OPINION
SMITH, Circuit Judge.I.
We issue this opinion in support of our order filed on January 9, 2008 in the mat*224ter of United States v. Wecht. WPXI, Inc., PG Publishing Company, doing business as Pittsburgh Posh-Gazette, and Tribune-Review Publishing Co. (collectively, the “Media-Intervenors”), filed a motion challenging an order of the United States District Court for the Western District of Pennsylvania announcing jury selection procedures to be used at an impending criminal trial. Specifically, the Media-In-tervenors challenged the District Court’s decisions (1) to empanel an anonymous trial jury, and (2) to conduct voir dire through use of a written questionnaire and without venirepersons physically present in an open courtroom until the pool of prospective jurors was reduced to 40. In our January 9 order, we vacated the District Court’s order to the extent that it restricts public access to the names of trial jurors or prospective jurors.1 We denied all other relief sought.
II.
On January 20, 2006, a grand jury returned an 84-count indictment against Dr. Cyril H. Wecht. As we noted in deciding an earlier interlocutory appeal in this matter, “[t]he 84-count indictment asserts that [Wecht] unlawfully used his public office as the coroner of Allegheny County, Pennsylvania, for private financial gain.” United States v. Wecht, 484 F.3d 194, 198 (3d Cir.2007). The offenses charged included theft of honest services, mail and wire fraud, and theft from an organization receiving federal funds. The case was assigned to Judge Arthur Schwab of the U.S. District Court for the Western District of Pennsylvania. During a pretrial conference on July 12, 2006, the parties reported to the District Court that they had agreed to use a 24-page questionnaire containing 69 questions in the jury selection process.2
The following day, on July 13, 2006, the Board of Judges for the Western District of Pennsylvania entered an administrative order directing that “all jurors shall be identified in court during the jury selection process by his/her assigned juror number ONLY. A prospective juror shall no longer be identified by or identify himself or herself by name.” In re Jury Administration Procedures, Misc. 06-211 (W.D.Pa. July 13, 2006). The order further provided that “any and all juror lists generated by this Court for use in the jury selection process shall be deemed confidential and property of the Court and shall not be removed from the Court at any time.” Id. The juror lists were available only to counsel who were required to execute a receipt for the list and to return it upon completion of jury selection.
*225A day later, on July 14, Judge Schwab issued a pretrial order addressing the use of the jury questionnaire and establishing procedures to be employed during voir dire. Section A of the order noted that the summons to be issued to prospective jurors would be mailed together with the final juror questionnaire, a cover letter from the Court, and instructions. Section B of the order pertained to the jury questionnaire procedure. Paragraph 5 of Section B of the order stated: “Pursuant to the decision of the Board of Judges of this District, counsel shall not have access to the names and addresses of the prospective jurors. Therefore, Jury Administrator Morder is instructed to remove and retain the last page of the Jury Questionnaire setting forth the prospective jurors’ names and current addresses.” Although this directive was more restrictive than the July 13, 2006 standing order, neither Wecht nor the Government objected. Paragraph 6 of Section B of the order scheduled a hearing for September 19, 2006, to permit counsel to review the completed jury questionnaires, albeit without the names and addresses of the jurors, and to confer among themselves as to prospective jurors that they did not believe should be part of the venire. Jury selection was scheduled for October 11, 2006.
In mid-September, after hearing oral arguments in the first Wecht appeal, this Court granted a stay of the trial pending its resolution of the various appeals. United States v. Wecht, No. 06-3098 (3d Cir. Sept. 15, 2006). As a result, the 300 venire members who had been summoned were released. After we issued our mandate, Judge Schwab entered an order on November 26, 2007, announcing the jury selection procedures he would follow. The order stated in the first paragraph that the “jury will be anonymous.” The Court directed the jury administrator to issue 400 summonses. This time, however, the summons would not be accompanied by the juror questionnaire. Instead, the venire members would be called in groups of sixty to appear in the jury assembly room where they would receive the standard jury orientation and complete the jury questionnaire fashioned for this case. During the afternoon of each session, the jury administrator would provide four copies of the completed jury questionnaire to the District Court. The Court would then provide the questionnaire to counsel to review for the purpose of making challenges for cause and obtaining information for peremptory challenges. The District Court instructed that it would rule on any challenges for cause each afternoon or the following morning. Final selection of the petit jury was tentatively scheduled to begin on January 23, 2008, during which the parties would exercise their peremptory challenges.
Wecht objected inter alia to jury anonymity, the limitation on access to the questionnaires, and the fact that the District Court’s order did “not indicate whether the voir dire questions will be given to each group in open court or even at all prior to the for cause determinations.... ” In a subsequent filing, Wecht objected to the removal of the signature page from the jury questionnaire and requested that each prospective juror be subject to voir dire in open court. On December 4, the Media-Intervenors filed a petition with the District Court objecting to the anonymous jury and the lack of in-person, public voir dire in open court.
On December 21, 2007, Judge Schwab issued a 64-page order reviewing the procedural history relative to jury selection and voir dire. Order of Court Re: Jury Selection, Voir Dire, And Other Pretrial Issues, No. 2:06-cr-00026-AJS (W.D.Pa. Dec. 21, 2007) (“Dec. 21 Order”). In this order, Judge Schwab reiterated his earlier *226declaration that the “jury will be anonymous,” but noted that he preferred the term “innominate jury.”3 Dec. 21 Order at 21 n. 5. Additionally, Judge Schwab opined that the voir dire process would provide the parties with ample information about the jurors. Dec. 21 Order at 21 n. 5. After setting forth the factors to be considered in deciding whether to empanel an anonymous jury, Judge Schwab adhered to his earlier ruling to empanel such a jury. Dec. 21 Order at 28-30. Noting that the Media-Intervenors had challenged the Board of Judges’ order (Misc.06-211) as unconstitutional, he disclaimed any reliance on this order as the basis of his decision. Dec. 21 Order at 18 n. 4.
In his December 21 order, Judge Schwab also explained that “[t]he court will be reviewing the Jury Questionnaire in open court at the same time as counsel, so rulings may be made on the record, with media in attendance to hear the reason(s) for each ‘for cause’ dismissal, by juror number.” Dec. 21 Order at 33. The order further noted that “the Court, as it always has done throughout this case, will provide the media access to the Courtroom during the six (6) day initial jury selection process (including all rulings on the record re: ‘for cause’ decisions); and during voir dire process of the pool of 40 jurors using the individual voir dire questions.... ” Dec. 21 Order at 37.
Judge Schwab explained that “the final jury selection process will commence on January 23, 2008, and copies of the completed Jury Questionnaires of the pool of 40 prospective jurors will be returned only to the counsel, parties, and the Court (with a copy of the last page of the Jury Questionnaire identifying the names and addresses in order by juror number).” Dec. 21 Order at 44. During this final stage of the process, the District Judge noted that he would ask ten voir dire questions of each prospective juror. Dec. 21 Order at 40-41 (listing the questions). The order specified that the Media-Intervenors would, at the conclusion of the trial, be given access to review the jury questionnaire, excluding the last page which contained the juror’s name and address. Dec. 21 Order at 34-35. The order contains no indication that the Media-Intervenors would be given access to the names or addresses of the prospective and trial jurors at any time before or after Wecht’s trial.
The Media-Intervenors timely appealed the December 21 order. They moved for summary reversal under Third Circuit Internal Operating Procedure (“I.O.P.”) 10.6 or, in the alternative, for a stay of jury selection. See Media’s Emergency Motion, United, States v. Wecht, No. 07-4767 (3d Cir. Dec. 26, 2007).4 The Government filed a response opposing the motion (Government’s Response, United States v. *227Wecht, No. 07-4767 (3d Cir. Jan. 2, 2008)5), and Wecht filed a Response supporting it (Wecht’s Response, United States v. Wecht, No. 07-4767 (3d Cir. Jan. 2, 2008)6). The motions panel referred the matter to a merits panel on January 2, 2008, deeming the parties’ submissions on the motion to be their legal briefs and offering the parties an opportunity to file supplemental briefs. See Order, United States v. Wecht, No. 07-4767 (3d Cir. Jan. 2, 2008). After the Media-Intervenors filed a Reply to the Government’s Response (Media’s Reply, United States v. Wecht, No. 07-4767 (3d Cir. Jan. 4, 2008)7 ), all of the parties submitted letters stating that they would rely on their previous filings as their briefs on the merits.8 In an order filed on January 9, 2008, we declared that
[t]o the extent that the District Court’s order restricts access of Media Interve-nor-Appellants and defense counsel to the names of prospective jurors who participate in the selection process prescribed by the District Court, those provisions of the order are VACATED. We leave to the discretion of the District Court the method and timing of disclosure of juror and prospective jurors’ names, except that disclosure of those names shall be made prior to the swearing and empanelment of that jury.9
Order, United States v. Wecht, No. 07-4767 (3d Cir. Jan. 9, 2008). We denied any other relief relative to the voir dire proceedings. Id. We denied the motion for a stay as moot because our order was issued prior to the commencement of trial. Id.
III.
We have jurisdiction to review the Media-Intervenors’ motion under 28 U.S.C. § 1291 and the “collateral order” doctrine.10 See Cohen v. Beneficial Indus. Loan Corp., 337 U.S. 541, 546-47, 69 S.Ct. 1221, 93 L.Ed. 1528 (1949). Under 28 *228U.S.C. § 1291, “[t]he courts of appeals ... shall have jurisdiction of appeals from all final decisions” of district courts. Ordinarily, this rule “prohibits appellate review until conviction and imposition of sentence” in a criminal case. Flanagan v. United States, 465 U.S. 259, 263, 104 S.Ct. 1051, 79 L.Ed.2d 288 (1984). The category of “final decisions” subject to appellate review under § 1291 also includes, however, “collateral orders” that (1) “conclusively determine the disputed question,” (2) “resolve an important issue completely separate from the merits of the action,” and (3) are “effectively unreviewable on appeal from a final judgment.” Coopers & Lybrand v. Livesay, 437 U.S. 463, 468-69, 98 S.Ct. 2454, 57 L.Ed.2d 351 (1978). When deciding whether an order is appeal-able as a collateral order under § 1291, we give this provision of the statute a “practical rather than a technical construction.” Cohen, 337 U.S. at 546, 69 S.Ct. 1221.
In United States v. Schiavo, we held that a District Court’s order restricting the media from publishing certain information about a criminal trial was appealable under the collateral order doctrine because it “determined a matter independent of the issues to be resolved in the criminal proceeding itself, bound persons who were non-parties in the underlying criminal proceeding and had a substantial, continuing effect on important rights.” 504 F.2d 1, 4-5 (3d Cir.1974) (en banc). In United States v. Cianfrani, we applied Schiavo to permit collateral order jurisdiction over the media’s challenge to orders excluding the public and the press from a pretrial hearing and sealing the record. 573 F.2d 835, 845 (3d Cir.1978) (citing Schiavo, 504 F.2d at 4). The Government has challenged our exercise of jurisdiction, relying in large part on Flanagan, in which the Supreme Court held that we must interpret the requirements of the collateral order doctrine “with the utmost strictness in criminal cases” because of “the compelling interest in prompt trials” and the delays that an appeal is likely to create.11 465 U.S. at 265, 104 S.Ct. 1051. Flanagan instructs us to defer appeal until final judgment in a criminal case unless the matter involves “an asserted right the legal and practical value of which would be destroyed if it were not vindicated before trial.” Id. at 266, 104 S.Ct. 1051 (citations and internal quotation marks omitted). At the time Flanagan was decided, the Supreme Court had found only three types of pretrial orders to meet the requirements of the collateral order doctrine: (1) an order denying a motion to reduce bail; (2) an order denying a motion to dismiss an indictment on double jeopardy grounds; and (3) an order denying a motion to dismiss an indictment on speech or debate grounds. Id. at 265-66, 104 S.Ct. 1051. In Sell v. United States, 539 U.S. 166, 176-77, 123 S.Ct. 2174, 156 L.Ed.2d 197 (2003), the Court recognized that an order to forcibly medicate a defendant during trial also meets the requirements for collateral order jurisdiction. What these orders have in common is that neither an acquittal, a post-trial reversal of a conviction, nor any other result can adequately redress the harm that these orders cause. See Flanagan, 465 U.S. at 266-67, 104 S.Ct. 1051. The Supreme Court has not addressed, post-Flanagan, whether a right of access claim raised by a media outlet in a criminal case would satisfy the collateral order doc*229trine.12 The issue before us, therefore, is not only whether the instant order would be appealable under the principles of Schiavo and Cianfrani but also whether Flanagan has effectively overruled these cases.
Because Flanagan has its greatest impact on the third requirement of the collateral order doctrine (i.e., that the decision must be “effectively unreviewable on appeal from a final judgment”), we address this requirement first. We conclude that it would be impossible for us to vindicate the public’s asserted right of access if we foreclosed appeal of this matter until after the final judgment. As with the orders discussed in Flanagan and Sell, the potential harm caused by an improper order restricting the public’s right of access to a criminal trial is not adequately redressable on appeal after final judgment, regardless of the trial’s outcome. We have observed in similar cases that “contemporaneous review [of judicial proceedings] by the public ‘is an effective restraint on possible abuse of judicial power.’ ” United States v. Smith, 787 F.2d 111, 113 (3d Cir.1986) (quoting Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 596, 100 S.Ct. 2814, 65 L.Ed.2d 973 (1980) (Brennan, J., concurring in the judgment)); see also Republic of Philippines v. Westinghouse Elec. Corp., 949 F.2d 653, 660 (3d Cir.1991); United States v. Criden, 648 F.2d 814, 821 (3d Cir.1981). Knowledge of jurors’ identities aids public review by enabling the public to “verify the impartiality of key participants in the administration of justice.” In re Globe Newspaper Co., 920 F.2d 88, 94 (1st Cir.1990).
The Government argues that we can fully vindicate any such right via post-trial release of information, allowing us to defer review until the final judgment. Although post-trial release of information may be better than none at all, the value of the right of access would be seriously undermined if it could not be contemporaneous. See, e.g., Grove Fresh Distribs. v. Everfresh Juice Co., 24 F.3d 893, 897 (7th Cir.1994) (“To delay or postpone disclosure undermines the benefit of public scrutiny and may have the same result as complete suppression.”). We do not suggest, of course, that public disclosure of information related to judicial proceedings must always be contemporaneous, given the interests in security and the other grave concerns that might outweigh the right of access in a particular case. Indeed, “stronger reasons to withhold juror names and addresses will often exist during trial than after a verdict is rendered.” Globe Newspaper, 920 F.2d at 91 (emphasis in original). But the value of contemporaneous disclosure, as opposed to post-trial disclosure, is significant enough to justify our immediate review of the matter under the collateral order doctrine. Accordingly, we decline to hold that Flanagan undermines *230our conclusion in Schiavo and Cianfrani that right of access claims are immediately appealable.
We turn now to the other requirements for application of the collateral order doctrine. The first requirement is that the order must “conclusively determine the disputed question.” Coopers & Lybrand, 437 U.S. at 468, 98 S.Ct. 2454. We cannot review any decision that is “tentative, informal or incomplete.” Cohen, 337 U.S. at 546, 69 S.Ct. 1221. The District Court’s December 21, 2007 order is a conclusive determination that the names of prospective jurors will not be available to the Media-Intervenors at any time before or after the trial and that voir dire will be conducted by written questionnaire until the pool of jurors is reduced to forty.13 We reach this conclusion based on the fact that the 64-page order explicitly considers and rejects the Media-Intervenors’ objections to this procedure. There is no reason to believe that any subsequent developments would have led the District Court to reconsider its conclusion.14 The Government does not dispute that the District Court’s order conclusively determined the question.15 See Government’s Response at 8-9 (“the order regarding jury selection procedures may satisfy the first requirement of the collateral order doctrine, ie., conclusively determining the disputed question ... ”). Accordingly, we conclude that the District Court’s order satisfies the first requirement of the collateral order doctrine.
The second requirement is that the order “resolve an important issue completely separate from the merits of the action.” Coopers & Lybrand, 437 U.S. at 468, 98 S.Ct. 2454. This is sometimes divided into two sub-requirements: (a) the issue must be important; and (b) the issue must be completely separate from the merits of the action. The Supreme Court has defined an important issue as one involving interests that are “weightier than the societal interests advanced by the ordinary operation of final judgment principles,” Digital Equip. Corp. v. Desktop Direct, 511 U.S. 863, 879, 114 S.Ct. 1992, 128 L.Ed.2d 842 (1994), or one that is “serious and unsettled,” Cohen, 337 U.S. at 547, 69 S.Ct. *2311221.16 We believe that the question of when a district court may withhold the names of jurors and the content of voir dire proceedings from the public during a criminal trial is important enough to satisfy the first sub-requirement. The District Court’s order implicates the public’s right of access to judicial proceedings, which is a constitutional interest of sufficient weight to permit the possibility of departing from ordinary final judgment principles. Like the orders at issue in Schiavo and Cian-frani, the instant order has “a substantial, continuing effect on important rights.”17 See Cianfrani, 573 F.2d at 845; Schiavo, 504 F.2d at 5. Moreover, as we make clear in our substantive discussion below, the precise question in this case is unsettled. Thus, the sub-requirement that the issue must be “important” is satisfied.
The issue is also completely separate from the merits of the action, i.e., Wecht’s guilt or innocence. We have repeatedly held that orders restricting public access to information are separate from the underlying issues in criminal trials. See, e.g., Cianfrani, 573 F.2d at 841 (concluding that the media’s appeal of an order barring the public from a pretrial suppression hearing and sealing the record of that hearing was separate from the merits of the underlying criminal proceeding); Schiavo, 504 F.2d at 5 (concluding that an order purporting to enjoin newspapers from publishing information “determined a matter independent of the issues to be resolved in the criminal proceeding itself’). The Government objects that the anonymity of the jury is “intimately tied to the merits” because “the jury is the entity that will decide the ultimate issues of guilt or innocence in this case.” Government’s Response at 9. But the relevant question is whether the issues presented in the Media-Intervenors’ right of access claim are tied to Wecht’s guilt or innocence, not whether the appealed issue and the merits involve the same “entity.” In Sell v. United States, the Supreme Court found that the issue of whether the defendant must be forcibly medicated in order to stand trial was separate from the merits, even though the District Court had found that the medication served “the government’s compelling interest in obtaining an adjudication of defendant’s guilt or innocence.” 539 U.S. 166, 174-76, 123 S.Ct. 2174, 156 L.Ed.2d 197 (2003); see also 539 U.S. at 192, 123 S.Ct. 2174 (Scalia, J., dissenting) (agreeing that the medication order resolved “an important issue separate from the merits” despite concluding that the order did not satisfy the third prong). Similarly, in this case, the District Judge may have believed that establishing jury anonymity would aid the determination of *232guilt or innocence, but the anonymity issue is nonetheless separate from any issue presented by the merits. Thus, we believe that the sub-requirement that the issue be “completely separate from the merits of the action” is satisfied in this case.18
The Government presents two additional challenges to our ability to hear this appeal. First, the Government argues that the media lacks standing because it “does not have a public right of access under the First Amendment or the common law to the jurors’ names and/or the jurors’ questionnaires” and thus suffered no “injury in fact.” Government’s Response at 11-12. Because we conclude that the Media-In-tervenors have a right to the jurors’ names, for the reasons discussed below, we reject this argument and conclude that the Media-Intervenors have standing.19 The Media-Intervenors do not have standing, however, to challenge directly the constitutionality of the Board of Judges’ order (Misc.06-211), because the District Court did not rely on this as a basis for its decision.20 See Dec. 21 Order at 18 n. 4. Second, the Government argues that as a prudential matter, we should dismiss the Media-Intervenors’ appeal as untimely because they did not appeal until December 2007, shortly before trial was scheduled to begin in January 2008. Government’s Response at 12-13. The Government asserts that the District Court had made its intention to establish an anonymous jury clear in July 2006, one-and-a-half years before the Media-Intervenors appealed. The Media-Intervenors respond that the July 2006 order did not clearly establish that the jury would be anonymous, and in any event was not a final order since it explicit*233ly indicated that it was subject to modification. Media’s Emergency Motion at 6-7. Because the media acts as a surrogate for the public in asserting a right of access, see Richmond Newspapers v. Virginia, 448 U.S. 555, 572, 100 S.Ct. 2814, 65 L.Ed.2d 973 (1980), we decline to reject the appeal even assuming arguendo that the Media-Intervenors were not diligent in asserting this right.
IV.
The Media-Intervenors seek reversal of the order because it requires that the prospective and trial jurors be anonymous and because it creates a voir dire process that relies solely on written questionnaires without jurors being physically present in the courtroom prior to reduction of the venire to a pool of forty. Because they rely primarily on arguments that the First Amendment creates a right of access that requires disclosure of jurors’ names and the conducting of voir dire in open court,21 we briefly review the right of access jurisprudence of the Supreme Court and our court.
In Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 100 S.Ct. 2814, 65 L.Ed.2d 973 (1980), the Supreme Court22 held that “the right to attend criminal trials is implicit in the guarantees of the First Amendment” because “without the freedom to attend such trials, which people have exercised for centuries, important aspects of freedom of speech and ‘of the press could be eviscerated.’ ” 448 U.S. at 580, 100 S.Ct. 2814 (quoting Branzburg v. Hayes, 408 U.S. 665, 681, 92 S.Ct. 2646, 33 L.Ed.2d 626 (1972)). The Court said that this right encompassed both a “right of access” and a “right to gather information,” and that the media’s right is no less important than that of the general public. Id. at 576-77 & n. 12, 100 S.Ct. 2814. In Press-Enterprise Co. v. Superior Court, 464 U.S. 501, 104 S.Ct. 819, 78 L.Ed.2d 629 (1984) (“Press-Enterprise I ”), the Supreme Court held that this right of access and to gather information applies to voir dire in criminal trials as well. 464 U.S. at 508, 104 S.Ct. 819. The Court explained that “[t]he presumption of openness may be overcome only by an overriding interest based on findings that closure is essential to preserve higher values and is narrowly tailored to serve that interest” and that “[t]he interest is to be articulated along with findings specific enough that a reviewing court can determine whether the closure order was properly entered.” 464 U.S. at 510, 104 S.Ct. 819.
To determine what aspects of a criminal trial are subject to a presumptive right of public access under the First Amendment, the Court created the “expe*234rience and logic” test in Press-Enterprise Co. v. Superior Court, 478 U.S. 1, 8-9, 106 S.Ct. 2735, 92 L.Ed.2d 1 (1986) (“Press-Enterprise II ”). This test requires courts to weigh two “complementary considerations.” Id. at 8, 106 S.Ct. 2735. Under the “experience” prong, a court considers “whether the place and process have historically been open to the press and general public.” Id. Under the “logic” prong, a court considers “whether public access plays a significant positive role in the functioning of the particular process in question” by, inter alia, enhancing “both the basic fairness of the criminal trial and the appearance of fairness so essential to public confidence in the system.” Id. at 8-9, 106 S.Ct. 2735 (citation omitted). If an aspect of a criminal trial “passes these tests of experience and logic, a qualified First Amendment right of public access attaches.” Id. at 9-10, 106 S.Ct. 2735. As Press Enterprise I made clear, even when such a right of access exists, it is merely presumptive and may be overcome if the District Court articulates specific facts that justify closure.
Whether there is a First Amendment right to have access to a particular aspect of a judicial proceeding is a question of law that we review de novo. See United States v. Antar, 38 F.3d 1348, 1356-57 (3d Cir.1994). Outside of the First Amendment context, we use an abuse of discretion standard to review a District Court’s decisions regarding jury anonymity, United States v. Scarfo, 850 F.2d 1015, 1023 (3d Cir.1988), and its conduct of voir dire, Butler v. City of Camden, 352 F.3d 811, 815 (3d Cir.2003).23 To the extent that we consider whether the District Court has articulated findings sufficient to overcome a presumptive right of access under the First Amendment, however, we conduct “substantially broader” review that “includes independent consideration of the district court’s order and the factual findings inferred from the evidence before it.” In re Capital Cities/ABC, Inc.'s, 913 F.2d 89, 92 (3d Cir.1990) (citing Bose Corp. v. Consumers Union, 466 U.S. 485, 499, 104 S.Ct. 1949, 80 L.Ed.2d 502 (1984); New York Times Co. v. Sullivan, 376 U.S. 254, 285, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964)); see also Antar, 38 F.3d at 1357; United States v. Smith, 787 F.2d 111, 113 n. 1 (3d Cir.1986).
V.
The Media-Intervenors argue that the First Amendment requires disclosure of the jurors’ names prior to empanelment of the jury in this case. This question is one of first impression in our circuit.24 *235Following the framework established in Press-Enterprise II, we will examine first whether the “experience and logic” test supports a conclusion that the presumptive right of access to criminal proceedings includes a right of access to jurors’ names. See Press-Enterprise II, 478 U.S. at 7-10, 106 S.Ct. 2735. If so, we will examine whether this presumption of openness is overcome by particularized findings in the record “establishing the existence of a compelling government interest,” and “demonstrating that absent limited restrictions on the right of access, that other interest would be substantially impaired.” Antar, 38 F.3d at 1359 (citing Press-Enterprise II, 478 U.S. at 15, 106 S.Ct. 2735).
A.
The first question before us is whether the “experience and logic” test establishes the existence of a presumptive First Amendment right of access to obtain the names of both trial jurors and prospective jurors prior to empanelment of the jury. We conclude that it does.
1. Experience
In Press-Enterprise I, the Supreme Court traced the development of the jury selection process from the days “before the Norman Conquest,” and concluded that “since the development of trial by jury, the process of selection of jurors has presumptively been a public process with exceptions only for good cause shown.” 464 U.S. at 505-08, 104 S.Ct. 819. Although this historical evidence helps to show that voir dire is traditionally a public process, it does not necessarily establish that the jurors’ names were also known to the public. The Media-Intervenors point out that “there is no suggestion in ... Press-Enterprise I that the names of the jurors were not equally open to the public as the other parts of the voir dire process.” Media’s Emergency Motion at 11-12. This is true, but the opinion contains no suggestion to the contrary, and we are reluctant to draw conclusions solely based on the Court’s silence about a question that was not before it.
Because juries have historically been selected from local populations in which most people have known each other, however, the traditional public nature of voir dire strongly suggests that jurors’ identities were public as well. Case law and legal commentary confirm this suggestion. See In Re Baltimore Sun Co., 841 F.2d 74, 75 (4th Cir.1988) (“When the jury system grew up with juries of the vicinage, everybody knew everybody on the jury ...,” so requiring the public disclosure of the names of trial and prospective jurors upon empanelment is “no more than an application of what has always been the law... .”)25; David Weinstein, Protecting A Juror’s Right to Privacy: Constitutional Constraints and Policy Options, 70 Temp. L.Rev. 1, 30 (1997) (“The names of jurors have been available to the public throughout the history of the common law.... Moreover, the records of early jury trials evince little concern for protect*236ing juror anonymity.”); Robert Lloyd Ras-kopf, A First Amendment Right of Access to a Juror’s Identity: Toward a Fuller Understanding of the Jury’s Deliberative Process, 17 Pepp. L.Rev. 357, 370 (1990) (“An examination of historical tradition indicates that jurors’ identities and places of residence traditionally have been known to the public.”). We find it significant that instances of courts withholding jurors’ names appear to be very rare before the 1970s. See, e.g., Ephraim Margolin & Gerald F. Uelmen, The Anonymous Jury: Jury Tampering By Another Name?, 9 Crim Just. 14, 14 (1994) (“Juror anonymity is an innovation that was unknown to the common law and to American jurisprudence in its first two centuries.”). Neither the District Court nor the Government cite any such cases in their discussion of the “experience” prong. See Dec. 21 Order at 26; Government’s Response at 16-18. Moreover, none of the federal and state opinions that analyze the “experience” prong have cited any pre-1970s cases in which jurors’ names were not publicly known prior to empanelment. See, e.g., United States v. Black, 483 F.Supp.2d 618, 623-26 (N.D.Ill.2007); Commonwealth v. Long, 592 Pa. 42, 922 A.2d 892, 901-03 (2007); Beacon Journal v. Bond, 98 Ohio St.3d 146, 781 N.E.2d 180, 193 (2002); Gannett Co., Inc. v. State of Delaware, 571 A.2d 735, 743-48 (Del.1990). Such cases exist,26 but they are rare.27 Based on the evidence before us, it appears that public knowledge of jurors’ names is a well-established part of American judicial tradition.
The Government’s strongest argument that there is no such tradition of openness is based on 28 U.S.C. § 1863(b)(7), which instructs District Courts to put into effect a jury selection plan that will
fix the time when the names drawn from the qualified jury wheel shall be disclosed to parties and to the public. If the plan permits these names to be made public, it may nevertheless permit the chief judge of the district court, or such other district court judge as the plan may provide, to keep these names confidential in any case where the interests of justice so require.
28 U.S.C. § 1863(b)(7) (2000) (emphasis added). When Congress enacted this provision in 1968, the accompanying legislative history explained that it “permits the present diversity of practice to continue. Some district courts keep juror names confidential for fear of jury tampering. Other district courts routinely publicize the names.” See In re Globe Newspaper, 920 F.2d 88, 92 (1st Cir.1990) (quoting H.R.Rep. No. 1076, 90th Cong., 2d Sess., reprinted in 1968 U.S.C.C.A.N. 1792, 1801). The Government argues that the *237statute and this passage from its legislative history suggest that, at least as of 1968, no strong tradition of juror name disclosure existed. Government’s Response at 16-17.
We are reluctant to afford the relevant language of the statute significant weight in this context. Nothing in the statute itself indicates whether Congress believed that allowing federal courts to withhold juror names was consistent with historical practice or a significant departure. The House Report is more explicit on this point, stating that the statute was merely intended to preserve the “diversity of practice” that existed in federal courts as of 1968. But the House Report is not an official expression of Congress’s views, and its persuasive value is limited because it provides no evidence to support its claim that such a “diversity of practice” existed at the time. Moreover, even if the House Report accurately characterizes the practices of federal district courts in 1968, it is nonetheless consistent with the proposition that withholding the names of jurors is a relatively recent phenomenon. Given the Supreme Court’s suggestion that a proper analysis of “experience” will evaluate trial practices as they have developed over the past millennium in courts at all levels, see Press-Enterprise II, 478 U.S. at 8, 106 S.Ct. 2735, we cannot discern whether jurors’ names have traditionally been public based on an assertion in legislative history forty years ago that some degree of “diversity of practice” existed in the federal system.
The reports published in 1968 and 1980 by the Committee on the Operation of the Jury System of the Judicial Conference of the United States are not to the contrary. See Report of the Committee on the Operation of the Jury System on the “Free Press-Fair Trial” Issue, reprinted in 45 F.R.D. 391 (1968); Revised Report of the Judicial Conference Committee on the Operation of the Jury System on the “Free Press-Fair Trial” Issue, reprinted in 87 F.R.D. 519 (1980). Both of the Committee’s reports recommend that District Courts adopt a rule that allows judges, “in a case which is likely to attract unusual publicity,” to issue a “special order” that “might be addressed” to a variety of subjects, including a direction that “the names and addresses of jurors or prospective jurors not be publicly released except as required by statute.”28 45 F.R.D. at 409-11, 87 F.R.D. at 529-31. We do not dispute that a trial judge has historically had the power to issue such an order in special cases. We conclude only that a tradition of openness exists and that anonymous juries have been the rare exception rather than the norm.
In short, we believe that the “experience” prong of the Press-Enterprise II test favors a conclusion that jurors’ names have traditionally been available to the public prior to the beginning of trial. If any significant evidence to the contrary exists, we have not discovered it in our review of case law and commentary on this question.
*238 2. Logic
We next consider whether presumptive public access to jurors’ names prior to empanelment “plays a significant positive role in the functioning” of the criminal justice system. See Press-Enterprise II, 478 U.S. at 8, 106 S.Ct. 2735. As the First Circuit explained in In re Globe Newspaper Co., the purposes served by the openness of trials and voir dire generally are also served by public access to the jurors’ names:
Knowledge of juror identities allows the public to verify the impartiality of key participants in the administration of justice, and thereby ensures fairness, the appearance of fairness and public confidence in that system. It is possible, for example, that suspicions might arise in a particular trial (or in a series of trials) that jurors were selected from only a narrow social group, or from persons with certain political affiliations, or from persons associated with organized crime groups. It would be more difficult to inquire into such matters, and those suspicions would seem in any event more real to the public, if names and addresses were kept secret. Furthermore, information about jurors, obtained from the jurors themselves or otherwise, serves to educate the public regarding the judicial system and can be important to public debate about its strengths, flaws and means to improve it.... Juror bias or confusion might be uncovered, and jurors’ understanding and response to judicial proceedings could be investigated. Public knowledge of juror identities could also deter intentional misrepresentation at voir dire.
920 F.2d at 94. Public access to jurors’ names is not without risks. First, when the names of jurors are public, friends or enemies of a criminal defendant may find it easier to influence the jury’s decision. In an extreme case, this could take the form of threats to the jurors or their family members. Second, if jurors know that the media will attempt to contact them or their families, they may resist serving on high-profile cases at all because they fear that their privacy will be threatened. Third, public knowledge of jurors’ identities might actually increase the risk of misrepresentation at voir dire, because some jurors will be tempted to lie in order to avoid the disclosure of embarrassing information.29
Despite these risks, we believe that the judicial system benefits from a presumption of public access to jurors’ names. A criminal jury trial vests twelve randomly-selected citizens with the power to decide the fate of someone who the state has targeted for prosecution. We cannot reconcile the Supreme Court’s conclusion that the public has the right to see the process in which this power is exercised (Richmond Newspapers) and to see the process that selects those who will exercise the power (Press-Enterprise I), with the conclusion that the public has no right to know who ultimately exercises this power. As the First Circuit said, “the prospect of criminal justice being routinely meted out by unknown persons does not comport with democratic values of accountability and openness.” Globe Newspaper, 920 *239F.2d at 98. Of course, in a given case, a risk of jury tampering or excessive media harassment may exist. But we are satisfied that district judges are well-positioned to address these risks on a case-by-case basis, and in such cases, to make particularized findings on the record “establishing the existence of a compelling government interest” and “demonstrating that absent limited restrictions on the right of access, that other interest would be substantially impaired.” Antar, 38 F.3d at 1359 (citing Press-Enterprise II, 478 U.S. at 15, 106 S.Ct. 2735). We do not consider these risks so pervasive as to overcome the benefits of public access. We must strike the balance in favor of presumptive public access to jurors’ identities.30
Finally, we believe that this presumptive First Amendment right of access to the identities of jurors attaches no later than the swearing and empanelment of the jury. Corruption and bias in a jury should be rooted out before a defendant has to run the gauntlet of trial. Public knowledge of the jurors’ identities is desirable in part because it can deter such corruption and bias. The value of any right of access, then, can only be diminished after trial has begun, and diminished even further once a verdict has been rendered by a corrupt or biased jury.31 We acknowledge that, in the words of the First Circuit, “stronger reasons to withhold juror names and addresses will often exist during trial than after a verdict is rendered.” Globe Newspaper, 920 F.2d at 91 (emphasis in original). But we do not believe that these reasons are so compelling that they negate altogether the existence of a First Amendment right of access to the names during trial. Rather, a presumption of openness exists at the latest at the time of the swearing and empanelment of the jury, regardless of the fact that a judge may find “stronger reasons” for overcoming this presumption during trial.
B.
We now consider whether the District Court articulated the necessary findings and consideration of alternatives to overcome the presumption that the jurors’ names should be publicly available. It provided three reasons for exercising its *240discretion32 to empanel an anonymous jury; we discuss these reasons in turn.
First, the District Court said that withholding the jurors’ names is necessary to prevent the media from publishing stories about them:
First, from the prospective [sic ] of the media, because the media requests the names and addresses of the potential jurors, if those requests were granted, there is certainly a real potential that the media would use those names (and addresses) to develop and publish stories about the prospective jurors, coupled with possible interviews of the potential jurors’ family members, coworkers, and friends. The media obviously does not want the jurors’ names as an intellectual exercise to file in some reporter’s electronic desk drawer. If they want the names, they want to do “reporting.” If the numerous excellent “investigatory” reporters in Western Pennsylvania obtain the names and home address of the jurors, detailed “background” stories, before and during the trial, are likely. The Court thus has serious concerns that the dissemination of stories about the prospective jurors (and especially the empaneled jury) would have a real impact on the jurors’ willingness to serve and, if selected, on the jurors’ abilities to remain fair, unbiased, and focused on this case.
Dec. 21 Order at 28-29. The prospect that the press might publish background stories about the jurors is not a legally sufficient reason to withhold the jurors’ names from the public. Although such stories might make some jurors less willing to serve or more distracted from the case, this is a necessary cost of the openness of the judicial process.33 The participation of jurors “in publicized trials may sometimes force them into the limelight against their wishes,” but “[w] e cannot accept the mere generalized privacy concerns of jurors” as a sufficient reason to conceal their identities in every high-profile case. See Globe Newspaper, 920 F.2d at 98. The District Court has not established that there is anything unusual about this case, aside from a locally prominent defendant, that makes the prospective jurors’ hypothetical privacy concerns more compelling than usual. The District Court’s statements amount to the sort of “conclusory and generic” finding that we have held to be insufficient to overcome the presumption of openness.34 See Antar, 38 F.3d at 1363.
Second, the District Court cited the possibility that friends or enemies of Wecht would attempt to influence the jurors:
Secondly, from the perspective of the defendant, if there is media coverage disseminating the names (and addresses) of the prospective jurors, that coverage would undoubtedly increase the risk of intimidation of those jurors as there is a probability that other individuals (not *241including the defendant himself) would contact those jurors in an attempt to either hurt or bolster defendant’s case. Just like the district court in the Scarfo case had concerns that persons hostile to defendant might have been inclined to harass the jurors, this Court also has real concerns that persons who are either hostile to, or enamored with, defendant would attempt to influence the jurors.
Dec. 21 Order at 29. This explanation is insufficient to justify withholding the names in this case. Again, this is a “con-clusory and generic” finding that cannot overcome the presumption that jurors’ names are public information. In fact, the District Court’s reasoning would justify anonymity in virtually every jury trial, whether or not it attracts media attention, since almost all defendants have friends and enemies who might be inclined to influence jurors. The District Court’s citation to United States v. Scarfo, 850 F.2d 1015 (3d Cir.1988), is revealing. In Scarfo, we noted that the defendant belonged to an “organized crime group,” had ordered “several murders” (including those of a judge and a prospective witness), and had attempted to bribe judges. Id. at 1017. In short, specific reasons existed in Scarfo to believe that friends of the defendant would threaten or bribe the jurors. The District Court in this case has not provided anything closely resembling the specific reasons offered in Scarfo.35
Finally, the District Court quoted extensively from a document that Wecht filed that purports to establish that he has acquired many enemies. This document points out that Wecht “has made countless cause and manner-of-death determinations” as a witness in “hundreds of homicide and other criminal trials,” some of which involve “the most serious offenders of society — violent criminals,” and that he remains a witness in “pending criminal homicide trials.” Dec. 21 Order at 29 (quoting Defendant’s Brief in Support of Motion For Clarification And/Or Modification of Trial Procedures and Scope of Exhibit and In Limine Rulings at 5, United States v. Wecht, No. 2:06-cr-00026-AJS (W.D.Pa. Dec. 4, 2007) (“Defendant’s Brief’)). In addition, the document says, he has participated in “high-profile” civil cases, including wrongful death actions. Id. at 30 (quoting Defendant’s Brief at 6). As a result, many people “may harbor ill will” or “bear animus” toward Wecht. Id. (citing Defendant’s Brief at 5-6). Moreover, the document says that Wecht’s son, as a “sitting family court judge in Allegheny County ... [,] makes judicial decisions that affect people in the most emotional and passionate areas of their fives,” and that some of these people “may feel wronged as a result of his judicial decisions.” Id. The document points out that “[tjhose individuals may find their way into Dr. Wecht’s pool of jurors.” Id.
The District Court cited the statements in this document to support its conclusion that, unless the jury is anonymous, Wecht’s many enemies might attempt to *242influence jurors. (The District Court also suggests, without citing any evidence, that “presumably ‘unknown’ friends” of Wecht might also attempt to influence the jury. Dec. 21 Order at 30.) Wecht made these statements, however, in support of the opposite conclusion: that the jury should not be anonymous because the defense and the media must be able to ensure that Wecht’s enemies do not enter the jury pool without being detected. Defendant’s Brief at 5-11. As we have explained, one of the purposes of access to jurors’ names is to make this type of investigation possible.36 Moreover, the quoted statements consist largely of speculation that people might be hostile toward Wecht; they describe no specific instances in which Wecht’s enemies or friends had threatened or harassed anyone. The mere fact that people might have passionate opinions about a defendant is not enough to justify an anonymous jury. The District Court must articulate some reason to conclude that the risks that such people pose to the jurors are serious and specific enough to justify depriving the public (and, in this case, the defendant) of knowledge of the jurors’ identities. Because the District Court did little more in this case than quote factual assertions that Wecht offered in opposition to jury anonymity, we conclude that it did not overcome the presumption in favor of disclosure.
VI.
The Media-Intervenors also challenge the voir dire procedure adopted by the District Court. They contend that by using Juror Questionnaires instead of in-court voir dire to make “for cause” determinations until the venire has been reduced to forty prospective jurors, the District Court violates their First Amendment right of access to voir dire proceedings.37 They do not request immediate access to the actual questionnaires but instead demand that the District Court conduct voir dire in open court in addition to using the questionnaires. Media’s Emergency Motion at 4 n. 1. According to the Media-Intervenors, they seek in-court voir dire in order to allow public access to “information traditionally revealed during voir dire (juror names, area where a juror lives, employment, family, etc.).” Id. at 16.
We reject the Media-Interve-nors’ request for two related reasons. First, unlike the Media-Intervenors’ request for the names of prospective jurors, the request for in-court voir dire is not merely a request for access to information, but a request that the District Court conduct a specific procedure and that the Media-Intervenors have access to that procedure. It is well-established that “the method of conducting the voir dire is left to the sound discretion of the district court.” Waldorf v. Shuta, 3 F.3d 705, 710 (3d Cir.1993). In a voir dire process involving 400 prospective jurors, we believe that a trial judge has discretion to conduct part of the process exclusively based on written questionnaires, so long as it is consistent with established procedural *243rules (e.g., Fed.R.CRIm.P. 24). Second, we believe that our order requiring the District Court to release the names of prospective jurors grants the Media-Interve-nors most of the relief they seek. Because the prospective jurors will not be anonymous, the Media-Intervenors will have available information to investigate and detect possible improper bias in “for cause” determinations, should they elect to do so.
VII.
In sum, we have articulated in this opinion the reasons supporting our January 9 Order vacating the provisions of the District Court’s order that restricted access of Media-Intervenors and defense counsel38 to the jurors’ names.
. Our order decreed that "juror and prospective jurors' names” shall be disclosed prior to the swearing and empanelment of the jury. The term "prospective juror” refers to a member of the venire. The term "juror” refers to a member of the venire who is chosen to be part of the actual trial jury. For the sake of clarity, this opinion will use the term "trial jurors” to describe people in the latter category. We also emphasize that, referring to the names of "juror[s] and prospective jurors,” we anticipated that the District Court's disclosure would distinguish between the names of trial jurors and those of prospective jurors. It appears that in the disclosure that followed the entry of our order, the District Court did not make such a distinction. See Notice of Filing Prospective Juror List, United States v. Wecht, No. 06-CR-26 (W.D.Pa. Jan. 23, 2008).
. In May of 2006, the Media-Intervenors moved to intervene in an effort to unseal certain court filings. In response to those motions, the Court unsealed certain documents. Wecht and the Government appealed several rulings, and Wecht filed a petition for mandamus seeking review of the District Court's denial of his motion to recuse. See United States v. Wecht, 484 F.3d 194 (3d Cir.2007). Pretrial preparation proceeded in the District Court during the pendency of those appeals.
. Judge Schwab provided two reasons for his preference for this term. First, he said that the term "anonymous” inappropriately connoted a "clandestine, forbidden, and obscure jury panel.” Dec. 21 Order at 21 n. 5 (quoting United States v. Carpa, 271 F.3d 962, 963 n. 1 (11th Cir.2001)). Second, he suggested that the jury was not really anonymous because the parties knew everything about the jurors but their names. Id. (citing United States v. Bowman, 302 F.3d 1228, 1236 (11th Cir.2002)). Judge Schwab nonetheless uses the terms "anonymous” and "innominate” interchangeably in his December 21 order. In this opinion, we will use the term "anonymous.”
. The full title of this application is "The Media’s Emergency Motion For Summary Reversal Of The District Court’s Order Dated December 21, 2007, Or, In The Alternative, Expedited Relief In The Form Of A Stay Of Jury Selection And Trial Proceedings Pending Disposition Of This Appeal.”
. The full title is "Government’s Response To The Media's Emergency Motion For Summary Reversal Of The District Court’s Order Dated December 21, 2007, Or, In The Alternative, Expedited Relief In The Form Of A Stay Of Jury Selection And Trial Proceedings Pending Disposition [Of] This Appeal.”
. The full title is "Cyril H. Wecht's Response To The Media’s Emergency Motion For Summary Reversal Of The District Court’s Order Dated December 21, 2007, Or, In The Alternative, Expedited Relief In The Form Of A Stay Of Jury Selection And Trial Proceedings Pending Disposition Of This Appeal.”
. The full title of the Media’s Reply is "The Media's Reply in Support of Emergency Motion for Summary Reversal of the District Court’s Order dated December 21, 2007, or, in the Alternative, Expedited Relief in the Form of a Stay of Jury Selection and Trial Proceedings Pending Disposition of this Appeal.”
. The dissent argues that we would have benefited from "additional briefing” given the fact that the parties' original filings focused on the Media-Intervenors’ request for summary reversal or a stay. Dissent, infra, at 58 n. 44. As noted above, we offered the parties an opportunity to file supplemental briefs after the matter was referred to a merits panel, and they chose to rely on their previous filings.
. We noted that Media-Intervenors were "not seeking access to the jurors' home addresses or the actual jury questionnaire.” Order at 2 n. 1, United States v. Wecht, No. 07-4767 (3d Cir. Jan. 9, 2008). For that reason, we do not address those issues here.
. In our January 9 order, we stated: "Appellate jurisdiction exists under the collateral order doctrine.” Order at 2, United States v. Wecht, No. 07-4767 (3d Cir. Jan. 9, 2008) (citing Cohen v. Beneficial Indus. Loan Corp., 337 U.S. 541, 69 S.Ct. 1221, 93 L.Ed. 1528 (1949)). Our dissenting colleague now argues that the collateral order doctrine did not provide us with jurisdiction to hear this appeal. See Dissent, infra, at 243-50. We are surprised that he did not raise this purported jurisdictional defect in his dissent to our Jan-uaiy 9 order.
. We note that our decision to grant review in this case does not threaten "the compelling interest in prompt trials.” 465 U.S. at 265, 104 S.Ct. 1051. Because we declined to grant the media’s request for a stay, and because releasing the jurors’ names did not create a time-consuming burden for the District Court, our ruling has posed no danger to the interests of either Wecht or the public in a speedy trial.
. In ABC, Inc. v. Stewart, 360 F.3d 90 (2d Cir.2004), the Second Circuit held that an order affecting the right of access in a criminal case was appealable under the collateral order doctrine. Id. at 97. Without mentioning Flanagan, the court deemed the order appealable for two reasons. First, the court said that the district court had "in effect allowed the Media Coalition to intervene in the pending criminal proceeding for the limited purpose of challenging” the order, and that the order was therefore final and appealable as to the intervenors. Id. (citing In re Herald Co., 734 F.2d 93, 96 (2d Cir.1984)). Second, the court said that because the Media Coalition's claims "could have been treated by the district court as a new civil case, as opposed to an intervention in the pending criminal case, and the orders would have been final in that case[,][n]o jurisdictional significance should attach simply because the district court chose to treat appellants as intervenors in the criminal proceeding.” Id. (citing In re New York Times Co., 828 F.2d 110, 113 (2d Cir.1987)).
. The Dec. 21 Order does not explicitly address whether the venirepersons will be present when counsel reviews the jury questionnaire in open court.
. The dissent argues that the December 21 order was not final because if the Media-Intervenors or the defendant had petitioned for modification of the December 21 order, “they may well have been successful.” Dissent, infra, at 246. Given the fact that the District Court had already considered and rejected their arguments, we find it unlikely that they could have achieved the desired modification by making these arguments again. Of course, it was theoretically possible that the District Court would have modified the December 21 order, either in response to a petition for modification or sua sponte. This possibility does not eliminate finality, however, because the first requirement of the collateral order doctrine may be satisfied when "there is no basis to suppose that the District Judge contemplated any reconsideration of his decision.” Moses H. Cone Mem. Hosp. v. Mercury Constr. Corp., 460 U.S. 1, 12-13, 103 S.Ct. 927, 74 L.Ed.2d 765 (1983). Because the District Court’s December 21 order indicates that it is intended to be the final word on the Media-Intervenors' objections, it was final for purposes of the collateral order doctrine.
.We recognize that the Government's failure to challenge the finality of the order is not dispositive, since it is always our duty to ensure that we are properly exercising jurisdiction. To the extent that the finality of the order is a factual issue, however, the Government may be in a better position than we are to evaluate whether the District Court would have considered a request for modification. Therefore, its decision not to challenge this requirement of the collateral order doctrine is notable.
. At times, courts have treated importance as a fully independent requirement of the collateral order doctrine. See, e.g., Nixon v. Fitzgerald, 457 U.S. 731, 742, 102 S.Ct. 2690, 73 L.Ed.2d 349 (1982) ("As an additional requirement, Cohen established that a collateral appeal of an interlocutory order must ‘[present] a serious and unsettled question.’ ” (quoting Cohen, 337 U.S. at 547, 69 S.Ct. 1221)). The exact role of the importance requirement in the analysis has been the subject of debate. See generally 15A Charles Alan Wright, Arthur R. Miller & Edward H. Cooper & et al., Federal Practice and Procedure § 3911.5 (2d ed. 1992).
. In a footnote, the Government "questions the importance of the Media's ability to write articles that include jurors' names as opposed to articles without them.” See Government’s Response at 9 n. 6. As we explain below, however, the release of juror names can be an important part of the public's right of access. In the words of the First Circuit in In re Globe Newspaper Co., "[kjnowledge of juror identities allows the public to verify the impartiality of key participants in the administration of justice, and thereby ensures fairness, the appearance of fairness and public confidence in that system.” 920 F.2d 88, 94 (1st Cir.1990).
.The Government cites several cases in support of a contrary conclusion. First, it cites Sell for the proposition that an issue satisfies the "completely separate” requirement only when it is separate "from questions concerning trial procedures.” 539 U.S. at 176, 123 S.Ct. 2174. Sell does not define a "trial procedure” or explain whether this means something different from being separate from the merits. In any case, we have discussed Sell above and concluded that it supports the conclusion that jury anonymity is completely separate from Wecht's guilt or innocence. Second, the Government cites United States v. McVeigh, 106 F.3d 325, 332 (10th Cir.1997), which held that an order barring victim-impact witnesses from observing the guilt phase of a capital trial was not separate from the merits of the action. We decline to rely on McVeigh, however, because it interprets the Criminal Appeals Act, 18 U.S.C. § 3731, rather than the collateral order doctrine. 106 F.3d at 329; see also id. at 331 (“... when the government seeks review in a criminal case, concerns unaddressed by Cohen come into play.”). Third, the Government cites dicta from United States v. Green, 407 F.3d 434, 438 (1st Cir.2005), which addresses the government's attempt to appeal from a pretrial order establishing two separate juries for guilt and sentencing. Like McVeigh, Green involves a government appeal in a criminal case, and thus would not be controlling even if it were a holding from our own circuit rather than dicta from another.
. Contrary to the Government’s suggestion, the Media-Intervenors are not seeking access to the questionnaires, so we need not consider whether they have a “right” to access them for standing purposes. Pursuant to Press-Enterprise I, the Media-Intervenors have at least a presumptive right of access to voir dire proceedings, so we conclude that they have standing to the extent that they claim that they challenge the District Court's voir dire rulings on right of access grounds.
. The District Court states later in its order that "in accordance with Misc. Rule 06-211, the Court (and no counsel or party) will neither read nor state the names or addresses of prospective jurors in open court, nor will they provide the media or any person or party access to the names or addresses of the prospective or empaneled jurors.” Dec. 21 Order at 37. Despite this, it appears that the District Court did not rely on Misc. 06-211 even if it might have acted "in accordance with” it. Thus, we will not consider the constitutionality of Misc. 06-211 in our evaluation of the District Court's order.
. The Media-Intervenors also state that the common law or the Third Circuit's supervisory powers establish a right of access to criminal proceedings that includes disclosure of the jurors' names and voir dire in open court. Media's Emergency Motion at 10-11 (citing In Re The Baltimore Sun Co., 841 F.2d 74 (4th Cir.1988) (common law); United States v. Criden, 675 F.2d 550 (3d Cir.1982) (supervisory powers)). They do not, however, make a serious effort to develop these alternative grounds for a right of access aside from their citations to these cases. Thus, we will focus only on their argument that a right of access exists under the First Amendment.
. Although no opinion in Richmond Newspapers commanded a majority, seven of the eight justices who participated in the case recognized that the First Amendment embodies a right to attend criminal trials. See 448 U.S. at 558-81, 100 S.Ct. 2814 (plurality opinion); id. at 584-98, 100 S.Ct. 2814 (Brennan, J., concurring in judgment); id. at 598-601, 100 S.Ct. 2814 (Stewart, J., concurring in judgment); id. at 601-04, 100 S.Ct. 2814 (Blackmun, J., concurring in judgment).
. We point out that the First Amendment right of access that the Media-Intervenors assert is distinct from a defendant’s Sixth Amendment right to challenge the use of an anonymous jury.
. Some have argued that the Supreme Court settled this question in Press-Enterprise I by implying that jurors’ names are an inseparable component of voir dire. Discussing a potential conflict between the First Amendment presumption of openness for voir dire, and jurors' privacy interests, the Court said:
When limited closure [of voir dire ] is ordered, the constitutional values sought to be protected by holding open proceedings may be satisfied later by making a transcript of the closed proceedings available within a reasonable time, if the judge determines that disclosure can be accomplished while safeguarding the juror’s valid privacy interests. Even then a valid privacy right may rise to a level that part of the transcript should be sealed, or the name of a juror withheld, to protect the person from embarrassment.
464 U.S. at 512, 104 S.Ct. 819 (emphasis added). Arguably, this passage can be "read to imply that jurors’ identities are part and parcel of voir dire, and as such are governed by the same principles of presumptive access.” Beacon Journal v. Bond, 98 Ohio St.3d 146, 781 N.E.2d 180, 192 (2002) (quoting David Weinstein, Protecting a Juror’s Right to *235Privacy: Constitutional Constraints and Policy Options, 70 Temple L.Rev. 1, 30 (1997)). See also Gannett Co., Inc. v. State of Delaware, 571 A.2d 735, 755 (1990) (Walsh, L, dissenting). Although this argument is plausible, we will not conclude from a single passage of Supreme Court dicta that the question is decisively settled. Rather, we treat the question as unsettled and proceed with the “experience and logic” analysis required by Press-Enterprise II.
. The Fourth Circuit's Baltimore Sun opinion ultimately rests on the common law rather than on the First Amendment. 841 F.2d at 76 n. 4. Nonetheless, its historical evidence is useful for our analysis of the “experience” prong.
. For example, in Hamer v. United States, 259 F.2d 274 (9th Cir.1958), the Ninth Circuit upheld a 1951 order of the United States District Court for the Southern District of California that prohibited revelation of the names and addresses of jurors and prospective jurors to anyone, including the defendant, prior to trial. Id. at 277-80.
. We agree with the observation of the dissenting judge in Gannett Co., 571 A.2d at 757-58, that even if a few cases exist in which courts kept jurors’ names private, this would not by itself prove that no tradition of openness exists. In that judge’s words:
One cannot conclude with certainty that in the entire history of Anglo-American jurisprudence an anonymous jury was never im-panelled [sic] prior to the 1970s. Yet the majority appears to demand that degree of certainty before it would be willing to recognize a tradition of openness. Rather than requiring Gannett to show that a strong presumption of openness exists, the majority would ask it to prove that restrictions have never and could never have been imposed. By contrast, the United States Supreme Court has never required such an impossible standard of proof.
571 A.2d at 757-58 (Walsh, J., dissenting).
. The dissent states that "[t]he Committee recommended that each District Court adopt a rule providing for special measures to be taken in cases likely to receive significant media attention, including” a direction that jurors’ names and addresses be withheld. Dissent, infra, at 254. The dissent’s language ("special measures to be taken ”) suggests that the Committee recommended that District Courts adopt a rule requiring judges to withhold jurors’ names and addresses in high-profile cases. For the sake of clarity, we note that the Committee said that "[sjuch a special order might be addressed to some or all of the following subjects,” including the withholding of names and addresses. 45 F.R.D. at 409 (emphasis added).
. Although these risks may be greater when the jurors’ identities are made public during trial, they often exist even when the jurors’ identities will remain secret until the end of trial. First, jurors might be reluctant to convict a defendant who is known to be dangerous for fear of post-trial retaliation from the defendant's friends. Second, jurors might be reluctant to serve on a jury at all if they think that they will be the subject of media attention post-trial. Third, jurors might lie during voir dire because they fear that, after the trial, sensitive information will be revealed.
. We acknowledge that our conclusion in the instant case may be inconsistent with our dicta in United States v. Scarfo, 850 F.2d 1015 (3d Cir.1988), in which we said that "anonymity would seem entirely consistent with, rather than anathema to, the jury concept" and that “the probable merits of the anonymous jury procedure are worthy, not of a presumption of irregularity, but of disinterested appraisal by the courts.” 850 F.2d at 1023. Unlike the instant case, Scarfo did not involve a First Amendment challenge and did not apply the Press-Enterprise II test to determine whether jurors’ names should be presumptively public. We do not challenge Scar-fo's conclusion that a district court’s decisions about anonymity should be reviewed for abuse of discretion when no one has raised a valid First Amendment challenge. See id.
. The dissent claims that "[cjorruption could just as easily be rooted out post-trial as it could pre-trial.” Dissent, infra, at 257. Although we agree that corruption could be rooted out post-trial, it is far more desirable to discover it pre-trial. It is neither fair nor efficient to subject a defendant to a second trial because the jury in the first trial was tainted. Moreover, post-trial discovery of corruption shakes public confidence in the validity and finality of criminal jury verdicts. Of course, we do not suggest that a public right of access to jury names is the most effective method for uncovering corruption or bias in jury selection before a trial begins. Voir dire, conducted by the parties and the court, has traditionally been the primary method for accomplishing this. Nonetheless, we believe that public access plays an important role in the criminal justice system by allowing the public to verify, before a trial has begun, that the trial will proceed with an impartial jury.
. The District Court found that the media had no First Amendment right to obtain the jurors' names, and therefore described its decision to withhold the names as an exercise of discretion that balanced "competing constitutional interests.” Dec. 21 Order at 26-27.
. The District Court appears to believe that no good can come from any story published about a juror. As we noted above, however, press investigation of jurors might be beneficial in some cases by, for example, revealing possible sources of juror bias or deterring misrepresentation during voir dire.
.Taken to its logical conclusion, the District Court’s argument would allow judges to withhold the names of jurors in every case that might attract media attention. In fact, aside from the reference to "Western Pennsylvania” reporters, any other court could copy the District Court’s statement verbatim to justify an anonymous jury in any high-profile case.
. The District Court said in a footnote that it was "aware of the arguments made by the government regarding defendant's alleged witness intimidation,” and referred to a letter that two individuals sent to the court asking that the jury be anonymous in light of threatening letters that they had received from Wecht. Dec. 21 Order at 30 n. 9. But the District Court denied that it was relying on these alleged threats "as the basis for its decision to empanel an innominate jury,” id., and did not include any findings of fact about them in the record. Thus, even assuming arguendo that these alleged threats provide a justification for an anonymous jury, we cannot rely on them because the District Court was required to place "findings on the record which clearly established that closure was necessary to protect an overriding interest.” Antar, 38 F.3d at 1361.
. Voir dire may detect prospective jurors with obvious connections to Wecht, even if they are anonymous. But if we accept Wecht’s assertion that he has incurred the animus of countless "friends” of those who were affected adversely by his participation in trials, some of these connections might not be obvious. Moreover, any person who feels passionately enough about Wecht to threaten jurors would presumably be willing to lie about his or her connections to Wecht during voir dire, protected by anonymity, in order to earn a spot on Wecht’s jury.
. We need not conduct an "experience and logic” inquiry into whether a public right of access to voir dire proceedings exists, because Press-Enterprise I established that this right exists. 464 U.S. at 508, 104 S.Ct. 819.
. Suffice it to say, the fact that the accused here supports the Media-Intervenors’ appeal — which often is not the case, see, e.g., ABC, Inc., 360 F.3d at 98-99; Black, 483 F.Supp.2d at 620 — bolsters the result we reach today.