UNITED STATES DISTRICT COURT AT SEATTLE KAELI GARNER, et al., Cause No. C21-0750RSL Plaintiffs, ORDER GRANTING IN v. PART DEFENDANTS’ MOTION TO COMPEL AMAZON.CO, INC., et al., Defendants. This matter comes before the Court on “Defendants’ Motion to Compel.” Dkt. # 106. Having reviewed the memoranda, declarations, and exhibits submitted by the parties, the Court finds as follows: A. Interrogatories Seeking Identification of Plaintiffs’ Recordings and the Subset of Recordings at Issue in this Case Defendants seek to compel plaintiffs to identify audio recordings that contain their voice and, in particular, the audio recordings are at issue in this case. Interrogatory 5 requests that each plaintiff identify by date, time, and content each Alexa audio recording associated with their Amazon account that contains their voice. Interrogatories 6, 7, and 8 (and in some cases Interrogatory 9) are designed to identify those recordings that each plaintiff contends were private and confidential conversations that were not directed to Alexa. Plaintiffs object on the grounds that Amazon has refused to produce audio recordings for all but five named plaintiffs and that the parties were in the midst of negotiating a production, review, and transcript annotation process when Amazon precipitously filed this motion to compel. The Court declines to evaluate the merits of Amazon’s objections to plaintiffs’ discovery requests in the context of this motion to compel. Plaintiffs may not refuse to produce responsive documents or information simply because they believe that Amazon has failed to meet its discovery obligations. The only relevant issue is whether Amazon’s refusal to make the audio recordings and transcripts available has prevented plaintiffs from responding to Amazon’s discovery requests. To a certain extent, that appears to be the case. At various points, Amazon argues that plaintiffs and their counsel have always had access to their audio recordings and transcripts through an on-line Amazon portal or the Alexa app (Dkt. # 106 at 11) and, contrarily, that there is reason to suspect that plaintiffs and their counsel are not authorized to access the information at all (Dkt. # 110 at 6). The difference appears to involve the way in which the Alexa account is held. If a named plaintiff is the account holder and no other plaintiff is seeking access to the account (as is the case for five of the twenty-one named plaintiffs)1, the account holder not only has password access to the digital files, but Amazon has been willing to assume that counsel is authorized to obtain the information requested and has produced the audio recordings and
1 While this motion was pending, eleven of the named plaintiffs voluntarily dismissed their claims against defendants. The numbers set forth in this Order reflect the pre-dismissal status of the claims. transcripts for those five plaintiffs. With regards to the twelve plaintiffs who live with another plaintiff, however, Amazon has refused to provide the audio recordings and transcripts for the account even though the account holder has requested them through properly-served discovery requests, apparently on the ground that a non-account holder plaintiff joined in the request. While this position is baffling, it is undisputed that the account holders have access to the information through the on-line Amazon portal and/or Alexa app and may share that access with their non-account holder spouses. Plaintiffs make no effort to show that this form of access is insufficient to allow the co-habitating plaintiffs to respond to the interrogatories seeking identification of each plaintiff’s recordings and the subset of recordings at issue in this case. Thus, with regards to at least seventeen named plaintiffs, there is no impediment to their providing full and complete responses to the interrogatories. There is a third group of plaintiffs who apparently cannot access the information necessary to respond, however. With regards to plaintiffs Kaeli Garner, Ricky Babani, and Caron Watkins,2 their claims are based on the allegation that they were recorded by Alexa devices owned by non-parties. Although there is some indication that the account holders have authorized Amazon to produce the audio recordings and associated transcripts in discovery, there is no evidence that the account holders have turned over their passwords or other access codes to plaintiffs such that they could obtain the records themselves. Amazon continues to 2 The Court is aware that the number of named plaintiffs discussed in this paragraph totals twenty but has been unable to ascertain into which category the last plaintiff falls. refuse to provide audio recordings or transcripts related to accounts of non-parties. These three plaintiffs will not be compelled to provide documents or information they do not have and to which they have no right of access. Plaintiffs argue that it would be unduly burdensome to require them to review all of their audio recordings and/or to annotate all of their Alexa transcripts and that such a requirement would not be proportional to the needs of the case. These assertions are wholly unsupported. There can be no doubt that the information sought is essential to plaintiffs’ claims and/or Amazon’s defenses. Each plaintiff’s familiarity with and use of Alexa will shed light on the adequacy of Amazon’s disclosures, the user’s awareness that recordings were made, stored, and reviewed, and the user’s control over the recordings. Each plaintiff will ultimately have to show that communications were improperly recorded: requiring them to identify the recordings that form the basis of their wiretap claims is therefore entirely proportional to the needs of the case. Nor is there any indication that the account holder plaintiffs or their non-account holder spouses have attempted to determine the number or length of the recordings or transcripts at issue. In the absence of such basic information (or some other metric from which one could make a reasonable estimate of the number of hours it would take to review, annotate, and produce the requested materials), a claim of undue burden is unpersuasive given the importance of the information. Finally, plaintiffs argue that the motion to compel responses to these interrogatories is premature because the parties had not reached an impasse. That does not appear to have been the case. Defendants requested that plaintiffs identify any recording of their voice as well as the subset of those recordings that form the basis for their claims. As part of the meet and confer process, defendants identified sources for the audio recordings and related transcripts and proposed that plaintiffs simply annotate the transcripts rather than typing out responses to individual interrogatories. Plaintiffs did not accept that proposal, instead insisting that Amazon produce the recordings and transcripts for review, at which point plaintiffs would be willing to discuss the extent to which they would provide responses. Having rejected defendants’ proposal and steadfastly refused to provide any responsive information, defendants appropriately sought the Court’s intervention. B. Verified Amended Responses Prior to Depositions The parties agree that plaintiffs shall provide verified amended responses from each plaintiff no later than three weeks before their scheduled deposition. It is so ordered. C. Request for Production (“RFP”) 15: Alexa Settings Defendants request that plaintiffs produce “[d]ocuments, communications, and things sufficient to show any Settings that have been set on each Alexa-enabled Device that You used to access the Alexa Voice Service.” They have clarified that they are requesting that plaintiffs render in a tangible form (i.e., take a screenshot of) the electronically-stored information regarding their account settings that is visible and accessible via the Alexa app or amazon.com. Plaintiffs argue that they sh
Free access — add to your briefcase to read the full text and ask questions with AI
UNITED STATES DISTRICT COURT AT SEATTLE KAELI GARNER, et al., Cause No. C21-0750RSL Plaintiffs, ORDER GRANTING IN v. PART DEFENDANTS’ MOTION TO COMPEL AMAZON.CO, INC., et al., Defendants. This matter comes before the Court on “Defendants’ Motion to Compel.” Dkt. # 106. Having reviewed the memoranda, declarations, and exhibits submitted by the parties, the Court finds as follows: A. Interrogatories Seeking Identification of Plaintiffs’ Recordings and the Subset of Recordings at Issue in this Case Defendants seek to compel plaintiffs to identify audio recordings that contain their voice and, in particular, the audio recordings are at issue in this case. Interrogatory 5 requests that each plaintiff identify by date, time, and content each Alexa audio recording associated with their Amazon account that contains their voice. Interrogatories 6, 7, and 8 (and in some cases Interrogatory 9) are designed to identify those recordings that each plaintiff contends were private and confidential conversations that were not directed to Alexa. Plaintiffs object on the grounds that Amazon has refused to produce audio recordings for all but five named plaintiffs and that the parties were in the midst of negotiating a production, review, and transcript annotation process when Amazon precipitously filed this motion to compel. The Court declines to evaluate the merits of Amazon’s objections to plaintiffs’ discovery requests in the context of this motion to compel. Plaintiffs may not refuse to produce responsive documents or information simply because they believe that Amazon has failed to meet its discovery obligations. The only relevant issue is whether Amazon’s refusal to make the audio recordings and transcripts available has prevented plaintiffs from responding to Amazon’s discovery requests. To a certain extent, that appears to be the case. At various points, Amazon argues that plaintiffs and their counsel have always had access to their audio recordings and transcripts through an on-line Amazon portal or the Alexa app (Dkt. # 106 at 11) and, contrarily, that there is reason to suspect that plaintiffs and their counsel are not authorized to access the information at all (Dkt. # 110 at 6). The difference appears to involve the way in which the Alexa account is held. If a named plaintiff is the account holder and no other plaintiff is seeking access to the account (as is the case for five of the twenty-one named plaintiffs)1, the account holder not only has password access to the digital files, but Amazon has been willing to assume that counsel is authorized to obtain the information requested and has produced the audio recordings and
1 While this motion was pending, eleven of the named plaintiffs voluntarily dismissed their claims against defendants. The numbers set forth in this Order reflect the pre-dismissal status of the claims. transcripts for those five plaintiffs. With regards to the twelve plaintiffs who live with another plaintiff, however, Amazon has refused to provide the audio recordings and transcripts for the account even though the account holder has requested them through properly-served discovery requests, apparently on the ground that a non-account holder plaintiff joined in the request. While this position is baffling, it is undisputed that the account holders have access to the information through the on-line Amazon portal and/or Alexa app and may share that access with their non-account holder spouses. Plaintiffs make no effort to show that this form of access is insufficient to allow the co-habitating plaintiffs to respond to the interrogatories seeking identification of each plaintiff’s recordings and the subset of recordings at issue in this case. Thus, with regards to at least seventeen named plaintiffs, there is no impediment to their providing full and complete responses to the interrogatories. There is a third group of plaintiffs who apparently cannot access the information necessary to respond, however. With regards to plaintiffs Kaeli Garner, Ricky Babani, and Caron Watkins,2 their claims are based on the allegation that they were recorded by Alexa devices owned by non-parties. Although there is some indication that the account holders have authorized Amazon to produce the audio recordings and associated transcripts in discovery, there is no evidence that the account holders have turned over their passwords or other access codes to plaintiffs such that they could obtain the records themselves. Amazon continues to 2 The Court is aware that the number of named plaintiffs discussed in this paragraph totals twenty but has been unable to ascertain into which category the last plaintiff falls. refuse to provide audio recordings or transcripts related to accounts of non-parties. These three plaintiffs will not be compelled to provide documents or information they do not have and to which they have no right of access. Plaintiffs argue that it would be unduly burdensome to require them to review all of their audio recordings and/or to annotate all of their Alexa transcripts and that such a requirement would not be proportional to the needs of the case. These assertions are wholly unsupported. There can be no doubt that the information sought is essential to plaintiffs’ claims and/or Amazon’s defenses. Each plaintiff’s familiarity with and use of Alexa will shed light on the adequacy of Amazon’s disclosures, the user’s awareness that recordings were made, stored, and reviewed, and the user’s control over the recordings. Each plaintiff will ultimately have to show that communications were improperly recorded: requiring them to identify the recordings that form the basis of their wiretap claims is therefore entirely proportional to the needs of the case. Nor is there any indication that the account holder plaintiffs or their non-account holder spouses have attempted to determine the number or length of the recordings or transcripts at issue. In the absence of such basic information (or some other metric from which one could make a reasonable estimate of the number of hours it would take to review, annotate, and produce the requested materials), a claim of undue burden is unpersuasive given the importance of the information. Finally, plaintiffs argue that the motion to compel responses to these interrogatories is premature because the parties had not reached an impasse. That does not appear to have been the case. Defendants requested that plaintiffs identify any recording of their voice as well as the subset of those recordings that form the basis for their claims. As part of the meet and confer process, defendants identified sources for the audio recordings and related transcripts and proposed that plaintiffs simply annotate the transcripts rather than typing out responses to individual interrogatories. Plaintiffs did not accept that proposal, instead insisting that Amazon produce the recordings and transcripts for review, at which point plaintiffs would be willing to discuss the extent to which they would provide responses. Having rejected defendants’ proposal and steadfastly refused to provide any responsive information, defendants appropriately sought the Court’s intervention. B. Verified Amended Responses Prior to Depositions The parties agree that plaintiffs shall provide verified amended responses from each plaintiff no later than three weeks before their scheduled deposition. It is so ordered. C. Request for Production (“RFP”) 15: Alexa Settings Defendants request that plaintiffs produce “[d]ocuments, communications, and things sufficient to show any Settings that have been set on each Alexa-enabled Device that You used to access the Alexa Voice Service.” They have clarified that they are requesting that plaintiffs render in a tangible form (i.e., take a screenshot of) the electronically-stored information regarding their account settings that is visible and accessible via the Alexa app or amazon.com. Plaintiffs argue that they should not have to provide screenshots of their Alexa settings because (a) it would require them to create documents that do not currently exist, (b) the information is already in defendants’ possession, custody, and control, and (c) their current settings will not “show with any degree of certainty what the Alexa’s settings were in the years leading up to the filing of this lawsuit.” Dkt. # 108 at 9. Plaintiffs’ arguments are not well-taken. Rule 34(a) authorizes requests for electronically-stored images or data in the responding party’s possession, custody, or control. Plaintiffs do not dispute that they have possession, custody, or control over the requested data, and converting the date from its stored electronic medium into a screenshot is not unduly burdensome and specifically authorized under Rule 34(a). The fact that Amazon has the same setting information on its servers does not make RFP 15 improper. Under Rule 26(b)(2)(C)(i), the Court is required to limit discovery if the documents, data, or information “can be obtained from some other source that is more convenient, less burdensome, or less expensive.” It is plaintiffs’ burden to show that a protective order is warranted, and they offer no evidence that Amazon’s access to individual customer settings is more convenient, less burdensome, or less expensive than having plaintiffs access and produce screenshots of their account settings page. Alves v. Riverside Cnty., 339 F.R.D. 556, 559 (C.D. Cal. 2021). 3 In addition, screenshots of plaintiffs’ Alexa settings would show exactly what plaintiffs see and can access through the user interface on the Alexa app or amazon.com. As defendants point out, “[a] picture tells a thousand words,” and the format and ease in which
3 The Court has not considered defense counsel’s hearsay assertion that the account settings information “is not as easily and readily accessible to Amazon because Amazon separates access to different types of data across different teams.” Dkt. # 107 at ¶ 17. this data can be accessed by plaintiffs is almost as important as the settings each plaintiff has chosen. Dkt. # 106 at 15. With regards to the relevance objection, the fact that the screenshot of current settings does not definitively establish what plaintiff’s past choices may have been does not preclude the requested discovery. Plaintiffs’ current choices offer a starting point for deposing them regarding their current selection, their familiarity with available options, and whether any changes were made over time. In addition, the screen shot is evidence of what each individual plaintiff has chosen today, after learning of Amazon’s allegedly wrongful conduct. And, as set forth above, the form and accessibility of these choices may be relevant to determining notice and knowledge. D. Search Term Hit Reports Three days before Amazon filed this motion to compel, plaintiffs agreed to “provide hit reports and their ESI Search Methodology this week,” i.e., before August 19th. Dkt. # 109 at 14. They have now done so, and no further production will be compelled. E. Timing of Production On August 15, 2022, plaintiffs agreed to “produce another batch of documents in August and [to] substantially complete their productions by September 16, 2022.” Dkt. # 109 at 16. Amazon was under the impression that plaintiffs would complete their production by the September date (Dkt. # 107-14 at 3) and asked that plaintiffs clarify what they meant by “substantially complete” (Dkt. # 107-14 at 2). When plaintiffs did not respond, Amazon included a request that plaintiffs be compelled to complete their production by September 16th in this motion. Plaintiffs have agreed to produce all non-privileged documents responsive to the agreed- upon search terms by September 16th. It is so ordered.
For all of the foregoing reasons, defendants’ motion to compel (Dkt. # 106) is GRANTED in part and DENIED in part. All plaintiffs except Kaeli Garner, Ricky Babani, and Caron Watkins shall identify all of their audio recordings and annotate the corresponding transcripts to identify communications at issue in this litigation, as requested in Interrogatories 5-8 (and possibly 9). Plaintiffs shall also supplement their responses to RFP 15, produce all non- privileged documents responsive to the agreed-upon search terms by September 16th, and provide verified amended responses from each plaintiff no later than three weeks before their scheduled deposition.
Dated this 3rd day of October, 2022.
Robert S. Lasnik United States District Judge