Statement of reasons for voting to deny the government’s petition for rehearing en banc.
As prelude, I observe that issuance of this decision was in violation of this court’s Internal Operating Procedures. The motions division had issued an unpublished Memorandum Opinion and Judgment which effectively functioned as a Writ of Mandamus. Thereafter, the division decided to publish the decision. Accordingly, it was circulated to the nonparticipating judges as required by Part VIII E of our I.O.P.’s. The release of this unpublished decision, however, was not in accordance with Part VIII D of those procedures, which permits a Memorandum Opinion and Judgment to be released only
when a division unanimously determines that a judgment of the trial court ... should be affirmed or enforced, and this court’s decision ... (3) does not alter [or] modify ... an existing rule of law; ... and (6) does not find any error of law. (Emphasis added.)1
Because an error of law was found, i.e., restriction on cross-examination at the pre*218liminary hearing and refusal to hear a witness called by the accused, there has not been compliance with Part VIII D. Additionally, in this decision, the motions division applies diluted precedent (if precedent at all2) which, in one highly important aspect, is inapposite to the present preliminary hearing rule. The present rule is one markedly different from the rule which existed when those cases cited as precedent were decided. Because these flaws so burden this decision and because it has been accurately characterized as a narrow, fact-focused decision with no significant impact on the administration of justice, I deem it unworthy of en banc rehearing. On the other hand, should this decision produce fallout, another case may provide the appropriate opportunity for further review. A close examination of this decision reveals its flaws and narrow scope.
I. Grand Jury Interference
My major criticism of this decision is its intrusion into the orderly method of Grand Jury process. It applies extraordinary process in an undeserving case.
In Washington v. Clemmer, 119 U.S.App.D.C. 216, 339 F.2d 715 (1964), an injunction — in the nature of a writ of prohibition — did issue against Grand Jury action to prevent mooting the ease in the court of appeals. For that same reason, the division here initially prohibited any Grand Jury process. In its decision, however, that prohibition has been modified, at government request, but only to permit evidence to be presented to the Grand Jury. The Grand Jury still may not return an indictment or decline to do so until the preliminary hearing is completed. This is an unwarranted and unjustifiable intrusion by this court, which — on the other hand— secures little or no legitimate benefit to the accused. See Circuit Judge Danaher’s separate statement in Washington v. Clemmer. Id. at 221, 339 F.2d at 720. Then Circuit Judge Burger’s separate statement in that case, id. at 225, 339 F.2d at 724, is most apt here:
A Grand Jury is an independent body and neither this court nor any other court ought to interfere gratuitously with its deliberations, especially for the sole purpose of preventing mootness of an issue because some members of the court were seemingly anxious to reach it.
Id. n. 1.
It should be noted that the division is unable to justify this intrusion and makes no attempt to do so. Under the All Writs Act, 28 U.S.C. § 1651 (1982), the division’s action is hardly “agreeable to the usages and principles of law.” Id. Both Fed.R. Crim.P. 5(c) and Super.Ct.Crim.R. 5(d)(2) require prompt preliminary hearing. Those rules are grounded on expressed public policy that there is a “public interest in prompt disposition of criminal cases.” The Grand Jury should never have been gagged, as the division ordered, over three months ago.
II. The Predicate “Error”
Furthermore, the underlying predicate of the division’s order was erroneous. I quarrel with the holding that cross-examination was erroneously limited on the narrow issue of probable cause and that the accused had a right in this case to call the witness on that same issue. The opinion’s passing reference in its fourth paragraph to the possibility that the effect of the proffer was to “cast a cloud on Hoffman’s reliability as an informant” is not to be read as anything more than an observation or characterization of the potential but unidentified proof. As footnote 3 makes clear,3 the inquiry at the preliminary hearing is limited to probable cause, not general credibility *219or “other matters.” The division, having chosen to omit reference to the first sentence of Super.Ct.Crim.R. 5(d) (negating discovery as a purpose for preliminary hearings) cannot and does not purport to have affected that proscription. Thus, complete deference to that provision is still to be expected at preliminary hearings so that they do not become deposition and discovery proceedings.
Moreover, the “cloud” that the division perceived hovering over the finding of probable cause is non-existent. The proffer as to the eyewitness’ testimony was that she had been the focus of a love triangle between the accused and the decedent for some time. Though she had lived in a “common law” relation with the accused, she was living with the decedent when the accused shot him. On numerous occasions that witness had made criminal charges against the accused, but had dropped them. It was also proffered that the witness would state that the decedent had in the past threatened the accused’s life.
The latter is only relevant to an affirmative defense not available to the accused at this stage and not even raised by him. Thus, that evidence is now quite irrelevant. The other proffered facts have no diluting or negative impact on probable cause. Indeed, they enhance the showing of probable cause for they establish identity of the accused and motive for the killing. Whether there is a reasonable doubt as to the witness’ credibility must await another day. The proffer is inadequate to cast any cloud on the finding of probable cause at Poteat’s preliminary hearing. In re R.D.S., 359 A.2d 136, 139-40 (D.C.1976).4
III. The Decision Is Not Precedent
There are a number of reasons why this decision is so infirm as to be of no future authority. First, it fails, in its original form, even to mention Super.Ct.Crim.R. 5(d). Even with the addition of footnote 3, the decision still fails to take proper account of that rule. The rule — an approved modification of Fed.R.Crim.P. 5 — specifically provides that “[t]he purpose of a preliminary examination is not for discovery.” This provision was adopted for the express purpose of preventing a preliminary hearing from degenerating into a deposition and discovery exercise and to confine the inquiry to the narrow question of probable cause. In any event, since the reopened hearing in this case must be so narrowly confined to probable cause and exclude all “other matters,” it will, in the long run, be a shallow exercise unworthy of extraordinary writ remedy.
Discovery was perceived to be a primary purpose for expanding the preliminary hearings in the twenty-year-old cases relied upon by the division. Therefore, subsequent rule changes in both the federal and Superior Court rules (Rules 5 and 5.1 and Rule 5(d) and (e), respectively) have significantly limited judicial authority to tinker with preliminary hearings. Both the federal rule and the Superior Court's rules now provide that if the defendant is indicted “the preliminary examination shall not be held.” (Emphasis added.) See Fed.R. Crim.P. 5(c) and Super.Ct.Crim.R. 5(d)(2). Thus, the purpose of the rule as it now exists is to moot arguable defects in Rule 5 proceedings by prompt indictment and to overrule Blue v. United States, 119 U.S.App.D.C. 315, 342 F.2d 894 (1964) and Ross v. Sirica, supra, both post-indictment decisions relied upon by the division. Those decisions have, in any event, been overruled by Gerstein v. Pugh, 420 U.S. 103, 119, 95 S.Ct. 854, 865, 43 L.Ed.2d 54 (1975) (conviction is not to be vacated for preliminary hearing defects).
Moreover, the irregularity in promulgation of this decision strips it of precedential value. Where a proposed opinion is circulated to nondivision judges before release, an opportunity for their input is provided. *220This not only produces collegial decisions, but often great changes in proposed opinions. The division here inadvertently closed its eyes to this salutary process and now stands by its flawed decision. Our internal circulation rule is a prerequisite to decisions being accorded full status as precedent (see M.A.P. v. Ryan, 285 A.2d 310 (D.C.1971)). Because of the violation of that rule, plus the division opinion’s utter failure to take account of highly relevant rules changes and full binding precedent (e.g., M.A.P. v. Ryan,5 supra, and United States v. Davis, 330 A.2d 751 (D.C.1975)), it is not entitled to any precedential weight.
In addition, All Writs relief is always a matter of discretion. Schlagenhauf v. Holder, 379 U.S. 104, 85 S.Ct. 234, 13 L.Ed.2d 152 (1964). Thus, one division of the court is not bound to grant that relief simply because another division saw fit to do so in an earlier case.
Both the United States and the District of Columbia, which filed an amicus memorandum in support of the en banc petition, view the unpublished decision of the division as having no precedential effect. But it is not altogether certain anyone had expected the division to publish it. Indeed, the Public Defender Service, which filed an amicus memorandum opposing en banc reconsideration, takes the view that this decision is a “narrow, fact-focused holding,” with “no important policy considerations or systematic interests” warranting “further evaluation.” (Memorandum of Public Defender Service at 1). I accept their position that this decision is of no importance in the administration of justice.6 My colleagues seem to have done so for they deny rehearing en banc consistent with that significant concession by the Public Defender Service.
If, on the other hand, this court is in the future to interject itself into the Grand Jury function by reviewing preliminary hearings before that body is permitted to act, Superior Court may rightly seek to do away with such hearings completely. Gerstein v. Pugh, supra, 420 U.S. at 120-22, 95 S.Ct. at 866-67, permits and recognizes. the efficiency in such a step. Alternatively, such tampering with Grand Jury process should become the focus of Supreme Court review.
. No one contends that the Internal Operating Procedures do not apply to our original All Writs Act proceedings because of their literal application to a trial court “judgment." By the very nature of such extraordinary proceedings application of our I.O.P.’s is compelled.
. Ross v. Sirica, 127 U.S.App.D.C. 10, 380 F.2d 557 (1967), is of very doubtful value as precedent since six judges of the court did not agree with its holding. See then Circuit Judge Burger’s separate statement, id. at 22, 380 F.2d at 569, noting “an accurate ‘head count’”; Washington v. Clemmer, 119 U.S.App.D.C. 216, 339 F.2d 715 (1964) suffers from the shadow of infirmity cast by Ross three years later.
. Footnote 3 was added before publication but after the decision was issued.
. We are not told by the division how the proffer is relevant to the concededly narrow issue of probable cause. Indeed, as I have observed, the division does not even identify the evidence. To do so would require the impossible — the division would have to demonstrate its relevance.
. Ironically, the M.A.P. decision was itself a probable cause hearing case on the merits. We refused to follow the federal circuit court’s holding in Brown v. Fauntleroy, 143 U.S.App.D.C. 116, 442 F.2d 838 (1971), that a juvenile was entitled to a probable cause hearing before trial when he was released to family custody.
. Presumably, this position was taken after institutional deliberation and recognition that a shift of position would be most unseemly.