(dissenting, with whom Cordy, J., joins). There are at least two different types of historical cell site location information (CSLI). Telephone call CSLI (the type sought by the Com*259monwealth and ordered by the court in this case) provides the approximate physical location (location points) of a cellular telephone only when a telephone call is made or received by that telephone. Registration CSLI (the type not sought by the Commonwealth or ordered by the court, and therefore the type not at issue in this case) provides the approximate physical location of a cellular telephone every seven seconds unless the telephone is “powered off,” regardless of whether any telephone call is made to or from the telephone. Telephone call CSLI is episodic; the frequency of the location points depends on the frequency and duration of the telephone calls to and from the telephone. Registration CSLI, for all practical purposes, is continuous, and therefore is comparable to monitoring the past whereabouts of the telephone user through a global positioning system (GPS) tracking device on the telephone, although it provides less precision than a GPS device regarding the telephone’s location. The court recognizes the differences between telephone call CSLI and registration CSLI, and then conducts its analysis under art. 14 of the Massachusetts Declaration of Rights as if those differences have no constitutional consequence or as if the court ordered the production of registration CSLI. I believe that those differences have fundamental constitutional consequence with respect to both the reasonableness of the expectation of privacy under the third-party doctrine and the extent of the intrusion on privacy, and therefore I respectfully dissent.
In Smith v. Maryland, 442 U.S. 735, 743 (1979), the United States Supreme Court held, under what has become known as the third-party doctrine, that telephone users had no subjective expectation of privacy in the telephone numbers they dialed because they “typically know that they must convey [the telephone numbers they call] to the [telephone company; that the [telephone company has facilities for recording this information; and that the [telephone company does in fact record this information for a variety of legitimate business purposes.” The Court also declared that, even if the defendant “did harbor some subjective expectation that the [telejphone numbers he dialed would remain private, this expectation is not ‘one that society is prepared to recognize as “reasonable.” ’ ” Id., quot*260ing Katz v. United States, 389 U.S. 347, 361 (1967). The Court noted that it “consistently has held that a person has no legitimate expectation of privacy in information he voluntarily turns over to third parties.” Smith v. Maryland, supra at 743-744, citing United States v. Miller, 425 U.S. 435, 442 (1976) (bank depositor has no legitimate expectation of privacy in financial information “voluntarily conveyed to . . . banks and exposed to their employees in the ordinary course of business”). See, e.g., Couch v. United States, 409 U.S. 322, 335-336 (1973) (individual may not invoke privilege against self-incrimination under Fifth Amendment to United States Constitution to protect financial and tax records held by his accountant); United States v. White, 401 U.S. 745, 752 (1971) (plurality opinion) (government agents did not violate Fourth Amendment to United States Constitution by listening to conversations between defendant and cooperating witness that they heard because witness secretly wore transmitter); Hoffa v. United States, 385 U.S. 293, 302 (1966) (Fourth Amendment does not protect “wrongdoer’s misplaced belief that a person to whom he voluntarily confides his wrongdoing will not reveal it”); Lopez v. United States, 373 U.S. 427, 437-438 (1963) (same, applied to secret consensual recording of attempted bribe to Internal Revenue Service agent). In each of these cases, the Supreme Court held that an individual enjoys no constitutional protection from the government obtaining information that he voluntarily furnished to a third party to advance the individual’s interest, whether that purpose is to make a telephone call, engage in banking transactions, prepare tax forms, confess wrongdoing, or attempt to pay a bribe.
In the instant case, the court acknowledges that we have “applied in substance the Supreme Court’s third-party doctrine” but have also recognized that art. 14 may, under certain circumstances, provide more substantive protection than is provided under the third-party doctrine. Ante at 244, citing Commonwealth v. Blood, 400 Mass. 61, 68 n.9 (1987), and Commonwealth v. Buccella, 434 Mass. 473, 484 n.9 (2001), cert. denied, 534 U.S. 1079 (2002). The court declares that “we see no reason to change our view that the third-party doctrine applies to traditional telephone records,” but concludes that “the distinctive characteristics of cellular telephone technology and *261CSLI” require the conclusion that, under art. 14, the third-party doctrine should not be applied to CSLI. Ante at 251-252.
The “distinctive characteristics” that the court identifies that lead to this conclusion, however, are characteristics of registration CSLI, not the telephone CSLI that the Commonwealth sought in this case and that the court ordered to be produced. Because nearly everyone now carries a cellular telephone and because CSLI “tracks the location of a cellular telephone user,” the court claims that “[i]t is evident that CSLI implicates the same nature of privacy concerns as a GPS tracking device,” and is arguably even more intrusive of privacy because it tracks the location of the cellular telephone carried on the person of its user rather than the location of his or her vehicle. Ante at 246, 248, 249. The court essentially contends that cellular telephone users are speaking on their cellular telephone so often that telephone CSLI will provide nearly as many location points as a GPS tracking device or registration CSLI, so that telephone CSLI is analogous to a GPS device in a cellular telephone user’s pocket. The court rests this contention on its own experience (“As anyone knows who has walked down the street or taken public transportation in a city like Boston, many if not most of one’s fellow pedestrians or travelers are constantly using their cellular telephones as they walk or ride,” see ante at 246) and on data regarding the annual volume of voice minutes used on cellular telephones in the United States (“In 2012, there were 2.3 trillion voice minutes of use on wireless devices such as cellular telephones in the United States,” see ante at note 29).
But this contention is empirically incorrect. According to the Wireless Association, the same source the court relies on for the annual number of voice minutes, there are more cellular telephones in the United States than United States residents.1 Therefore, the total United States population conservatively estimates the number of cellular telephones in active operation in the United States. If we take the annual volume of voice *262minutes (2.3 trillion), and divide it by the multiple of the total United States population in July, 2012 (313.87 million), and the number of minutes in a year (525,600), we learn that cellular telephone users spoke on the telephone in 2012 only 1.4 per cent of the day (0.01394). See CTIA: The Wireless Association, Wireless Quick Facts (Nov. 2013), http://www.ctia. org/your-wireless-life/how-wireless-works/wireless-quick-facts (last viewed Feb. 14, 2014); United States Census Bureau, Annual Estimates of Resident Population: 2013 Population Estimates (Dec. 2013), http://factfinder2.census.gov/faces/t-ableservices/jsf/pages/productview.xhtml?src=bkmk (last viewed Feb. 14, 2014). This rough estimate is corroborated by a survey conducted by PriceWaterhouseCoopers, which found that the average cellular telephone subscriber is speaking on the telephone .673 minutes per month, or 1.5 per cent of the day (0.01536). PriceWaterhouseCoopers, Real Time: The Growing Demand for Data, 2012 North American Wireless Industry Survey 36 (Apr. 2013), http://www.pwc.com/en__US/us/ industry/communications/publications/assets/pwc-north-american-wireless-industy-survey-2012.pdf (last viewed Feb. 14, 2014). Therefore, while it may seem as if Americans are always talking on their cellular telephones, they are actually doing so less than two per cent of the day. Therefore, there is a world of difference between telephone CSLI and registration CSLI in terms of the location points they will reveal and the degree to which they will intrude on personal privacy.
The telephone CSLI obtained in this case is much closer to the “traditional telephone records” that, the court agrees, are still governed by the third-party doctrine. While we have long accepted that the Commonwealth may obtain cellular telephone toll records without a search warrant supported by probable cause, it bears noting that the information revealed by those records intrudes deeply on personal privacy. Just as registration CSLI can “provide an intimate picture of one’s daily life,” by revealing “the people and groups they choose to affiliate with and when they actually do so,” ante at 248, quoting State v. Earls, 214 N.J. 564, 586 (2013), so, too, can telephone toll records, which can be used to identify who one speaks with on the telephone and how often.
*263Before cellular telephones, when telephones were located only in one’s home or business, traditional telephone records effectively revealed the location of the telephone user at the time of the call; if a person made a telephone call from a home telephone, the person was at home.2 Therefore, location information is not unique to telephone call CSLI; what has changed is the mobility of the telephone. I recognize that, because of the mobility of a cellular telephone, telephone call CSLI will provide many location points outside a user’s home or place of business, and these location points may provide a patchwork that will intrude on the user’s privacy to the extent that they reveal where the user is located when making or receiving calls on the telephone. But this patchwork of location points, while intrusive of privacy, is less intrusive than the patchwork of personal affiliations that can be learned from traditional telephone toll records. I also recognize that the degree of intrusion on privacy will depend on the number of calls the user makes and receives. But this is also true about traditional telephone records; the more telephone calls a person makes and receives, the more will be revealed regarding the persons the individual speaks with and the frequency of those calls.
Telephone CSLI, like telephone toll records, also fits within the traditional justification for the third-party doctrine. Every person who uses a cellular telephone recognizes that the location of the telephone matters in determining whether there is cellular service and, where there is such service, in determining the quality of the telephone connection, which is why at least one cellular telephone company advertises “more bars in more places.” Therefore, every person who uses a cellular telephone recognizes, at least implicitly, that a cellular telephone company must identify the location of a cellular telephone, as well as the telephone number called, before a call can be successfully made from a cellular telephone. Accordingly, although a cellular telephone user may not know that the telephone company records and keeps this information, or want it kept, the user should know *264that location information, as well as the telephone number, must be provided to the telephone company whenever he makes or receives a telephone call. See In re Application of the U.S. for Historical Cell Site Data, 724 F.3d 600, 611 (5th Cir. 2013) (information, such as telephone user’s location, that cellular telephone company needs to route telephone communications “appropriately and efficiently” falls within third-party doctrine).
I agree with the court and Justice Sotomayor that, in this digital age, “it may be necessary to reconsider the premise that an individual has no reasonable expectation of privacy in information voluntarily disclosed to third parties.” Ante at note 35, quoting United States v. Jones, 132 S. Ct. 945, 957 (2012) (Sotomayor, J., concurring). In the context of cellular telephone records, I would not extend the third-party doctrine to include information that is not necessary to the successful completion of a telephone call, and therefore would not apply the third-party doctrine to registration CSLI. A person obtains a cellular telephone for the purpose of making and receiving telephone calls, not to permit the telephone company or another third party to track the user’s location when the person is not using the telephone. Telephone CSLI is obtained by the telephone company to advance the telephone user’s interest in making or receiving telephone calls and is necessary to the successful completion of those telephone calls, while registration CSLI is not necessary because a telephone call may successfully be made even if the telephone had been powered off moments before.3 Where telephone toll records are covered by the third-party doctrine, and where location information is as necessary as the telephone number itself to the successful completion of a telephone call, I cannot find any principled reason why this *265doctrine would not also apply to telephone call CSLI.4 The court claims that there is a “significant difference” between cellular telephone toll records and CSLI in that “[n]o cellular telephone user . . . voluntarily conveys CSLI to his or her cellular service provider in the sense that he or she first identifies a discrete item of information or data point like a telephone number . . . and then transmits it to the provider.” Ante at 249-250. The difference is less than the court claims. First, it has been many years since a telephone caller had to make a call by dialing the operator at the telephone company, providing the operator with the number to be called, and asking the operator to connect you with that number. Today, a telephone caller no more voluntarily conveys a number to the telephone company than he voluntarily conveys his location to the telephone company, but he implicitly knows that the telephone company’s computers, need to know both for the call to be successfully connected. Second, for incoming telephone calls, the person receiving the call does not dial any number or otherwise convey any number, but the telephone number of the caller is nonetheless included in the cellular telephone toll records. Telephone CSLI is certainly different from telephone toll records, and provides different private information, but if the principle justifying the third-party doctrine for telephone toll records is that the information is necessary to the successful completion of telephone calls, there is no principled reason why the third-party doctrine should apply to telephone toll records but not to telephone call CSLI.5
Separate and apart from the third-party doctrine, the court’s analogy of CSLI to GPS tracking devices affixed to auto*266mobiles also is far weaker with telephone call CSLI than with registration CSLI. As I have noted, GPS tracking is continuous; registration CSLI is nearly so, providing location points for the cellular telephone every seven seconds. Telephone call CSLI is episodic, not continuous, and therefore its location points are not a continuous or continual line, but simply a patchwork of points. The extent to which that patchwork can reveal an intelligible picture of where the user goes and whom the user visits, and therefore the degree of intrusion on privacy, will depend both on the frequency of telephone calls and the duration of the CSLI request. We concluded in Commonwealth v. Rousseau, 465 Mass. 372, 382 (2013), “that under art. 14, a person may reasonably expect not to be subjected to extended GPS electronic surveillance by the government, targeted at his movements, without judicial oversight and a showing of probable cause.” We did not decide “how broadly such an expectation might reach and to what extent it may be protected,” or what duration less than thirty days would suffice as “extended” GPS electronic surveillance. Id.
Where that durational line is drawn, that is, determining when a locational surveillance (whether through GPS or CSLI) becomes so intrusive as to constitute an invasion of the reasonable expectation of privacy, is critical in finding the appropriate balance between personal liberty and legitimate law enforcement interests. No CSLI, whether it be telephone CSLI or registration CSLI, may be obtained by the Commonwealth without obtaining judicial authorization under 18 U.S.C. § 2703(d) (2006), and such authorization requires a showing of “specific and articulable facts showing that there are reasonable grounds to believe that the contents of a wire or electronic communication, or the records or other information sought, are relevant and material to an ongoing criminal investigation.” This may be “essentially a reasonable suspicion standard,” In re Application of the U.S. for an Order Pursuant to 18 U.S.C. § 2703(d), 707 F.3d 283, 287 (4th Cir. 2013), but it is merely reasonable suspicion that the CSLI records “are relevant and material to an ongoing investigation,” not reasonable suspicion that the user of the telephone has committed, is committing, or is about to commit an offense. Therefore, where the police *267receive informant information of uncertain reliability that a particular gang was involved in a shooting, the police under this standard may follow up on that lead by gathering the CSLI records at or around the time of the shooting for each gang member to determine whether they were in the vicinity of the shooting, even if they have nothing more to suggest that any particular gang member participated in the shooting.
Where a search warrant is required, however, the standard becomes “probable cause to believe ‘that a particularly described offense has been, is being, or is about to be committed, and that [the CSLI being sought] will produce evidence of such offense or will aid in the apprehension of a person who the applicant has probable cause to believe has committed, is committing, or is about to commit such offense.’ ” Ante at 256, quoting Commonwealth v. Connolly, 454 Mass. 808, 825 (2009). Because of the probable cause requirement and, more importantly, because there must be probable cause that the CSLI will produce evidence implicating the telephone user in a crime, the police will not be able to obtain a search warrant unless they already have obtained significant other information implicating the telephone user in a crime. Therefore, if a search warrant were required for all CSLI, regardless of duration, the police would not be able to use CSLI in my hypothetical shooting case to identify or eliminate possible suspects. A search warrant may appropriately be required where the CSLI, because of its duration and the number of location points it will identify, will reveal so much about the private life and personal affiliations of the telephone user as to invade the reasonable expectation of privacy, but it is not appropriate where the duration will reveal only where the telephone user was at a particular time or over a brief period of time. And, if the search warrant requirement is given inappropriate breadth, it will significantly diminish the ability of law enforcement to solve and to prove crimes, which so often depends on proving the whereabouts of a suspect at the time of the crime through his or her cellular telephone location.6
Because the court treats the telephone CSLI at issue in this *268case as if it were registration CSLI, and fails to recognize that this distinction is critical to the applicability of the third-party doctrine, to the adaptation of our GPS tracking jurisprudence, and to the determination whether the duration of CSLI surveillance invades the reasonable expectation of privacy, I respectfully dissent. Because the court order in this case allowed only for production of telephone CSLI over a two-week period, not registration CSLI, I would conclude under the third-party doctrine that the defendant had no reasonable expectation of privacy in his location points when he was making or receiving telephone calls, and reverse the judge’s allowance of the motion to suppress.
“Wireless penetration,” defined as “[the number] of active units divided by the total [United States] and territorial population,” in December, 2012, was 102.2 per cent. CTIA: The Wireless Association, Wireless Quick Facts (Nov. 2013), http://www.ctia.org/your-wireless-life/how-wireless-works/ wireless-quick-facts (last viewed Feb. 14, 2014).
Of course, just as with cellular telephones, the owner of the telephone line may not be the person using the telephone, or it may be used by multiple persons. The only difference between the traditional home telephone and a cellular telephone is that the latter is more likely to have a single user.
The court, quoting State v. Earls, 214 N.J. 564, 587 (2013), declares: “People buy [cellular telephones] to communicate with others, to use the Internet, and for a growing number of other reasons. But no one buys a [cellular telephone] to share detailed information about their whereabouts with the police.” Ante at 250-251. It is true that no one buys a cellular telephone “to share detailed information about their whereabouts with the police,” but it is also true that no one buys a cellular telephone to share with the police the telephone numbers of the calls they are dialing and receiving. In terms of the third-party doctrine, the meaningful distinction between telephone cell site location information (CSLI) and registration CSLI is not the cellular telephone owner’s willingness to share private information with the police.
I offer no opinion as to whether the third-party doctrine should apply to CSLI obtained when a cellular “smartphone” is using the Internet.
The court contends that there is the “probability that, as CSLI becomes more precise, cellular telephone users will be tracked in constitutionally protected areas,” namely one’s home. Ante at 253. The court noted, however, that the GPS “type of location tracking is not at issue here,” ante at note 21, and there is nothing in the record to suggest that CSLI is likely to become so precise in the immediate future that it will identify where inside a home a person is located. The theoretical possibility that telephone call CSLI may enable the police in the future to track a person’s movement within the home is not an independent ground to require a search warrant supported by probable cause.
The court appears to recognize this concern where it declares that “it would be reasonable to assume that a request for historical CSLI of the type at issue in this case for a period of six hours or less would not require the *268police to obtain a search warrant in addition to a § 2703(d) order.” Ante at note 37. However, because the court characterized this as a reasonable assumption rather than a safe harbor rule, and prefaced it with the declaration that “this is not an appropriate case in which to establish a temporal line of demarcation between when the police may not be required to seek a search warrant for historical CSLI and when they must do so,” prosecutors who procure CSLI for six hours or less through a § 2703(d) order without also obtaining a search warrant still risk the suppression of this CSLI evidence. The court need not in this case “establish a temporal line of demarcation,” but it should have established a safe harbor well within that “temporal line” to protect the invaluable investigative use of short-term CSLI.