OPINION OF THE COURT
A. LEON HIGGINBOTHAM, Jr., Circuit Judge.This is an appeal from an order of the district court denying Philadelphia Newspapers, Inc., Westinghouse Broadcasting and Cable, Inc., and CBS, Inc. (“appellants”) permission to copy audiotapes admitted into evidence at the trial of seven former Philadelphia police officers. The district court also denied appellants access to transcripts of the tape recordings that had been given to the jury. Applying the principles established in United States v. Criden, 648 F.2d 814 (3d Cir.1981), the district judge determined that he had “no reasonable alternative to this course of action.” Bench Opinion at 12A. For the reasons set forth below, we will reverse the order of the district court.
I.
On May 3, 1984 fifteen officers and former officers of the Philadelphia Police Department were indicted on charges stemming from an alleged city-wide practice of extorting protection money from operators of illegal video poker machines and “numbers” lotteries. The indictment charged each defendant with conspiracy, racketeering, and multiple counts of extortion. In addition, the indictment charged James Martin, then the Deputy Commissioner of the Philadelphia Police Department, with three counts of obstructing a criminal investigation through the payment of “hush money” to Albert Ricci, a Philadelphia police officer who was named as a defendant in an initial indictment handed down in March, but who later pleaded guilty and agreed to testify for the government. Martin and Joseph DePeri, a Chief Inspector at the time of the indictment, were also charged with obstructing justice by destroying records relating to payoffs made by video poker machine vendors. The trial of Martin, DePeri and five other defendants ended on August 10, 1984 with verdicts of guilty on all counts. The trial of the eight remaining defendants is scheduled to begin on November 5, 1984.
The government’s case, which included the testimony of Albert Ricci and numerous victims of the alleged extortion conspiracy, centered on former Lieutenant Joseph Alvaro. Alvaro resigned from the force and agreed to cooperate with the FBI in exchange for immunity after his name surfaced in the ongoing federal investigation of corruption in the Philadelphia Police Department. In addition to Alvaro’s in-court testimony, the jury heard tape recordings of face-to-face and telephone conversations between Alvaro and — principally—Martin, DePeri, and Ricci. While jurors were provided with headsets and transcripts to aid them in following the conversations, courtroom spectators could only listen to the tapes over the loudspeakers.1 Appellants, owners of two local newspapers and three broadcast stations, promptly moved for permission to copy the audiotapes and for access to the transcripts, for the purpose of disseminating these materials to the public. They now appeal from the district court's denial of their motions. Counsel for one of the remaining eight defendants entered an appearance as an appellee, but deferred to the Fraternal Order of Police, which filed a *967brief as amicus curiae. We granted appellants’ petition for expedited consideration.
II.
Appellants contend that the district court misapplied the principles of United States v. Criden, 648 F.2d 814 (3d Cir.1981) (hereinafter cited as Criden 7)2 in denying their request for access to the tape recordings and transcripts. Thus, our analysis must begin with a brief review of that case.
Criden I arose out of the FBI undercover operation popularly known as “Ab-scam”. George X. Schwartz, then President of the Philadelphia City Council, Harry P. Jannotti and Louis C. Johanson, Philadelphia City Council members, and Howard L. Criden, a Philadelphia attorney, were indicted on charges of bribery and related offenses. At the trial of Schwartz and Jannotti, which the district court had severed from the other cases, video and audiotapes made by the FBI were prominent items of proof. The district court released transcripts of the tapes, but denied the three major networks and a local broadcaster permission to copy the tapes for rebroadeast to the public.
On appeal we acknowledged that release of the tapes was a matter committed to the discretion of the district court, but we explained that the exact scope of review of a discretionary matter is dependent on “the reason why that category or type of decision is committed to the trial court’s discretion in the first instance.” 648 F.2d at 817. We went on to say:
In the matter before us on this appeal, the decision whether to release the tapes was not dependent in the main on particular observations of the trial court. Therefore, the trial court’s decision is not accorded the narrow review reserved for discretionary decisions based on firsthand observations, and we must consider both the relevance and weight of the factors considered____
We note that until or unless guiding rules have become fixed, it is important that the exercise of discretion be accompanied by the trial court’s articulation of the factors considered and the weight accorded to them,____
648 F.2d at 818-19.
Applying this standard of review, we reversed the order of the district court in Criden I. We found that the district court had given too little weight to the common law right of the public to inspect and copy judicial records, see Nixon v. Warner Communications, 435 U.S. 589, 98 S.Ct. 1306, 55 L.Ed.2d 570 (1978), and “the significant interest of the public in observation, participation, and comment on the trial events,” 648 F.2d at 823. We held that these two factors together created “a strong presumption that material introduced into evidence at trial should be made reasonably accessible in a manner suitable for copying and broader dissemination.”3 Id. Of the factors the district court cited as militating against release, we dismissed some as irrelevant or entitled to little weight,4 and concluded that others could be addressed by *968means short of withholding the tapes wholesale.5
Here the district court has stated three reasons for its decision to deny access: (1) the transcripts sought by appellants were not admitted into evidence (Bench Opinion at 4-5); (2) the public interest in these proceedings is not as significant as the interest in Abscam (Bench Opinion at 7-8); and (3) the “enhanced publicity” caused by release of the requested materials would likely make it impossible to impanel a fair and impartial jury for the trial of the eight remaining defendants (Bench Opinion at 11). As we did in Criden I, “[w]e must consider both the relevance and weight of the factors considered.” 648 F.2d at 818. We do so with the added guidance of the legal rules that emerged in Criden I.
1. Access to Materials Not Admitted Into Evidence
The first consideration the district court mentioned in denying appellants’ request was the fact that the transcripts provided to the jury were not admitted into evidence. “Thus,” the district court concluded, “any common law right to access to evidence introduced at trial should not apply to tape transcripts.” Bench Opinion at 5. We believe that the district court’s view of the common law right is too narrow, and that the strong presumption in favor of access established in Criden I applies to these transcripts.
It is true, as the district court noted, that our holding in Criden I extended only to “material introduced into evidence at trial____” 648 F.2d at 823. In that case, however, the district court did release transcripts of the video and audiotapes to the public, 648 F.2d at 822, and the appeal raised only the issue of permission to copy and rebroadcast the evidence itself.6 Thus, we had no occasion to consider whether the strong presumption in favor of access applied to judicial records and documents other than evidence. That issue is now squarely before us.
The strong presumption established in Criden I was based on two factors: the common law right of access to judicial records, and the public interest in “observation, participation, and comment____” 648 F.2d at 823. We find that both of these factors support a strong presumption in favor of access to these transcripts. The common law right of access is not limited to evidence, but rather encompasses all “judicial records and documents,” Warner Communications, 435 U.S. at 597, 98 S.Ct. at 1311-1312. It includes “transcripts, evidence, pleadings, and other materials submitted by litigants ____” Comment, All Courts Shall Be Open: The Public’s Right to View Judicial Proceedings, 52 Temple L.Q. 311, 337-38 (1979). The public interest in monitoring judicial proceedings7 also supports a presumption in favor of access. Although representatives of the media were present at the trial and were able to take notes on the recorded conversations as they were played to the jury, this procedure has obvious limitations. The public interest can best be vindicated by the release of complete and accurate transcriptions, at the expense of the media applicants. We therefore hold that the strong presumption in favor of public access ap*969plies to these transcripts. Cf Criden I, 648 F.2d at 828 (“It would unduly narrow the right of access were it to be confined to evidence properly admitted, since the right is based on the public’s interest in seeing and knowing the events which actually transpired.”).
In so holding we do not suggest that the fact that requested materials are not in evidence can never be a relevant consideration; we hold only that the district court erred in treating it as a dispositive consideration. Cf. In re Application of National Broadcasting Co., 653 F.2d 609, 614 (D.C.Cir.1981). Where proffered evidence is found inadmissible because it is unreliable, or because it is more prejudicial than probative, the dangers of broad dissemination may substantially outweigh any benefits. This is not such a case — the requested transcripts were deemed sufficiently reliable and helpful to be given to the jury. In these circumstances the fact that the transcripts are not in evidence carries little weight.
2. Extraordinary Public Interest
The district court distinguished Criden I on the basis that “while there is a legitimate public interest in the proceedings greater than that attendant to the usual criminal case, it does not rise to that interest surrounding Abscam.” Bench Opinion at 7-8. Although the extraordinary public interest in Abscam was noted in Criden I, 648 F.2d at 822, we do not believe that the court intended that this should be an important factor in the balancing process. Estimates of a case’s public significance will inevitably vary among segments of the community, not to mention from judge-to-judge. In any event, we do not find that the general public interest in these proceedings differs markedly from the public interest in the Criden proceedings.
In distinguishing Criden I, the district court noted three facts: that this, unlike Criden, is not a case of “national dimension;” that “elected members of Congress are not involved;” and that “the investigative techniques used by the Government are not extraordinary.” Bench Opinion at 7. As we read Criden I, however, that decision was not influenced by any “national dimension” or because members of Congress were involved. Indeed, although other Abscam prosecutions did involve members of the House and Senate, and did cut across several jurisdictions, Criden involved only the prosecution in Philadelphia of three Philadelphia City Councilmen and a local attorney. And though the FBI investigation conducted here was perhaps not as elaborate as Abscam, we believe that there are other factors present here that make this case of at least equal importance to the public. The integrity and efficiency of the police are matters that hit very close to home for most citizens. In this case police officers — including some of the highest-ranking in the city — with frontline law enforcement responsibility are accused of racketeering, extortion, conspiracy, and obstruction of justice. The fact that it is the federal authorities prosecuting these local officials is also a matter of legitimate public concern. Finally, we note that this is a rare case where some former police officers are cooperating with the government and testifying against fellow officers. In sum, we believe that this is a case of extraordinary moment, and that the presumption in favor of access to trial materials is no weaker here than it was in Criden I.
3. Effect on Subsequent Proceedings
The most important factor the district court considered, as was also true in Criden I, was the effect that release of the requested materials would have on subsequent related proceedings, and in particular on the prospects for impanelling a fair and impartial jury for the trial of the eight remaining defendants. For us this issue— the impact on subsequent proceedings — is by far the most difficult aspect of this case. For here we are all, whether as counsel, district court, or appellate judges, merely forecasting the uncertain future. We are looking at an opaque crystal ball, attempting to make predictions when we can never *970know what all the future variables will be as to how these tapes or transcripts will be used by various representatives of the media. It is even more difficult to forecast with certainty the impact of such additional publicity. Under some circumstances, these materials, if reprinted and rebroadcasted, may seem as stale and uninteresting as last week’s bread; yet, in other circumstances the materials can be reworked to bring a heightened sense of consciousness and drama to some of the events in this case.
We said in Criden I:
Considerations of the effect of publicity on a jury are ordinarily matters on which the trial court’s judgment is entitled to considerable deference by an appellate court. The trial court has already supervised the selection of one jury and any particular problems which the news coverage in this highly publicized matter created for the selection process would have come to the trial court’s attention. Therefore, if the trial court had already experienced difficulty in jury selection, we would be faced with an actual rather than conjectural factor militating against release of the tapes for rebroadcast. However, the trial court’s experience was the contrary____ Accordingly, the danger to defendants’ fair trial rights on a possible retrial is not based on the trial court’s experience in this case, but rather on its conjecture about possible future difficulties.
648 F.2d at 827. We concluded that the district court had given too little weight to the impact of skillfully conducted voir dire examination as an antidote for the effects of publicity. 648 F.2d at 828.
Here the district court encountered some difficulty in selecting the first jury. Voir dire, conducted by the court, took approximately four days. “A large proportion of the individuals questioned had been exposed to some type of publicity regarding this case and alleged police corruption. A not insubstantial number of those individuals stated they would have difficulty being fair and impartial because of their exposure to pre-trial publicity.” Bench Opinion at 9-10. Based on this experience, the district court concluded that the “enhanced publicity” that would result from release of the requested materials “is likely to make it impossible to obtain a fair and impartial jury” for the trial of the eight remaining defendants. Bench Opinion at 11.
The district court’s concern for the fair trial rights of defendants in subsequent proceedings is clearly entitled to much weight, and its experience in selecting the first jury shows that publicity in this case may have been unusually pervasive. Compare In re Application of National Broadcasting Co. (United States v. Myers), 635 F.2d 945, 953 (2d Cir.1980) (“[Ajbout half of those summoned for jury selection had no knowledge of Abscam, and only a handful had more than cursory knowledge.”); United States v. Haldeman, 559 F.2d 31, 62 n. 37 (D.C.Cir.1976) (in banc) (“Most of the venire simply did not pay an inordinate amount of attention to Watergate.”), cert. denied, 431 U.S. 933, 97 S.Ct. 2641, 53 L.Ed.2d 250 (1977). On the basis of this record, however, we cannot accept the district court’s conclusion that “enhanced” publicity is likely to make it impossible to impanel a fair and impartial jury for the upcoming trial if appellants’ requests are granted.
The requested materials can hardly be said to contain the sort of lurid or inflammatory matter usually present in cases where publicity has been found to work a denial of fair trial rights. See, e.g., Sheppard v. Maxwell, 384 U.S. 333, 86 S.Ct. 1507,16 L.Ed.2d 600 (1966); Irvin v. Dowd, 366 U.S. 717, 81 S.Ct. 1639, 6 L.Ed.2d 751 (1961). The tapes have already been played in open court, and have been widely reported in the media.8 None of the partic*971ipants in these conversations are defendants currently awaiting trial, and — with few isolated exceptions — the relevance of these recorded statements to the guilt or innocence of the remaining defendants is unclear.
Moreover, we find here a number of factors that we believe make this an even stronger case for releasing the requested materials than Criden I. There we reversed the district court for not permitting media representatives to copy the videotapes even though the trial judge had released transcripts to the press and public. In this case, with but one exception, the trial judge refused to release any of the transcripts to the press — even though the jury had been provided those transcripts. Our examination of the transcripts in Criden I and the transcripts in this case reveal no basis to distinguish between the two in terms of either significance or “titilation”. If it was appropriate to release the transcripts in Criden I, it was error to deny access in this case. Secondly, we must not forget that in Criden the tapes sought were videotapes and not merely audiotapes, as are involved in this case. In Criden, Judge John P. Fullam of the district court wrote:
The first point to be emphasized, in attempting to achieve a correct exercise of the Court’s discretion in this case, is the very great difference between videotape evidence and other forms of evidence. The hackneyed expression, “one *972picture is worth a thousand words” fails to convey adequately the comparison between the impact of the televised portrayal of actual events upon the viewer of the videotape and that of the spoken or written word upon the listener or reader. The viewer of videotape becomes virtually a participant in the events portrayed. Eyewitness testimony is often dramatic and convincing, but its effectiveness and convincing power are almost negligible in comparison with a film or videotape of actual events. When the videotape shows a crime actually being committed, it simply leaves nothing more to be said.
This is all to the good, when the videotape is presented in evidence in the course of a trial: The jury is plainly entitled to the best and most reliable evidence available, in its quest for the truth. If the truth is shocking, if the evidence is powerfully convincing, so be it; the accused has no justifiable complaint.
But dissemination of powerfully convincing evidence beyond the confines of the trial arena can, in some circumstances, cause serious and irreparable harm, both to persons whose interests are entitled to protection, and to the judicial process itself.
United States v. Criden, 501 F.Supp. 854, 859-60 (E.D.Pa.1980), rev’d, 648 F.2d 814 (3d Cir.1981).
Nowhere in Criden I did we reject or dispute the above specific findings of the trial judge as to the potential for a greater prejudicial impact with the use of videotapes as compared to audiotapes or transcripts. Indeed, we found that “[rjeplication of the trial court experience may contribute to the public’s understanding of the events which were the subject of the trial proceedings, and thereby enhance its ‘comprehension of the functioning of the entire criminal justice system ...’” 648 F.2d at 824 (quoting Nebraska Press Assn. v. Stuart, 427 U.S. 539, 587, 96 S.Ct. 2791, 2816, 49 L.Ed.2d 683 (1976)). If these “public forum values,” 648 F.2d at 822, served by release outweighed the potential prejudicial impact of the videotapes in Criden I, then — a fortiorari — the same values served by release of audiotapes and transcripts that replicate the trial court experience outweigh the lesser dangers created by their release.
Under these circumstances, we must deem the district court’s conclusion highly speculative. Though, in our constitutional scheme, “[n]o right ranks higher than the right of the accused to a fair trial,” Press-Enterprise Co. v. Superior Court, — U.S.-, 104 S.Ct. 819, 823, 78 L.Ed.2d 629 (1984), speculative threats to that right have never been sufficient to overcome either first amendment rights to attend and report on trials, Richmond Newspapers v. Virginia, 448 U.S. 555, 100 S.Ct. 2814, 65 L.Ed.2d 973 (1980); Nebraska Press Assn., supra, or the common law right of access to trial materials, Criden I, 648 F.2d at 826-828; Myers, 635 F.2d at 953.9
*973When rights such as these come in conflict, courts must maintain a delicate balance that will adequately preserve the protection afforded by the one without unduly impinging upon the freedom preserved by the other. The trial court bears the primary responsibility for striking the proper balance, and “must take strong measures to ensure that the balance is never weighed against the accused.” Sheppard, 384 U.S. at 362, 86 S.Ct. at 1522. In this regard we have stated our belief that “the appropriate course to follow when the spectre of prejudicial publicity is raised is not automatically to deny access but to rely primarily on the curative device of voir dire examination ____” Criden I, 648 F.2d at 827.
Since the inception of our criminal justice system, courts have acknowledged the utility of skillfully conducted voir dire as a means of ascertaining a prospective juror’s impartiality. Early in our history, Chief Justice Marshall wrote:
Were it possible to obtain a jury without any prepossessions whatever respecting the guilt or innocence of the accused, it would be extremely desirable to obtain such a jury; but this is perhaps impossible, and therefore will not be required. The opinion which has been avowed by the court is, that light impressions which may fairly be supposed to yield to the testimony that may be offered, which may leave the mind open to a fair consideration of that testimony, constitute no sufficient objection to a juror; but that those strong and deep impressions which will close the mind against the testimony that may be offered in opposition to them, which will combat that testimony, and resist its force, do constitute a sufficient objection to him. Those who try the impartiality of a juror ought to test him by this rule.
United States v. Burr, 25 F.Cas. 49, 50-51 (C.C.Va.1807) (No. 14, 692g). This “testing” by voir dire remains a preferred and effective means of determining a juror’s impartiality and assuring the accused of a fair trial.
The district court’s patient and careful examination of potential jurors in this case is an excellent example of the power of voir dire to screen out persons who may not be impartial. And though we fully appreciate the difficulties this delicate task presents, we also note that the four days consumed by voir dire in this case is by no means extraordinary for a highly publicized criminal case. Compare Press-Enterprise Co., — U.S. at-, 104 S.Ct. at 824 (voir dire in underlying criminal case consumed more than six weeks); Irvin, 366 U.S. at 720, 81 S.Ct. at 1641 (four weeks); People v. Manson, 71 Cal.App.3d 1, 139 Cal.Rptr. 275, 288 (Dist.Ct.App.1977) (24 days), cert. denied, 435 U.S. 953, 98 S.Ct. 1582, 55 L.Ed.2d 803 (1978). Conversely, in many cases surrounded by pervasive multimedia publicity, juries have been selected relatively quickly. Thus, even after film of the President of the United States being shot, and his alleged assailant being apprehended, was widely disseminated in the media, a jury for the attempted murder trial was selected in just five days. See N.Y. Times, May 4, 1982, at A22, col. 1 (trial of John W. Hinckley, Jr.). In the Watergate conspiracy case, after more than a year of sensational revelations in the media, a jury was selected after eight days of voir dire examination. Haldeman, 559 F.2d at 65.
Though we recognize the prospect that certain types of additional publicity may make voir dire somewhat more difficult in the upcoming trial, we cannot conclude that such publicity “is likely to make it impossible to obtain a fair and impartial jury” under the supervision of the thoughtful and sensitive trial judge. In all due respect, we think that the trial judge was far too modest in evaluating his proven skill to preside over effective voir dire proceed*974ings that would impanel a jury and assure a fair trial to all of the parties.
CONCLUSION
For the reasons set forth above, we will reverse the judgment of the district court and remand for proceedings consistent with this opinion.
. Because courtroom spectators had difficulty hearing the first tape played over the loudspeaker, the district judge permitted representatives of the media to examine a transcript of it during a luncheon recess.
. Criden I remanded the matter to the district court for determination of whether dissemination of the requested materials was likely to cause such serious harm to third parties as to merit excision of certain portions. On remand the district court deleted all references to third parties. The media applicants again appealed, and our decision ordering more selective excision is reported as United States v. Criden, 681 F.2d 919 (3d Cir.1982) (Criden II). Our dissenting colleague, Judge Sloviter, authored the opinions in both Criden I and Criden II.
. We did not consider in Criden I whether the public has any right to these materials based on the first amendment, 648 F.2d at 820, 821 n. 6, and in view of our disposition of this case on nonconstitutional grounds we will intimate no view on that question now.
. These factors were the special impact of videotape evidence, 648 F.2d at 823-24; the "enhanced punishment" the defendants would suffer as a result of rebroadcast, 648 F.2d at 824-26; concern that the requested tapes had been improperly admitted into evidence, 648 F.2d at 828; concern that some of the material would be highly prejudicial to a subsequent defendant though inadmissible at his trial, id,; and fear that release of videotapes, like televising the trial itself, could disrupt the trial process, 648 F.2d at 828-29.
. We held that the use of voir dire, rather than withholding the tapes, would adequately protect the fair trial rights of defendants in subsequent related proceedings, 648 F.2d at 826-28, and that the privacy interests of third parties could be protected by selective excision of material likely to cause third parties severe harm, 648 F.2d at 829.
. Indeed, this appears to be true generally in cases involving requests by the media to copy video or audiotapes used as evidence. See, e.g., Nixon v. Warner Communications, 435 U.S. 589, 594, 98 S.Ct. 1306, 1310, 55 L.Ed.2d 570 (1978); United States v. Edwards, 672 F.2d 1289, 1291 (7th Cir.1982); Belo Broadcasting Corp. v. Clark, 654 F.2d 423, 427 (5th Cir.1981); In re Application of National Broadcasting Co., 653 F.2d 609, 611 (D.C.Cir.1981); In re Application of National Broadcasting Co. (United States v. Myers), 635 F.2d 945, 952 (2d Cir.1980). See generally Note, Access to Taped Evidence: Bringing the Picture into Focus, 71 Geo.L.J. 193 (1982).
. We recently discussed this interest in detail in the context of a civil proceeding. See Publicker Indus, v. Cohen, 733 F.2d 1059 (3d Cir.1984).
. For example, we take notice that the following account of one recorded conversation appeared in a leading Philadelphia newspaper while the first trial was in progress:
*971In the March 7 conversation played for the jury yesterday, Martin and Alvaro knew that the indictments were imminent and that Ricci, who had been a vice officer under Martin and Alvaro in the northwest police division, would be indicted. Ricci testified last week that he had handed Martin protection payoffs from gambling operations totaling $10,000 a month for seven months.
On the tape, Martin asked Alvaro what he thought Ricci’s legal fees would come to. "Joe, let’s say, what could it cost, 50?” he asked Alvaro.
"Yeah,” the federal informant responded.
"OK, so we’ll have to raise $50,000, that’s all,” Martin replied. "I ain’t gonna sit back and let the guy get it. If he’s gonna go to bat for me, got to go all out for him. There’s nothing I can do about it. ... It shouldn’t be no problem, to be honest with you, in a year, to raise just perhaps 60, 70 thousand. ... I’m totally depending on him to keep his mouth shut."
"I think you can count on him not to say anything,” Alvaro said.
“Well, you, he can count on me,” Martin said. "It's going to be an upfront thing. That way you don’t have to worry about trust.”
Ricci testified last week that, in February, March and April, Martin paid him a total of $2,000, and that Martin promised him $500 a week for his silence.
Later in the conversation, according to Alvaro’s testimony, Martin talked about the direction the federal investigation was taking. Martin specifically discussed a lucrative source of extortion money — "lead houses,” establishments where numbers bets are placed. Alvaro said Martin had told him that he was reluctant to continue presiding over the extortion but that he felt compelled to, if only to generate money for Ricci.
"I have to, you know, take the chance, that’s all,” Martin said on the tape. "I hate doing it and I’m trying to eliminate it as much as possible. ...
"What am I going to do, Joe?” the deputy police commissioner asked. “You know what I’m worried about? I’m worried about them lead houses. Sambor’s [Police Commissioner Gregore Sambor] getting all kinds of rumors____ I’ve got so much [obscenity] on my mind. You know, thinking all the time, all the time____ How the hell are we going to get them bastards away from here? If they don’t leave us alone, finish it up and get it over with, you know, this could go one for four [obscenity] years ... You know, I feel so helpless. How do you stop this [obscenity] thing? Got any ideas?”
“You don't, Jim,” Alvaro said. “There’s nothing you can do. No way you can stop it.”
Alvaro then suggested a crackdown on numbers operations for appearances’ sake, saying the Major Investigations Division must have other sources of extortion money.
"Jim, that can't be the only source,” he said.
“It isn’t, but it’s gonna create a lot of hollering and screaming,” Martin replied. "You ever tell a guy that was making $100,000 a month to stop?”
A few minutes later on the tape, Martin mused about the possibility that he could have become police commissioner.
"I’m a little sorry I didn’t take that number-one job when I had the shot."
"Why?” Alvaro asked.
"See, I could control what’s happening____
I would have said [obscenity] you, you know.” Philadelphia Inquirer, July 25, 1984, at 6-A, cols. 1-3 (omissions and bracketed material in the original).
. The dissent seeks to distinguish Criden I on three bases. First, according to the dissent, the possibility of a subsequent trial in Criden I was remote and therefore militated against a denial of access, whereas here we have “the actuality of a trial already scheduled to begin shortly.” Dissent at 975. We, however, do not read Criden I as treating this uncertainty as a critical factor. Rather, we said “there remains the possibility that Schwartz and Jannotti may have to be tried again in this district, and the effect of rebroadcast of the audio and videotapes on a possible future jury cannot be summarily dismissed.” 648 F.2d at 827.
Similarly, we find unpersuasive the dissent's second suggested distinction: that the district judge in this case, because of the difficulties in selecting the first jury, was better situated to predict the effects of "enhanced publicity" than was the trial judge in Criden. We do not see how the trial judge in Criden had any less of an opportunity to experience “the atmosphere of the courtroom and the demeanor of the jurors, prospective jurors, and witnesses____” Dissent at 976.
Finally, the dissent distinguishes Criden I on the ground that "here the district court did not deny access but merely postponed it.” Dissent at 976. Far from distinguishing Criden /, this appears to repudiate our statement that there are "certain practical considerations that must be taken into account before finding that the possible effect of such publicity in a possible *973retrial warrants denial of access to evidence already introduced. Effectively, this would re-suit in denial of the right to copy at a time when the issues remained a matter of public interest. Thus the educational and informational value of public observation of the evidence would never be available at a meaningful time.” 648 F.2d at 827.