisoners to figure out who cooperated. Should the federal judiciary therefore be enlisted to do more, or should individual judges and districts go farther, as some are doing now, by keeping the fact of cooperation and its effect on federal sentences secret or disguised?
The answer, I say, is, “No.” Transparency is paramount to the federal judicial role. England’s notorious 17th-century Star Chamber discredited the respectability of secret judicial proceedings forever. The Supreme Court says the press and the general public have a First Amendment “constitutional right of access to criminal trials,” and that underlying that First Amendment right is “the common understanding that ‘a major purpose of that Amendment was to protect the free discussion of governmental affairs’” and “to ensure that the individual citizen can effectively participate in and contribute to our republican system of self-government.” With respect to criminal sentencing, Congress has specified that federal judges “at the time of sentencing, shall state in open court the reasons” for imposing a particular sentence. The Supreme Court explains Congress’s open-court sentencing requirement as follows: “Confidence in a judge’s use of reason underlies the public’s trust in the judicial institution. A public statement of those reasons helps provide the public with the assurance that creates that trust.” It “not only assures reviewing courts (and the public) that the sentencing process is a reasoned process but also helps that process evolve” by providing that information to others. The American Law Institute’s recently revised Model Penal Code advances similar reasons for sentencing transparency — among them, the need to communicate the sentencing judge’s reasoning to others in the sentencing system (including other judges), and “to enhance the legitimacy of the sentencing process in the eyes of the offender, the victim, and the public.”
The U.S. Sentencing Commission recognizes that Congress requires “[t]he sentencing judge [to] state the reasons for reducing a sentence” for substantial assistance, but it asserts that a “court may elect to provide its reasons to the defendant in camera and in writing under seal for the safety of the defendant or to avoid disclosure of an ongoing investigation.” It is difficult to square the commission’s broad permission to hide a cooperation-driven reduction with Congress’s open-court language and the powerful arguments for public transparency. But regardless, explaining the sentencing in writing under seal should be the truly exceptional case; for all other cases, closing the courtroom for part of the sentencing rationale should be off the table. So too should a judicial explanation delivered at sidebar or in chambers, or a sealed transcript of what the judge said; these are poorly disguised attempts to avoid Congress’s open-court mandate.
We cannot be confident that sealed sidebars or chambers conferences will even solve the prison retaliation issue. It is impossible to eliminate all cooperation information because there are so many other available sources — for example, cooperation implications drawn from recurrent delays in sentencing a particular defendant; a reduced charge in a conspiracy; general or specific knowledge among co-defendants, their families, and associates; materials about a witness that a statute or the Constitution requires be provided to a defendant; a cooperator’s testimony at a trial or sentencing hearing — the list goes on. The value of judicial transparency should not be sacrificed to a goal whose achievability is so dubious.
If the federal judiciary is indeed to retreat from transparency and pursue secrecy or disguise for cooperation-driven sentences, I say that policy should be created not by individual judges, judicial districts, the Administrative Office, or the Judicial Conference, but through a process that invites public comment and criticism — for example, through congressional legislation that is subject to public debate (and judicial review for constitutionality), or through Rules amendments that are subject to outside input and public hearings (so far the Rules Committees have declined the secrecy proposals).
Some have suggested that it is acceptable to hide the role of cooperation in individual sentences as long as the Sentencing Commission makes aggregate data available. I disagree. Victims are entitled to know why a judge sentences the way he or she does in their particular case. The public (that means all affected communities including, for example, minority groups) is entitled to monitor the appropriateness of the sentence in individual cases — assessing the justice of the punishment and the behavior of the judge and prosecutor. Lawyers for other defendants need to know the judge’s and prosecutor’s sentencing rationales so that they can effectively argue for proportionality for their particular client. Without that information, prosecutors and the judge know how cooperation has affected various sentences — a knowledge that may well affect the next sentence — but defense counsel does not know unless he or she was counsel in the previous sentence(s). That is simply unfair.>
Now I do not mean to suggest that the specifics of cooperation or the existence of ongoing cooperation always need to be disclosed. That information appears in materials the lawyers file or things they say to the judge. If there is a strong countervailing interest, federal appellate cases permit some sealing of such materials for a limited time on a case-by-case basis. (I have granted such motions.) Importantly, however, protecting those specifics is distinct from the demand to disguise the judge’s sentencing rationale explaining that cooperation did in fact affect the sentence. Federal judges should candidly and honestly pronounce their sentencing reasoning in open court, including the fact (if not the extent or details) of cooperation and the effect it has on the sentence length. Sidebars and chambers conferences do not suffice. Cooperation’s role should not be hidden; we must not deceive the public with misleading docket entries or disguised transcript redaction.
Let me be clear: No federal judge wants to be responsible for the death or assault of a sentenced defendant who cooperated. The judge has determined the offender’s punishment, and it does not include violence in prison. But the judge’s role is limited. The judge cannot determine the facility BOP will select for a particular defendant and the resulting risks. The judge cannot disguise the nature of the crime of conviction — for example, a crime like child molesting that might provoke violence against the offender in prison. The judge cannot ensure the adequacy of prison medical care. These and other consequences are all outside the federal judiciary’s role. What the judge can do — must do — is preserve the American public’s trust in the integrity and transparency of the federal judicial system. Americans are entitled to know the role cooperation plays in federal criminal law and sentencing. If the threat of violence deters some defendants from cooperating, then the Justice Department must deal with that consequence in evaluating how it prosecutes cases, or find the resources and the way to help BOP do its job of making prisoners, including cooperating prisoners, safe.
At the end of the day, encouraging or discouraging cooperation is not the business of federal judges. That is the executive branch’s role. Judges constitute an independent branch of government with distinctive responsibilities. Our charge is to sentence convicted defendants fairly, based on all the facts and circumstances and the law, and to explain as clearly as possible to the public, the defendants, and the victims how we reach the sentence we pronounce. As some of us say, a sentencing proceeding is a community morality play where society’s values are publicly applied and affirmed. We should not let the violence of prisoners — even a violence that BOP apparently cannot control — drive federal sentencing underground.
 See Robert J. Conrad, Jr. & Katy L. Clements, The Vanishing Criminal Jury Trial: From Trial Judges to Sentencing Judges, 86 Geo. Wash. L. Rev. 99 (2018).
 Since Booker v. United States, 543 U.S. 220 (2005), the national percentage has ranged between 14.7% and 10.8%. U.S. Sentencing Commission, 2005 Sourcebook of Federal Sentencing Statistics 292-93; U.S. Sentencing Commission, 2017 Sourcebook of Federal Sentencing Statistics S-53 tbl.N. Fed. R. Crim. P. 35(b) also allows for a reduced sentence for substantial assistance after sentencing, upon a government motion filed within one year of the sentence. The numbers and percentages are larger if postsentencing motions for cooperation under Rule 35(b) are included. The Sentencing Commission reported that adding them to 5K1.1s for fiscal 2010 increased the percentage from 11.5% to 13.0%. U.S. Sentencing Commission, The Use of Federal Rule of Criminal Procedure 35(b) 8 (2016).
 At the other extreme, for Washington Western it is 2.6%; for Virginia Eastern, 3.1%. U.S. Sentencing Commission, 2017 Sourcebook of Federal Sentencing Statistics 264, 267.
 The government makes such a motion only where the cooperation results in “substantial assistance in the investigation or prosecution of another person who has committed an offense,” and the decision to make the motion is entirely discretionary with the government. Guideline 5K1.1. In a variant sentence, judges may give credit for cooperation that did not yield other convictions. I have been unable to locate Sentencing Commission data that isolate this factor.
 Alternatively for certain narcotics crimes, defendants who truthfully tell the prosecution all they know about offenses that were part of the same course of conduct or a common scheme or plan and meet certain other criteria can obtain a sentence below the statutory minimum as well as a two-level decrease in total offense level whether or not the information is useful. Guideline 5C1.2 (colloquially known as the safety valve). The number of drug cases where defendants qualify for the safety valve — 6,198 cases nationally, U.S. Sentencing Commission, 2017 Sourcebook of Federal Sentencing Statistics 115 tbl.44) — approaches the number of 5K1.l motions (7,128). To my knowledge, application of 5C1.2 has not been the subject of attention concerning “cooperation.”
 Booker, 543 U.S. at 246.
See Peugh v. United States, 569 U.S. 530, 531 (2013).
 From the Administrative Office of the United States Court on Policy Change Restricting Routine Public Disclosure of the Statement of Reasons and Revised Forms for Judgments in a Criminal Case (AO245B-AO245I), Aug. 13, 2001.
 Bureau of Prisons Program Statement No. 1351.05, Inmate Requests to Institution for Information § 513.40(a)(2)(d)(1), Sept. 19, 2002.
 According to the pilot program, there was no evidence of “harm resulting from remote public access.” David Rauma, Remote Public Access to Electronic Criminal Case Records: A Report on a Pilot Project in Eleven Federal Courts 4 (May 7, 2003).
 Caren Myers Morrison, Privacy, Accountability, & the Cooperating Defendant: Towards a New Role for Internet Access to Court Records, 62 Vand. L. Rev. 921, 957-58 (2009), describes the website’s operation in 2009.
 Mem. from Committee on Court Administration and Case Management and Committee on Criminal Law to District Judges and Magistrate Judges, Nov. 9, 2006.
 David L. Snyder, Nonparty Remote Electronic Access to Plea Agreements in the Second Circuit, 35 Fordham Urb. L. J. 1263, 1276 (2008). The public comment period lasted from early September until late October 2007. Id.
 United States Courts, Guide to Judiciary Policy § 350(b), Vol. 10.
 Judicial Conference of the United States, Second Report on the Adequacy of Privacy Rules Prescribed Under the E-Government Act of 2002 2 (Feb. 2011) (referring to the Apr. 13, 2010 Fordham conference detailed at 79 Fordham L. Rev. 1 (2010)).
 Mem. from Committee on Court Administration and Case Management, Committee on Criminal Law, & the Defender Services Committee to Article III Judges, Sept. 17, 2014 (noting that, “[d]ue to increased access to criminal case filings, there is a growing concern that cooperation information found in these files is being used to threaten or harm cooperating individuals, particularly those who are incarcerated”).
 United States v. McCraney, 99 F. Supp. 3d 651, 655 (E.D. Tex. 2015).
 With at least seven stages of formal comment and review, “it usually takes two to three years for a suggestion to be enacted as a rule.” James C. Duff, The Federal Rules of Practice and Procedure, Administrative Office of the U.S. Courts.
 Mem. from Committee on Court Administration and Case Management of the Judicial Conference of the United States, June 30, 2016 (providing interim guidance to district court judges).
 Mem. from the Administrative Office of the United States Courts Regarding the Formation of the Task Force on Protecting Cooperators, Feb. 10, 2017. The Task Force consists of judges from the Criminal Rules Committee, the Criminal Law Committee, and CACM, along with advisory participants from the Department of Justice, the Bureau of Prisons, the Federal Public Defenders, and a Clerk of Court.
 Advisory Committee on Criminal Rules, April 17, 2017 Minutes 4 (“The BOP working group report describes widespread attempts by inmates to determine if someone newly designated to a particular facility has been a cooperator. In many places a newly arriving inmate is asked for ‘his papers’ (whatever documents the inmate has, such as a PSR, sentencing minutes, judgment and commitment order, transcripts, etc.). If the inmate says he doesn’t have his papers, he is told to get them. As a result, inmates ask people outside the prison, often their relatives, to get their papers. There have also been an increasing number of requests by inmates asking the district courts to send their papers to them in prison.”). Accord United States v. McCraney, 99 F. Supp. 3d 651, 656 (E.D. Tex. 2015) (citing BOP Supervisory Intelligence Officer testimony that “inmates often write to friends and family members who are not incarcerated and ask them to research other inmates on LexisNexis, Google, or Pacer to determine whether they provided substantial assistance to the Government.”).
 I am surprised that news media interest in this issue has been so mild.
 Advisory Committee on Criminal Rules, April 24, 2018 Draft Minutes 273.
 Report of the Judicial Conference Committee on Court Administration & Case Management 14 (Sept. 2018).
 James C. Duff, Items of Interest: Director’s Report to the Judicial Conference of the United States 20, Sept. 13, 2018 (“[T]he Task Force recommended changes that included . . . adoption of a ‘Plea and Sentencing Folder’ approach to handling cooperation information in files and docket entries in the Case Management/Electronic Case Files (CM/ECF) and the Public Access to Court Records (PACER) systems.”).
 See note 27, supra, at 15.
 Secrecy is the Star Chamber’s abiding reputation, although historians point out that its secrecy was in fact limited. Daniel L. Vande Zande, Coercive Power and the Demise of the Star Chamber, 50 Am. J. Legal Hist. 326 (2010).
 Globe Newspaper Co. v. Superior Court, 457 U.S.596, 603-04 (1982).
 18 U.S.C. § 3553(c) (emphasis added)
 Rita v. United States, 551 U.S. 338, 356 (2007); Chavez-Meza v. United States, 138 S. Ct. 1959, 1964 (2018).
 Rita, 551 U.S. at 357–58. In the criminal trial context, Chief Justice Burger quoted Jeremy Bentham: “Without publicity, all other checks are insufficient; in comparison of publicity, all other checks are of small account.” Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 569 (1980) (quoting Jeremy Bentham, Rationale of Judicial Evidence 524 (1827)).
 Model Penal Code § 10.01(4) cmt. e (Am. Law Inst. 1962). As the Second Circuit said in United States v. Amodeo, 71 F.3d 1044, 1048 (2d Cir. 1995): “Without monitoring . . . the public could have no confidence in the conscientiousness, reasonableness, or honesty of judicial proceedings. Such monitoring is not possible without access to testimony and documents that are used in the performance of Article III functions.” (Amodeo was not a sentencing case.)
 United States Sentencing Commission, Guidelines Manual § 5K1.1, comment. (backg’d) (Nov. 2018).
 Ironically, the commission itself says the “advisory guideline system continues to assure transparency by requiring that sentences be based on articulated reasons stated in open court that are subject to appellate review.” USSG Ch. 1, Pt. A, intro. comment.
 Interpreting § 3553(c), the Second Circuit has found it improper to sentence in the judge’s robing room. United States v. Alcantara, 396 F.3d 189, 206 (2d Cir. 2005) (“It is clear that ‘open court’ as used in these rules refers to a courtroom to which the public has access.”). See also United States v. Molina, 356 F. 3d 269, 277 (2d Cir. 2004) (one of Congress’s goals in § 3553(c) was “to enable the public to learn why [the] defendant received a particular sentence”). Justice Department policy is also against courtroom closure, requiring 60-day recurrent review of sealing. 28 CFR § 50.9. See also S. Attorneys’ Manual 9-5.150.
 For many courtrooms these days, there are neither journalists nor public court-watchers in attendance; thus sealing the transcript effectively forecloses public access.
 18 U.S.C. § 3500 (The Jencks Act)
 Giglio v. United States, 405 U.S. 150, 154 (1972) (nondisclosure of a leniency promise to a witness violates due process).
 It is reported that prisoners quickly become adept at figuring out what various docketing devices, such as skipped numbers or missing documents, are designed to hide or disguise.
 Supreme Court and circuit court cases recognize a presumptive common law right of public access to judicial documents and a qualified First Amendment right of access to criminal proceedings. See, e.g., Press-Enterprise Co. v. Superior Court, 478 U.S. 1 (1986) (transcript of preliminary hearing); Press-Enterprise Co. v. Superior Court, 464 U.S. 501 (1984) (state court criminal jury selection proceedings); United States v. Kravetz, 706 F.3d 47, 57, 59 (1st Cir. 2013) (sentencing memoranda and letters to a judge). But where the lawyers are in agreement and media are not interested, there is no challenge to courtroom closure or seal.
 See supra note 30.
 g., Morrison, supra note 12, at 926.
 The task force seems to recognize the importance of lawyer access to judges’ sentencing reasoning in its proposal that lawyers have remote ECF access to all plea and sentencing folders so that they can make proportionality arguments for their respective clients. But judges who redact and seal their cooperation-affected sentencing reasoning will thwart that ability.
 See, e.g., Kravetz, supra note 45; United States v. Doe, 870 F.3d 991, 1001–02 (9th Cir. 2017). Arguably, Doe suggests a weakening of the openness requirement resulting from the FJC survey and the conference committee recommendation. Id. at 998–1000.
 It is not my purpose here to challenge the use of cooperation in criminal prosecutions. But it is not an unmitigated blessing. See, e.g., Alexandra Natapoff, Criminal Informants and the Erosion of American Justice (2009) (especially chapter 4); Miriam Heckler Baer, Cooperation’s Cost, 88 Wash. U. L. Rev. 903 (2011). The public needs to be informed of cooperation’s role in federal criminal prosecution not just in the abstract but in the specifics.