This is an appeal from the denial of a preliminary injunction in a suit under the First Amendment and the Sherman Act (15 U.S.C. Secs. 1 et seq.) to prevent the City of Indianapolis (which is coterminous with Marion County, Indiana, and thus extends beyond the old city limits to take in the surrounding suburbs) from enforcing the provisions of its cable television ordinance that relate to the award of cable television franchises.

Enacted in November 1979 as chapter 8 1/2 of the Code of Indianapolis and Marion County, the ordinance provides that any cable television system using any of the public ways of the City of Indianapolis must obtain a franchise from the City. The ordinance creates the following procedure for awarding such franchises. Applications are accepted only "following action by the [City-County] Council determining that a franchise should be granted for all or a portion of the City," and must be filed within a specified time after the Council's action. The application, which must be accompanied by a nonrefundable fee of $3,000, is referred to the Cable Television Committee of the Council for at least two days of hearings, and it then goes, along with the Committee's evaluation, to the Board of Public Works. The Board "may recommend a franchising contract with the applicant whose application represents the most desirable of all applications submitted for each area of the City," but the contract must not be exclusive, that is, must not forbid the City to franchise other cable television systems that would compete with an existing franchisee.

In determining which is "the most desirable" of all the applications, the Cable Television Committee and the Board must consider "all factors normally considered in any case in which the Committee or Board must make such a determination," including various technical and financial factors, the applicant's reputability, the quality of his promised service, and "any special factors ensuring that the applicant will carry out the purposes of this chapter and that an award of the franchise to the applicant is in the best interest of the City." Those purposes are elsewhere described as "orderly and responsible development of a system which will provide the people of the City with cable television service which is versatile, reliable, and efficient and which is available at affordable rates."

The Board's recommendation for the grant of a franchise goes to the City-County Council to be confirmed or denied by ordinance. If the franchise is confirmed, the franchisee must pay a fee of $25,000. Applicants rejected by the Board may appeal to the Council, which may direct the Board to reconsider its action.

In 1967, when Marion County was still an autonomous political subdivision, its governing body had awarded a cable television franchise (later assigned to the present franchisee, Indianapolis Cablevision, Inc.) for the unincorporated suburban areas outside what were then the city limits of Indianapolis. Since this franchise is not exclusive, it does not prevent the City of Indianapolis from franchising other cable television systems in those areas. But when, shortly after chapter 8 1/2 was passed, the City-County Council exercised its power to "open up" areas of the City of Indianapolis for franchising, it did so only for the old city and the incorporated towns outside it; the area served by Indianapolis Cablevision was excluded. Four companies applied for franchises in the areas that were opened up, but only one franchise was awarded, to American Cablevision of Indianapolis, Inc. Since its franchise covered the entire area not served by Indianapolis Cablevision, this meant that neither company faced competition from another cable television system. The City-County Council confirmed the award of the franchise to American Cablevision in February 1981.

Enter the plaintiff, Omega Satellite Products Company. A limited partnership that has operated in Indianapolis since May 1980, Omega resembles and competes with cable television systems but at first was not subject to chapter 8 1/2 because it did not use any public way. A conventional cable television system receives television signals at a satellite earth station located some distance from its subscribers and transmits the signals to them over a grid of cables either strung on telephone poles or run under the streets. Omega, in contrast, installs small satellite earth stations on the roofs of apartment house complexes where its subscribers live, which means that all of the cables that connect the earth station to the subscribers' television sets are located on private property rather than above or below the public streets. Omega provides the usual cable television mixture of local television stations, distant "superstations" such as WGN in Chicago, cable television networks for sports and other specialized programming, and a modest amount of original programming. The corporation that is the general partner in Omega has interests in conventional cable television systems elsewhere in the country, but Omega itself has never operated such a system, and although it was operating in Indianapolis during the period in which applications for cable television franchises were being considered under the new ordinance neither it nor any of its sister systems expressed an interest in getting a franchise.

Not till March 1981, the month after the Council had awarded a franchise to American Cablevision, did Omega begin to express such an interest. It wrote letters to the City's general counsel, the chairman of the City-County Council, and other officials inquiring whether it might get a franchise. It received no encouragement. It could not actually apply for a franchise; the time for applying ran from the date on which the Council had opened up a part of the City of Indianapolis for franchises, and had expired long ago. But Omega made no effort, beyond perfunctory correspondence, to persuade the Council to open up all or part of the City of Indianapolis for new franchises, though chapter 8 1/2 appeared to give the Council the power to do so--and if it did not, the Council could easily rectify the oversight by passing an amending ordinance, as it was later to do.

Omega became impatient and in May 1981, without getting the required permit from the City's Department of Transportation to use a public right of way, it ran a cable through a drainage culvert under West 38th Street in Indianapolis to connect two apartment complexes so that it would not have to put a separate earth station on the second. It took the law into its own hands because it knew that the Department of Transportation would not grant it a permit to use the culvert. The Department may not allow the public ways of Indianapolis to be used in violation of an ordinance, and Omega's use of the drainage culvert under West 38th Street was contrary to chapter 8 1/2 because by using a public way in this manner Omega became a cable television system within the meaning of the ordinance, yet had no franchise. The City argues that Omega's lawless behavior is sufficient reason to deny the preliminary injunction that it seeks. Of course Omega did not have to obey chapter 8 1/2 if it violates federal law, but the Department of Transportation might have denied a permit for independent and valid reasons, such as safety. While not happy with Omega's surreptitious violation of the permit requirement, we do not think it bears on whether a preliminary injunction should have been issued. Evidently the cable does not create any safety hazard or other problem apart from chapter 8 1/2 because the City has consented to the cable's being allowed to remain in place until this lawsuit is finally decided, which we take to be a concession that the only ground on which the Department could have denied a permit was chapter 8 1/2. In any event, it is not the proper office of the federal courts to punish Omega for its failure to get a permit. That is for the City to do if it wants to; evidently it does not want to.

The City discovered the illegal cable in a routine inspection of public ways and in January and February 1982 wrote Omega ordering it to remove the cable but also advising it that it could apply for a cable television franchise under chapter 8 1/2. The legal basis for this invitation is a little unclear but is apparently an ordinance, passed in July 1981, which amends chapter 8 1/2 by setting up a new unit of local government, the Cable Communications Office, and inviting it to refer to either of two committees of the City-County Council "specific applications for the granting or renewal of a franchise."

Omega neither filed an application for a franchise nor removed the cable but instead brought this lawsuit on March 11, 1982, and moved for a preliminary injunction forbidding the City to remove Omega's cable or to enforce the limitations in chapter 8 1/2 on the award of new franchises. Two weeks later, after an evidentiary hearing, the district court denied the motion. Trial was originally scheduled to begin on September 7, but by agreement of counsel has been postponed to March of next year. The City has agreed not to disturb the cable under West 38th Street until a final judgment is entered.

The decision to grant or deny a preliminary injunction involves a comparison of the probabilities, and consequences (public as well as private), of two types of error: granting an injunction to an undeserving plaintiff, that is, one who will not be able to establish a legal right to an injunction when the case is tried in full rather than in the necessarily hasty and incomplete hearing on the motion for preliminary injunction; and denying an injunction to a deserving plaintiff. See 11 Wright & Miller, Federal Practice and Procedure Sec. 2948 (1973). When the consequences and probabilities point in the same direction the decision is easy. If the plaintiff is more likely than the defendant to prevail in the full trial and would be harmed more by the denial of the preliminary injunction than the defendant would be harmed by its grant, clearly the preliminary injunction should be granted. If the defendant is more likely to prevail in the full trial and would be harmed more by the grant of the preliminary injunction than the plaintiff would be harmed by its denial, clearly it should be denied. But the two cases are not exactly parallel, and not only because the plaintiff has the burden of proof. In measuring the harm to the plaintiff if the preliminary injunction is denied the court must consider whether the harm can be repaired completely by an award of damages after trial--in which event even a deserving plaintiff will not be hurt by being denied a preliminary injunction.

If the consequences and probabilities do not point in the same direction, the analysis is more complex. If the harm to the plaintiff from denial of the preliminary injunction would be very great and the harm to the defendant from granting it very small, then the injunction should be granted even if the defendant has a better chance of prevailing on the merits than the plaintiff, provided the plaintiff's chances are better than negligible; and vice versa. "In general, the likelihood of success that need be shown will vary inversely with the degree of injury the plaintiff will suffer absent an injunction." Roth v. Bank of the Commonwealth, 583 F.2d 527, 538 (6th Cir.1978); see 11 Wright & Miller, supra, Sec. 2948 at pp. 453-55.

To decide where in this matrix the present case falls requires us to consider and if possible weigh both the consequences and the probabilities. We begin with the consequences. Omega argues plausibly that the closer the two existing franchisees come to completing their cable grids throughout their respective service areas the more difficult it will be for a new cable television system to attract subscribers. Once a system's grid is complete, the cost of serving an additional subscriber may be much lower than the cost to a new system of building out its grid to serve that subscriber. Such a cost advantage would translate into a competitive advantage for the established firm, enabling it to charge the new subscriber a remunerative price lower than a new competitor's cost of serving that subscriber.

Indianapolis Cablevision has been constructing its grid in the suburbs for many years and it is hard to believe that a few months are going to make a critical difference in the ability of a new system to compete with it. But American Cablevision has, ever since it received its franchise in February 1981, been moving rapidly to complete its grid, and every month that passes diminishes Omega's chances of mounting an effective competitive attack upon American. Since the effect of the passage of time on Omega's prospects is difficult to calculate, an award of damages after a full trial might not compensate it fully for the harm caused to it by delay. But this type of damage estimation problem is endemic in antitrust litigation; and while it involves an undoubted element of uncertainty which granting a preliminary injunction would avoid, we must also consider the uncertainty whether Omega is actually being delayed by not having a preliminary injunction.

Except for the cable under West 38th Street, Omega has never been in the cable television business as conventionally understood; its business, which chapter 8 1/2 does not touch, involves running cables inside a building or building complex from a satellite earth station on the roof. True, it has sister companies engaged in the cable television business, but they have shown no interest in that business in Indianapolis. Omega was active in Indianapolis in 1980, when the franchise award process was in full swing, but neither it nor any of its sisters joined the bidding. When Omega did become interested, after the award of a franchise to American Cablevision, its efforts to obtain a franchise were half-hearted. It never formulated and submitted to the City-County Council a detailed plan for a cable television system that might have induced the Council to reopen the franchise award process, as the Council easily could have done since neither of the franchisees had received exclusive franchises.

If there is thus some doubt whether Omega's aspirations for entering the cable television business have ever gone much beyond saving the cost of one satellite earth station by laying a cable under West 38th Street, there is even greater doubt whether it would enter the business full tilt on the strength of a preliminary injunction. Only the boldest of entrepreneurs would commit millions of dollars on the strength of a preliminary injunction to an investment that might be wiped out completely if a permanent injunction never issued.

Since Omega is unlikely to make serious efforts to enter the cable television business in Indianapolis, at least until the merits of its attack on the City's ordinance are determined after a full trial, the denial of a preliminary injunction will not harm it beyond depriving it of a strategic weapon to use in bargaining with the City. True, by the same token the grant of a preliminary injunction might not do much harm to the City either, even though it would make it impossible for the City to prevent the construction of new cable television systems until it passed a new ordinance, which it might be reluctant to do until the case was finally resolved after trial and maybe an appeal. If Omega would not build a cable television system in reliance on a preliminary injunction, probably no other company would. In that event the City would not be harmed by the grant of a preliminary injunction. But the fact that the grant of a preliminary injunction would not hurt the defendant would not justify granting it if its denial would not hurt the plaintiff. The plaintiff has the burden of proof; he must show some need for the injunction before it will be granted. The symmetry is imperfect for another reason. We are pretty confident that Omega will not begin to wire up Indianapolis on the strength of a mere preliminary injunction, and thus that it will suffer no harm from denial of an injunction. We are less confident that no other cable television entrepreneur would be bold enough to do so, and if any did then the injunction would cause harm to the City.

We could stop with this observation and affirm the district court's denial of the preliminary injunction without any discussion of the merits were it not for two considerations. The first is the desirability of providing the district court with some guidance for the conduct of the trial, a factor stressed in Dos Santos v. Columbus-Cuneo-Cabrini Medical Center, 684 F.2d 1346, 1352 (7th Cir.1982). The second is the following relationship between the merits and the question of injury: if Omega has an overwhelming probability of winning this case when it is finally tried, and on grounds that would make clear its entitlement to a franchise, and we make all this clear in issuing the preliminary injunction that Omega seeks, Omega may be emboldened to go ahead and build its system on the strength of that injunction. This assumes, of course, what is uncertain, that Omega really has an interest in building a cable television system. But we cannot say for sure that it does not; so if there really is an overwhelming probability of its eventually winning this case and getting a franchise, that would argue for granting a preliminary injunction now. Therefore we will examine, though tentatively, the merits.

Omega does not have an overwhelming probability of winning on the Sherman Act count in the complaint. This is not because the City of Indianapolis surely is immune from the Sherman Act, as the City argues. Community Communications Co. v. City of Boulder, --- U.S. ----, 102 S. Ct. 835, 70 L. Ed. 2d 810 (1982), holds that the acts of a municipality are not immune from attack under the federal antitrust laws merely because the state has given its municipalities "the full right of self-government in both local and municipal matters," 102 S. Ct. at 836-37 n. 1; their acts are immune only if "undertaken pursuant to a clearly articulated and affirmatively expressed state policy," 102 S. Ct. at 842. Omega alleges a tacit agreement between the City and its two franchisees to exclude competing cable television systems from their areas, and nothing in any Indiana statute or other source of state policy suggests that the state actively desires local monopolies in the cable television industry.

But the statutory authority under which the City of Indianapolis acted in promulgating its cable television ordinance is considerably more specific than the "home rule" statute in Boulder. The City of Indianapolis is (or was at the time it acted) empowered to "regulate, inspect, license and prohibit services, advantages and property furnished directly to the homes of the general public which shall include ... television signals," Ind.Code Sec. 18-1-1.5-13(b), and to grant franchises to users of public ways, Ind.Code Sec. 18-1-21-5. Although the content of the regulation authorized by these sections is not prescribed, section 18-1-21-5, in confirming the municipality's powers and duties with regard to the safety, welfare, and convenience of its residents--including the power to set the terms of any franchise--could be thought to have approved traditional practices in public utility franchising.

True, the traditional presumption is against a municipality's having power to grant an exclusive license. See, e.g., Colen v. Sunhaven Homes, Inc., 98 So. 2d 501, 503-04 (Fla.1957). But in Indiana, though the early case of Citizens Natural Gas & Mining Co. v. Town of Elwood, 114 Ind. 332, 16 N.E. 624 (1888), adopted the presumption, a later case, State ex rel. Evansville Independent Tel. Co. v. Stickelman, 182 Ind. 102, 107, 105 N.E. 777, 779 (1914), emphasized the broad discretion of the municipal authorities over the licensing of the public ways and noted: "In the very nature of things there must be some limit to the number of telephone and light companies that can serve the public by the use of the streets of a town or a city for the location of poles, wires, and necessary appliances. If, as appellant contends, the municipal authorities must grant every applicant for a telephone or electric light franchise the right to use the streets on the same terms accorded corporations now using the same, such streets might be rendered of little use for other purposes." Id.; cf. Farmers' & Merchants' Co-op. Tel. Co. v. Boswell Tel. Co., 187 Ind. 371, 385-86, 119 N.E. 513, 517 (1918). This comes pretty close to saying that municipalities in Indiana are authorized to grant exclusive franchises. But it may fall short of the Supreme Court's requirement of "a clearly articulated and affirmatively expressed state policy" in favor of the specific restriction that is challenged. This issue will require further exploration in the district court.

Even if the City of Indianapolis is not immune from the Sherman Act, and even if it has as alleged tacitly agreed to give its franchisees exclusive territories--and although there is no evidence in the record of the preliminary injunction hearing of such an agreement we agree with Omega that the district court unduly hindered its efforts to obtain evidence of it through the discovery process--this would not necessarily prove a violation of the Sherman Act, though it might prove a violation of the ordinance. Competitors who agree to keep out of each other's sales territories commit a per se violation of the Sherman Act, but "vertically" imposed exclusive territories are illegal under the Sherman Act only if unreasonable, Valley Liquors, Inc. v. Renfield Importers, Ltd., 678 F.2d 742 (7th Cir.1982), and if there was such an agreement here it was vertical. The City of Indianapolis is not a competitor of Indianapolis Cablevision or American Cablevision but a representative of the potential customers of these companies. It has an interest in getting for its residents, the prospective consumers of cable television, the best service at the lowest price, though maybe not an undivided interest--some cities see cable television as a source of tax revenues as well as consumer benefits. Cf. City of Peoria v. General Electric Cablevision Corp., 690 F.2d 116 (7th Cir.1982). But we cannot, on the facts before us, assume that the City of Indianapolis is indifferent to consumer welfare, or dismiss the hypothesis that one way to promote that welfare might be--we do not say it would be--to give each franchisee an exclusive territory.

The cost of the cable grid appears to be the biggest cost of a cable television system and to be largely invariant to the number of subscribers the system has. We said earlier that once the grid is in place--once every major street has a cable running above or below it that can be hooked up to the individual residences along the street--the cost of adding another subscriber probably is small. If so, the average cost of cable television would be minimized by having a single company in any given geographical area; for if there is more than one company and therefore more than one grid, the cost of each grid will be spread over a smaller number of subscribers, and the average cost per subscriber, and hence price, will be higher.

If the foregoing accurately describes conditions in Indianapolis--again a question on which the record of the preliminary injunction proceeding is sketchy at best--it describes what economists call a "natural monopoly," wherein the benefits, and indeed the very possibility, of competition are limited. You can start with a competitive free-for-all--different cable television systems frantically building out their grids and signing up subscribers in an effort to bring down their average costs faster than their rivals--but eventually there will be only a single company, because until a company serves the whole market it will have an incentive to keep expanding in order to lower its average costs. In the interim there may be wasteful duplication of facilities. This duplication may lead not only to higher prices to cable television subscribers, at least in the short run, but also to higher costs to other users of the public ways, who must compete with the cable television companies for access to them. An alternative procedure is to pick the most efficient competitor at the outset, give him a monopoly, and extract from him in exchange a commitment to provide reasonable service at reasonable rates. See section 8 1/2-61 of the Indianapolis Code. In essence Omega's antitrust allegations accuse the City of Indianapolis of having taken this alternative route to the monopoly that may be the inevitable destination to which all routes converge.

Although there is language in some Supreme Court opinions, notably National Society of Professional Engineers v. United States, 435 U.S. 679, 688, 690, 98 S. Ct. 1355, 1364, 55 L. Ed. 2d 637 (1978), to the effect that the only thing to consider in deciding whether a practice violates the Sherman Act is the effect on competition, it is unlikely that the Court meant to overturn the established proposition that the antitrust laws do not require the impossible--a competitive market under conditions of natural monopoly. See, e.g., Hecht v. Pro-Football, Inc., 570 F.2d 982, 990-91 (D.C.Cir.1977); United States v. Aluminum Co. of America, 148 F.2d 416, 430 (2d Cir.1945) (L. Hand, J.); Fishman v. Wirtz, 1981-2 Trade Cas. p 64,378 at pp. 74,766-74,769 (N.D.Ill.1981); Otter Tail Power Co. v. United States, 410 U.S. 366, 389, 93 S. Ct. 1022, 1035, 35 L. Ed. 2d 359 (1973) (Stewart, J., dissenting). If a market has room for only one firm, it would be an effort worthy of King Canute to keep two firms in it. True, a cartel among the competing firms in such a market would be illegal per se, even if the alternative was the destruction, through unbridled competition, of all but one. See United States v. Trans-Missouri Freight Ass'n, 166 U.S. 290, 330-32, 338-40, 17 S. Ct. 540, 554-55, 557-58, 41 L. Ed. 1007 (1897). But that is because the cartel, by holding a price umbrella over the least efficient firms, would retard the evolution of the market toward its optimal state, which would be to have only one firm.

It is also true that the antitrust laws protect competition not only in, but for, the market--that is, competition to be the firm to enjoy a natural monopoly, see United States v. El Paso Natural Gas Co., 376 U.S. 651, 84 S. Ct. 1044, 12 L. Ed. 2d 12 (1964), and by a modest extension competition to replace the existing natural monopolist. If the most efficient method of determining which firm should have the natural monopoly is a competitive process that will inevitably destroy the other firms, the antitrust laws presumably would forbid interference with that process. If, therefore, Omega were challenging an agreement between Indianapolis Cablevision and American Cablevision not to invade each other's territories, or to cooperate in repelling an invasion of either territory by a third party such as Omega, the antitrust laws would condemn the agreement as an artificial interference with a natural, if in a sense destructive, competitive process. The case would be little different from the cartel agreement struck down in the Trans-Missouri case, supra.

But that is not Omega's complaint. It is rather that officials of the City of Indianapolis--the representatives of the consumers of cable television--have granted the existing franchisees de facto exclusive franchises. If consumers want to be spared a competitive free-for-all, though, and to this end, through their elected representatives, grant an exclusive franchise to a natural monopolist, it may be that no greater objection can be made in the name of the antitrust laws than was made to the full-requirements contract in Tampa Elec. Co. v. Nashville Coal Co., 365 U.S. 320, 81 S. Ct. 623, 5 L. Ed. 2d 580 (1961), and rejected by the Supreme Court--especially now that the Court has told us that consumer welfare is to be the lodestar in interpreting and applying those laws, see e.g., Reiter v. Sonotone Corp., 442 U.S. 330, 343, 99 S. Ct. 2326, 2333, 60 L. Ed. 2d 931 (1979). There is no precedent for condemning an exclusive franchise in these circumstances, at least out of hand, and it would be perilous to assume that this or any other court will soon create one. But that is not to say that Omega will not prevail on its Sherman Act count at trial. It may be able to prove that the City officials were not acting in the consumer interest, that cable television in Indianapolis is not a natural monopoly, that exclusive franchising is a needlessly restrictive way of dealing with natural monopoly, or other propositions that separately or together might establish a violation of the Sherman Act under the Rule of Reason. All we hold is that on this record Omega has not established a sufficient probability of prevailing to persuade us that it is entitled to a preliminary injunction on the basis of its Sherman Act claim.

We turn now to the First Amendment challenge to the ordinance. Omega is engaged in the dissemination of speech within the meaning of the First Amendment, both by transmitting programs originated by television stations and cable television networks and by originating its own modest programs. Omega is not so bold as to contend that this fact alone entitles it to build a cable television system in Indianapolis without having to get permission from the City. The Supreme Court has interpreted the First Amendment to allow more government regulation of television than of theaters, movies, books, newspapers, or open-air addresses. Compare Red Lion Broadcasting Co. v. FCC, 395 U.S. 367, 89 S. Ct. 1794, 23 L. Ed. 2d 371 (1969), with Miami Herald Pub. Co. v. Tornillo, 418 U.S. 241, 94 S. Ct. 2831, 41 L. Ed. 2d 730 (1974). True, the double standard was developed with respect to broadcast rather than cable television, and the cases articulating the standard emphasize the need to regulate broadcast frequencies in order to prevent frequency interference, a problem that does not arise with cable television. See, e.g., Red Lion, supra, 395 U.S. at 388-89, 89 S. Ct. at 1805-06. But cable television involves another type of interference--interference with other users of telephone poles and underground ducts. Moreover, the apparent natural monopoly characteristics of cable television provide, as we have seen, an argument for regulation of entry (though it should be noted that while today most newspaper markets are natural monopolies, no one thinks that entry into those markets could be regulated without creating profound First Amendment problems). Another explanation of the different treatment under the First Amendment between television and other media is the universal access to the home that television enjoys and a resulting felt need to protect children. See FCC v. Pacifica Foundation, 438 U.S. 726, 748-49, 98 S. Ct. 3026, 3039-40, 57 L. Ed. 2d 1073 (1978). This consideration is independent of whether the television signal comes into the home over the air or through a coaxial cable.

Probably there are enough differences between cable television and the nontelevision media to allow more government regulation of the former. See Community Communications Co. v. City of Boulder, 660 F.2d 1370, 1377-80 (10th Cir.1981). But it does not follow that a city may limit entry into cable television by any criteria or procedures that it chooses to employ. The City of Indianapolis could not deny a franchise to Omega because Omega carried programs critical of the Republican Party. Nothing of that sort is alleged, though. The allegation is that the franchise procedure in chapter 8 1/2 is so costly, multi-layered, political, and devoid of standards that they make the entry of a new cable television system unnecessarily difficult and thereby create the potential for denying a franchise on invidious grounds, such as disagreement with the political slant of the prospective entrant. Now that the initial period for the award of cable television franchises has expired, no new franchise can be granted until the City-County Council decides to open up the City or an area within the City to new franchises. The ordinance establishes no criteria to guide that decision. It does not even make clear how a would-be applicant is supposed to engage the Council's interest in the question of reopening the City to new franchisees, though the 1981 amendment implies that you do this by getting in touch with the Cable Communications Office. If the Council does decide to open up the City again, the applicant must pay a nonrefundable $3,000 fee just to get his application filed. The application will then be considered by subordinate units of local government, notably the Cable Franchise Board, which under the 1981 amendment succeeded to the functions performed by the Board of Public Works under the original ordinance. The list of criteria that the Board is to use in picking the "most desirable" application is incomplete and some of the criteria are vague, and anyway the Board's recommendation is not binding on the Council, which must confirm that recommendation by an ordinance before the franchise can actually be awarded--so that before Omega can get a cable television franchise it will have to run the gauntlet of the Council twice. Again, no guidelines are given the Council for exercising its discretion to confirm or reject the Board's recommendation.

There is, however, a big difference between the danger of an abuse and the abuse itself; and it is a fair question how far the courts should go in making municipalities rewrite their cable television ordinances to prevent dangers that may be largely hypothetical. It is true that in other areas of speech the balance has been struck in favor of requiring specific criteria, for example, in the regulation of access to public parks for speech-making. See, e.g., Niemotko v. Maryland, 340 U.S. 268, 272, 71 S. Ct. 325, 327, 95 L. Ed. 267 (1951), where the Supreme Court described how one city decided who could use its parks for public meetings: "No standards appear anywhere; no narrowly drawn limitations; no circumscribing of this absolute power; no substantive interest of the community to be served." But chapter 8 1/2 is not so bad as this, and it is uncertain as we have said how far the principles that have been applied to the old-fashioned methods of public address should apply to cable television. It is true that the procedure set up in chapter 8 1/2 for getting a franchise is cumbersome and lacks specific standards. But it is not obvious that the award of cable television franchises can be managed with the precision and simplicity appropriate to granting access to public parks for speechmaking. Maybe chapter 8 1/2 is the best possible--or at least a constitutionally adequate--accommodation between regulatory feasibility and the policy of the First Amendment. That is a matter to be explored at trial. Missing from the record of the preliminary injunction proceeding is any indication of how other cities have struck the balance. Portions of the Indianapolis ordinance are it is true patterned on the FCC's model cable television ordinance, Note following 47 C.F.R. Sec. 76.31, but unfortunately the model does not contain suggested provisions for determining who should receive a franchise.

In any event, the preliminary injunction was correctly denied. If the ordinance were likely to be invalidated on the ground that the City of Indianapolis may not limit the entry of cable television companies at all, then if the preliminary injunction were granted Omega might very well go ahead and build its system, assuming that it really wants to; and in such a case it could argue that the denial of the preliminary injunction would harm it irrevocably by increasing its competitors' headstart. But if chapter 8 1/2 is invalid under the First Amendment (a question we emphatically do not decide) it is so because it lacks adequate standards and procedures, not because a city may not limit the entry of cable television companies. If the enforcement of chapter 8 1/2 is enjoined, that will simply send the City back to the drawing board to draft a new cable television ordinance, with new limitations on entry; and Omega will have no assurance that it will be able to obtain a franchise under the new standards. Because of this uncertainty Omega is unlikely to build a cable television system in Indianapolis until the new ordinance is passed; and that is unlikely to happen before the case is finally decided after a full trial, no matter what we say today.

We realize the Supreme Court has said that "the loss of First Amendment freedoms, even for minimal periods of time, unquestionably constitutes irreparable injury." Elrod v. Burns, 427 U.S. 347, 373, 96 S. Ct. 2673, 2689, 49 L. Ed. 2d 547 (1976) (plurality opinion). But there are at least three ways to interpret this deceptively simple statement. The first, which is suggested both by the reference to time and an explanatory footnote, see id. at 374 n. 29, 96 S. Ct. at 2690 n. 29, is that timing is often a vital component of effective communication--news can lose its newsworthiness in hours. The second interpretation is that a violation of the First Amendment imposes social costs that might not be reflected in a preliminary-injunction balancing process that emphasized the private costs to the proponent and opponent of the injunction. The third is that to require proof that the speech which has been cut off had value greater than the harm that a preliminary injunction would cause would be inconsistent with the basic premise of the First Amendment, which is that the value of particular ideas is to be left to the marketplace of ideas to determine, Abrams v. United States, 250 U.S. 616, 630, 40 S. Ct. 17, 22, 63 L. Ed. 1173 (1919) (Holmes, J., dissenting), rather than to judges and other officials. Fortunately, we do not have to choose among these versions of the Elrod proposition. All presuppose that some curtailment of speech will occur if the preliminary injunction is denied. But we believe that, even if the preliminary injunction Omega seeks were granted, Omega would do no disseminating in Indianapolis, beyond what it is doing today with its satellite earth stations and its single cable underneath West 38th Street, at least until after the case is tried and a final judgment entered.

AFFIRMED.