In Episode 169, Kelly Twigger discuss the latest ruling on hyperlinked files from United States Magistrate Judge Lisa Cisneros in the In re Uber Techs class action, following up from Episode 141 on the same case in In re Uber Techs., Inc. Passenger Sexual Assault Litig.
Introduction
Welcome to our Case of the Week segment on our podcast. My name is Kelly Twigger. I am the Principal at ESI Attorneys, a law firm for ediscovery and information law, and the CEO and founder at eDiscovery Assistant, where we take the insights from our practice and provide a knowledge platform for you to leverage the power of Electronically Stored Information (ESI). Thanks so much for joining me today.
We are T-minus five days and counting from the start of LegalWeek in New York City, and our team from eDiscovery Assistant will be there and we would love to see you. We’ll be hosting coffee and donuts at our suite at the Hilton on Tuesday and Wednesday mornings and setting up meetings for those interested in seeing the platform, as well as hearing about our roadmap for 2025 and seeing what’s new in the platform. We also have some very exciting news to share about the future of the company, so please stop by and see us in our suite. You can sign up here.
Now, on with the show.
This week’s decision is the latest in the saga of hyperlinked files and when a responding party has to provide documents shared at links in communications, regardless of the application that they are shared from. We are revisiting the discovery dispute-rich case of In re Uber Technologies Passenger Sexual Assault Litigation in front of United States Magistrate Judge Lisa Cisneros. This decision is from March 3, 2025 and represents the first of three decisions on hyperlinked files just this month, in March, to resolve the issues raised by plaintiff following Uber’s production of documents.
We’ve covered decisions from this case previously, most recently on Episode 141 of the Case of the Week. In that episode, I walked you through Magistrate Judge Cisneros’ order regarding the language to be included in the parties’ ESI protocol on hyperlinked files. Recall that Uber leveraged Google Mail and Google Apps for its employees and also archived data into Google Vault. There is a detailed discussion in Episode 141 of those various technologies and the ability to collect information from those technologies that is assumed for purposes of this decision, so if you need to you can go back and take a look at that one.
Background and Facts
The March 3, 2025 decision from the Court comes following a joint letter from the parties. We’ve talked multiple times on the Case of the Week about the benefits of using a letter dispute process with courts and how you should engage with your judges to use that type of process if you feel like you’re going to have a situation with a lot of discovery issues.
In this ruling, the Court noted initially that the disputes the issues before it implicate the parties’ ESI protocol and reiterated its ruling from May 2024 in which the Court held that a hyperlinked document is an attachment for purposes of the ESI protocol in this litigation, because it is akin to a traditional email attachment where the email message and hyperlinked document reflect a single communication at a specific point in time.
That’s a very important point here — that Magistrate Judge Cisneros has said that a hyperlinked document is an attachment because it is akin to a traditional email attachment where the email message and hyperlinked document reflect a single communication at a specific point in time. Now that is different than other judges who have found that hyperlinked files are not an attachment. So who you’re in front of and how informed your judge is is going to depend on how this issue goes.
Here, the Court also noted that the ESI protocol that the parties had agreed upon and submitted did not require Uber to produce contemporaneous versions of all hyperlinked files. That came after Uber successfully argued that it was not technologically possible for it to provide contemporaneous versions of hyperlinked documents from Google Vault. But the Court found that Uber was able to provide contemporaneous versions from Google Drive. If you go back and look at Episode 141, it’ll give you a detailed discussion of the differences between Google Vault and Google Drive and the technology that’s available to be able to pull contemporaneous versions and metadata from Google Drive, but not from Google Vault.
But the ESI protocol here did require Uber to produce contemporaneous versions of documents that were not in Vault but in the Google Drive, and the protocol also required Uber to provide metadata linking the hyperlinked document to the email where it is referenced. And that’s what’s really critical. Because, as we know, when we see an email with a hyperlink in it, we just have that link, we just have that URL. We don’t know whether that’s an active link, we don’t know whether that document still exists. And in the context of a production, we don’t have any metadata that links a subsequently produced hyperlinked document back to that message, or I should say that communication, whether it’s a chat, an email or anything where a link is included.
With that, referencing her May 2023 order, Magistrate Judge Cisneros stated that “The lesson from the Court’s prior order, which bears on the current dispute, is that, while attachments are broadly defined to include hyperlinks, the production of all hyperlinked documents is not required in light of the pervasive technical challenges.”
Analysis
With that background, let’s turn to the issues before the court. There are several here. First, the plaintiffs sought metadata and underlying documents for hyperlinks that referenced sources other than Google Drive. And this is where the language of the protocol becomes key. Uber argued that the protocol only focuses on providing hyperlinked files from Google Drive and that it was not technologically feasible for them to provide metadata for all documents containing hyperlinks. Plaintiffs argued that the protocol language required Uber to provide sufficient metadata to establish parent-child relationships between the documents and their attachments, regardless of whether or not they linked to Google Drive documents. So the whole issue here is the protocol specifically addresses Google Drive documents, but it doesn’t address hyperlinks for communications in other applications.
Looking at the language of the protocol, the Court found that even if the protocol could be found to require metadata for hyperlinked documents not in Google Drive, there is no obligation for a party to do that where it is not technologically feasible: “Uber cannot use a method of collection and processing that preserves a certain metadata relationship, if the method does not exist.” The Court also noted that it was not clear what metadata fields that the plaintiffs actually sought from the non-Google Drive hyperlinked files. From my perspective, if metadata about hyperlinked documents is laid out in the ESI protocol, it would be pretty consistent as to what it is that the plaintiffs actually were seeking here. But that’s another lesson — that you need to articulate to the court at every level exactly what you are asking for. Even though you’ve had the disputes before, you need to tell the court exactly what you’re looking for in every discovery dispute.
To resolve the issue, the Court ordered the parties to meet and confer and come up with a process by which the plaintiffs could request the bates number or the production of a hyperlinked document where they believe a hyperlinked document was potentially relevant. That process is in line with Magistrate Judge Susan van Keulen’s decision in McLaughlin v. Tesla, Inc. from 2023.
The Court here did not require Uber to produce documents for links to public websites, and that’s in keeping with Magistrate Judge Parker’s decision from Nichols v. Noom Inc. back in 2021. Interestingly, though, the Court made the following comment about the process for the parties to come up with: “Such a protocol may include a limit on the number of requests Plaintiffs may make if the parties believe that is appropriate. If there is a dispute as to the number of requests to allow, the Court is inclined to allow a large number.” I love this in situations when judges signal to the parties what their thinking is, so that the parties come up with something that the court will deem reasonable. In this case, it signals the fact that the Court sees this as affecting more than just a few documents and it says specifically that the Court is inclined to allow a large number of requests to Uber to produce either metadata or hyperlinked documents.
Finally, on this issue, the Court made a statement that is important to note because it shows that she knows the technology will continue to move forward and that, as the case progresses, new technology may allow the parties to provide more and better data, and she stated this:
Separately, the parties are also encouraged — but not required — to consider whether it is technologically feasible to include further useful information regarding hyperlinked non-Google Drive documents in future document productions, and whether doing so would be more efficient than providing such information through case-by-case requests.
It’s highly likely that technology will allow for more collection and production as this litigation moves forward. If it does, then the question will become whether Uber is required to go back and re-produce the same information that could not be produced previously. And that, to me, is a whole new can of worms.
The second issue involved here between the parties raised Uber’s lack of production of hyperlinked files for documents outside of Gmail. The Court noted that the ESI protocol applies to hyperlinked files that “appear in communications” and ordered Uber to produce documents hyperlinked in Google Chat messages and for both parties to meet and confer to find a process to provide documents and metadata from hyperlinks in documents other than Gmail and Chat messages. So that is really saying okay, well, you parties contemplated Google Drive, but we didn’t talk about these other things as it relates to hyperlinks, but based on the language in the ESI protocol that everything that appears in communications related to hyperlinks, that means any source of ESI where hyperlinks are used. So, not just Google Drive, not just Gmail.
The last issue that the parties raised here before the Court included the hyperlinks issue in email threading. This one addressed Uber’s lack of production of email threads, but also relates to hyperlinked files in email threads. The ESI protocol for the matter allowed parties to use email threading for internal review, but required that no email would be withheld from production because it is included in an email thread.
This is the whole discussion that we’ve been through on email threading. If you have one email at the top, that’s the last included email, and then you have maybe 10 or 15 emails below it. Here, Uber is keeping out non-relevant emails from those email threads and the plaintiffs are disputing that issue. They’re also disputing the issue of where a relevant email that has a hyperlink or an email that they believe is relevant has a hyperlink, that email is left out of the production but the hyperlink later shows up in quoted information in another email thread. The question is whether or not the hyperlinked files in those email threads should be produced according to the protocol and according to what the judge is ordering here with regard to these other issues that we discussed.
Plaintiffs argue that where those emails are produced, Uber has not produced the documents or metadata for those included hyperlinks. Uber made an interesting argument here, stating that it treated links to those documents buried in email threads as traditional attachments — meaning that they would not necessarily be attached to the reply message in the way that, you know, an email might happen. So sometimes when I respond to an email that has an attachment, the attachment may be attached to the reply. Other times it may not be. It often depends on the email client and how your settings are configured. So, what Uber is arguing is that, hey, we treated these like traditional attachments, that the hyperlinked file wouldn’t necessarily be attached to a reply, so we didn’t produce it. That’s a sly argument on Uber’s part, but it’s a bit disingenuous when we just had a long discussion about the requirement to produce hyperlinked files from the ESI protocol. The Court disagreed with Uber, finding that while it can be true that traditional attachments are not always included in replies, they are almost always included when the email is forwarded.
Nevertheless, the Court identified the issue as this — and this is an important one, so pay attention.
Can a hyperlink within an earlier email be considered the equivalent of an attachment to a later email that quotes the earlier email? Under some circumstances, perhaps — but the question does not have a clear, universal answer, and the ESI Protocol does not speak to this issue directly.
With no clear path, the Court ordered the parties to meet and confer using the approach outlined on the earlier issues.
Takeaways
So what are our takeaways from this decision? Well, they are plentiful.
First, this case really reiterates the iterative nature of discovery and that you are always going to learn more as you go through the process. You’re always going to identify issues that you didn’t contemplate when you drafted the ESI protocol. It’s almost impossible to consider every issue when drafting that protocol. The parties here looked at what they had in front of them and negotiated an incredibly thoughtful and detailed protocol, but, as we see in this decision, it still had holes. It didn’t address what happens when a hyperlink is in an email thread where not all emails in the thread are relevant but the link is produced in a subsequent email, and it didn’t contemplate specifically what happens with hyperlinks in applications other than Google Drive.
The point here is that there are always going to be details that can’t be contemplated or that the parties fail to contemplate, and that’s why having an iterative process and language in your protocol to handle the general thought processes about how the parties wish to approach issues is key. That general language, like here on hyperlinks and email threading, acts as a guide for the court in making decisions on how to proceed on these complex issues. Be thoughtful about your protocol. It has ramifications all the way through your case. This decision that we looked at today is one of 37 discovery decisions in this class action case. It’s extensive and it’s difficult to cover all of the issues.
With regard to hyperlinks, today’s decision is another step in yet a consistent progression of case law on this issue, and it required the parties to come up with a process, which they did in a subsequent decision, allowing the plaintiffs to request documents and metadata on specific hyperlinks. What I love about Magistrate Judge Cisneros’ decision here, and it’s been consistent in her decisions on this matter, is that she is constantly looking at what is technologically feasible to be done.
The analysis about whether a party is entitled to the data is not the issue, and I don’t think it should be. It’s whether the technology exists to provide it. A party can only do what technology allows it to do, and a court is not going to require more than that. But the Court also noted here that if technology advances on these issues as the case progresses, that Uber will be required to provide data when it is able to do so.
That raises a question I noted earlier that will strike fear into the hearts of all producing parties. If technology evolves to allow for the production of contemporaneous versions of hyperlinked files and metadata, will a party be required to go back and reproduce that information later? That’s going to be a tough question for us to answer.
Conclusion
That’s our Case of the Week for this week. Be sure to tune in for our next episode, whether you’re watching us via our blog, YouTube, or downloading it as a podcast on your favorite podcast platform. You can also find back issues of Case of the Week on your favorite podcast platform and be sure to subscribe, as we’ll be adding new content apart from the Case of the Week segments. There are some exciting things coming with the podcast in the next couple of weeks, and we look forward to sharing those with you. Have a great week!
As always, if you have suggestions for a case to be covered on the Case of the Week, drop me a line. If you’d like to receive the Case of the Week delivered directly to your inbox via our weekly newsletter, you can sign up on our blog. If you’re interested in doing a free trial of our case law and resource database, you can sign up to get started.
