News

Unilateral GPS tracking via cell phone app violates Collective Agreement

In Brick and Allied Craft Union of Canada & Brick and Allied Craft Union of Canada, Local 31 v Terrazzo, the Ontario Labour Relations Board found that unilaterally implementing a digital timesheet tracking system on employee’s personal phones was in violation of two provisions of the collective agreement and also an unreasonable exercise of management rights. Tsheets is a digital timesheet app downloaded onto employee smart phones to allow them to electronically “clock in” and out of their shifts. The employer unilaterally implemented a digital timesheet app called “Tsheets” in November 2018. In response the Union brought a grievance application, arguing that this change violated the Collective Agreement in a number of ways.

The operation of the app, and its use by the employer, was summarized by the Board:

3. Tsheets is a digital timesheet application that allows employees to electronically record the time they start and end their workday on a cellular phone. An employee “clocks in” at the beginning of his shift and “clocks out” at the end. It requires the use of cellular data from the employee’s personal cellular plan.

4. In order to use Tsheets, the employee must download the Tsheets app on his smart phone. The employer installed Tsheets software on its computer in order to receive the data from the Tsheets app. The Tsheets app sends all the recorded data, through the Tsheets servers, to the employer.

5. Once a job is set up by the employer, Mr. Emiliani assigns a job number and job site in the Tsheets software. When the employee starts work, he is expected to open the app, use the pull-down menu to find the appropriate job number, and “clock in”.

6. The app is also capable of tracking the location of the employee (or more accurately, the location of the employee’s phone) through Global Positioning System (“GPS”) technology. By default, the GPS option on the app is enabled and transmits the location to the employer. If the employee disables the GPS on their phone, Tsheets is not capable of tracking the GPS.

7. To explain this in greater detail, if the GPS option is enabled on the employee’s phone, the phone sends GPS data to Tsheets while the employee is “clocked in”. That data is then sent, in real time, to the employer…

11. Tsheets allows the employer to determine which features it wants to use and which features should be unavailable. For example, the employer had the option of deactivating the GPS function. That is, even if the employee’s GPS function was activated on his phone, the employer could have made the settings for Tsheets to not track the employee’s GPS. It chose not to do so. The settings of Tsheets allowed GPS tracking for any employee who did not manually turn off the GPS tracking on their phone.

The employer had implemented use of the app to eliminate the time necessary to physically travel and collect time sheets in order to manage payroll. The Union objected that this change violated the Collective Agreement in five ways, only two of which the Board found it necessary to decide.

First, the Collective Agreement specifically considered the submission of timesheets, and provided that “A timesheet may be submitted by telephone, email or fax.” This language was exhaustive, and so requiring the use of Tsheets was inconsistent with the Collective Agreement.

Second, the Collective Agreement prohibited the use of personal cell phones during working hours on any project. The use of Tsheets, on the other hand, required the use of a personal cell phone during working hours, and so again was inconsistent with the Collective Agreement.

More importantly, however, the Board found that even if the use of Tsheets were not inconsistent with the Collective Agreement, it was an unreasonable exercise of management rights. First, the unilateral implementation of Tsheets forced employees to own, carry and use their personal cell phones at work. Second, employees were forced to pay for their own data in using Tsheets: the Board noted that:

45…While the employer asserted that employees can utilize Tsheets on public wifi networks (e.g. at a local restaurant), this seems impractical in the circumstances, since the employee is required to “clock out” at the end of their shift. To require an employee to relocate to an area where wifi was publicly available is not a practical response to avoid the use of personal data.

Most importantly, the Board found there to be notable privacy concerns with Tsheets:

46. Third, the storage, use and potential sharing of data by Tsheets is also of concern because the employees are not informed by the employer about how such data is protected or used. If looked at in isolation, the collection of “clocked in” and “clocked out” times might not give rise to any concerns. But, when that data is combined with GPS, including locations during breaks and lunches, a more complete picture of employee behavioural characteristics can be seen by Tsheets and any third party provider authorized by Tsheets to view the information.

47. It is unclear to me whether data such as facial recognition (a security option), passwords, Information Protocol (“IP”) addresses, online cookies, or other cell phone usage is tracked and stored by Tsheets. The Privacy Statement of Tsheets states that it collects device information, usage information, location information, content, camera and contacts, information from cookies and local storage. There was no evidence about the collection or use of this information. Thus, I am unable to draw conclusions about the privacy implications. It is obvious that there could be real privacy issues with allowing Tsheets (and possibly the employer) to having access to the data collected and controlled on personal cell phones.

48. It is also concerning that the possibility of such information being shared with third parties is not addressed in the employer’s policy or part of a detailed explanation to the union and the employees. The sharing of personal information with third parties is an issue employees would be reasonably concerned about and I would have expected the employer to thoroughly address it.

In addition, Tsheets automatically sent notifications to employees reminding them to “clock in”, but these notifications were sent to all employees, including those who were not scheduled to work or those who were off due to illness. This might have been reasonable if the employer issued cell phones to its employees, but “it is unreasonable for an employer to unilaterally maintain some control (e.g. sending notifications) over the personal cell phones of the employees” (para 51). Accordingly, the grievance was decided in favour of the union.

US court hold that “thumbs-up” emoji did not amount to consent for removal of child from country

In Bardales v. Lamothe, Judge Eli Richardson of the US District Court for the Middle District of Tennessee (Nashville) presided over an application under the Hague Convention on the Civil Aspects of Child Abduction, in which a Honduran father alleged that his child’s mother (from whom he was separated) had abducted their child from Honduras and taken him to the US, in violation of the custody arrangement in place. The father proved to the court’s satisfaction that he had been exercising his parental and custodial duties at the time the child was removed, and that his custodial rights under Honduran law had been breached by the mother’s removal of the child.

However, the mother defended, in part, on the basis that while the father had not consented to her removing the child, he had subsequently acquiesced to her keeping the child in the US permanently. In support of this she produced a series of text messages, in one of which she told the father that they had arrived in the US and he responded with a “thumbs-up” emoji. The father testified that this only meant that he was indicating that she should enjoy the company of her mother (with whom she was staying), and not that he was consenting to the child remaining there. In another text exchange the father stated that he was going to “work hard, get his papers and come to the United States,” though the father testified that he meant he was intending to come to the US to retrieve his son, and not to live there. In still another, the mother texted the father that their son “would be an important man” in the US, to which the father replied, “perfect.”

Judge Richardson found that as evidence of acquiescence, these texts were “ambiguous, at best,” and any inference of acquiescence that could be drawn was rebutted by the fact that the father had consistently made efforts to get the child back, in part by enlisting the help of the Honduran authorities and, eventually, bringing the court application. The mother had not proven, on a preponderance of the evidence, that the father had acquiesced. In the result, the court ordered that the child be returned to Honduras.

Bitcoin cleared by Ontario Securities Commission to issue final prospectus

In [Re] 31Q Corp. and the Bitcoin Fund, Commissioner Lawrence P. Haber heard a review in which Bitcoin Fund and 31Q Corp. requested that he set aside a decision of the Director of the Ontario Securities Commission’s Investment Funds & Structured Products branch denying a receipt for Bitcoin Fund’s final prospectus. The effect of the decision was that Bitcoin could not distribute its securities to the public in Ontario. The Bitcoin Fund is a non-redeemable public investment fund that plans to invest substantially all of its assets in bitcoin (using blockchain to record all transactions), while 3IQ manages the fund.

Noting that the decision was not about the merits of the plan as a potential investment and that it was not the Commission’s function to inoculate potential investors against risk, Commissioner Haber made some interesting preliminary remarks:

[5] This application is not about the merits of bitcoin as an investment. As with other classes of assets or undertakings or businesses underlying an issuer, the investment potential of these underlying assets, undertakings and businesses are outside the scope of securities regulation.

[6] Bitcoin is a novel asset in an emerging and evolving market. It is a risky asset. Markets for novel asset classes and securities evolve over time. Emerging markets for securities and asset classes look and feel very different from mature markets. As markets evolve and mature, they change, either through the efforts of the market participants or through government intervention or regulation, or both.

[7] Some novel asset classes and securities products fail. They become tulip bulbs or dot.com’s. Others succeed and become gold or the next great technology.

The decision below had been based in the Commission’s “broad, but not infinite” public interest jurisdiction, under which staff had raised three operational risks relating to the project:

A) Valuation and Market Manipulation

Bitcoin is a commodity, not an equity or other security and therefore the bitcoin market should be examined like other commodity markets as opposed to the standards applicable to securities markets (para 80). Market manipulation is a risk in all commodity markets (para 80), and the IFSP staff was not successful in demonstrating that bitcoin is more susceptible to manipulation than other commodity products (and not in the public interest).

B) Safeguarding of the Fund’s Assets

Bitcoin can be stolen and lost like other valuable commodities (para 85). However, the Bitcoin Fund will use a regulating Canadian trust company as its custodian (Cidel Trust Company) as well as a New York State trust company.as a sub-custodian (Gemini Trust Company, LLC), which is regulated by New York State. The safeguards in place with these two custodians, and reputable reports indicate that they have effective internal controls and cybersecurity practices, as well as disaster recovery plans (paras 87, 88).

C) Auditability of the Fund’s Financial Statements

“Investment funds that are reporting issuers are required to file financial statements that have been audited and contain an auditor’s report. When an investment fund relies on a service organization, the controls at the service organization are relevant to the investment fund’s audit.” (para 89)

Commissioner Haber concluded that the companies had a reputable auditor in place, and one which had experience auditing with regard to crypto-currencies. The operational risks identified were concerned not with the structure or management of the Fund, but with the asset itself. There are other means of acquiring bitcoin, and a ban by the Commission would not protect investors from the “unfair, improper or fraudulent practices” from which it seeks to protect the public from. The denial of the fund would also fail to promote fair and efficient capital markets and confidence in those markets. Accordingly, he held that the Director’s decision was set aside and the receipt for the final prospectus was ordered to be issued.

(with a contribution from Ella Hantho)

Edited dashcam video sufficiently reliable

In Dainov v. Lee, the Ontario Superior Court of Justice granted a motion for summary judgement to dismiss the action involving a minor car crash based solely the evidence from one of the car’s dashboard cameras and the parties’ testimonies. In essence, the only important evidence concerning the accident came from the dashcam, and unless that evidence was inadmissible the defendant would succeed. Because it was admitted, the plaintiff’s case was dismissed.

To be admissible the video had to be a fair and accurate representation of the events, and not impair trial fairness. The plaintiff objected to the use of the video on the basis of three arguments: 1) the video was edited to show only the first 26 seconds leading up to the collision; 2) the original video and the dashboard camera were no longer available, and; 3) the limited field of view of the dashboard camera made the video unreliable.

The judge rejected all three of these reasons. Although only a short period before the accident was on the video, the portion edited out was from another location and was not relevant. Although the plaintiff argued that portion of the video would have shown the road conditions on the night, the parties did not disagree about that. The plaintiff did not explain why the original would be required, and had not suggested the video was doctored in any way. With respect to the third argument, any camera has a limited field of view, but the narrow field in this case was sufficient to demonstrate the positions and speed of the vehicles involved, and clearly indicated the defendant gave the plaintiff a reasonable opportunity to avoid a collision.

Without first-hand evidence or testimony about the reliability of how it was taken, Google mapping images are not admissible

The Third District Court of Appeal of Florida in City of Miami v. Juanita Kho reversed and remanded a trial court’s decision that accepted a Google Maps photograph (presumably taken by a Google Street View car) into evidence without proof of authentication. The Court of Appeal instead held that a photo taken from Google Street View will be subject to the same authentication requirements of any other photograph. In Florida, this is addressed in s. 90.901 of the Florida Evidence Code.

The underlying dispute had to do with a trip-and-fall accident in 2010 on a sidewalk. The plaintiff alleged that a difference of 1 ¼ inches between the sidewalk and a patch created a “dangerous and defective condition” resulting in the trip and subsequent fall. To prove her case, the plaintiff had to prove actual or constructive knowledge of the situation on the part of the defendant city.

The plaintiff had attempted to demonstrate this constructive knowledge by using a Google image of the sidewalk at issue, which was dated November 2007. The incident occurred in 2010.

The defendant objected to the admission of any internet photograph or map unless it was properly authenticated. During the hearing, the trial court said that a Google image is not “self-authenticating”, would be treated like any other photograph and thus require authentication by testimony.

The plaintiff’s expert testified that there was no substantial difference between the Google photograph and the photograph that was taken of the same location on the date of the plaintiff’s fall. Because the expert had not visited the location of the fall prior to 2010, he was unable to claim that the site of the fall three years before, when the Google image was taken, was in fact similar to the date of the fall. The plaintiff also did not introduce any testimony from a Google representative who may be able to speak to the image or how it was created.

At trial, the court overruled the defendant’s objection to admitting the Google photo as evidence without any additional testimony.

On appeal, the court strictly followed Section 90.901 of the Florida Evidence Code. The Court of Appeal stated at page 4 of its decision that “a Google Maps image must be authenticated in the same manner as any other photographic evidence before it is admitted in evidence”. This can be done using one of two methods:

1. the “pictorial testimony” method, which requires a witness to have personal knowledge and testify that the image fairly and accurately depicts the scene, or
2. the “silent witness” method, which requires that photograph “may be admitted upon proof of the reliability of the process which produced the tap or photo.

The Google image was not authenticated by either of these methods in the trial court, so the first ground was not satisfied by the plaintiff’s expert who had not attended the scene before 2010.

On appeal, the plaintiff argued that the photograph may be authenticated under the “silent witness” method. This was not successful, because the plaintiff “did not present any evidence as to the operating capabilities or condition of the equipment used by Google Maps”, or the procedures that Google Maps’ employees used to take the photograph, or any other relevant factors listed in precedents dealing with photos.

The Court of Appeal reversed the trial court’s decision and remanded the case with instructions that judgment be entered for the defendant.

Limited rights to discovery prior to certification did not lead to production of full database of affected users.

Once again, privacy class action lawyers have suffered a setback on their path to certification in privacy breach cases. This time, it’s in Karasik v. Yahoo Inc. where plaintiffs’ counsel were rebuffed in their attempt to obtain access to the database of all of Yahoo!’s sixteen million customers in Canada. The case is rooted in the infamous data breaches of 2013 – 2016 and a putative class action was commenced in 2016 in the Ontario courts. Prior to certification, the plaintiff brought a motion to compel the production of Yahoo’s Canadian user database.

The motion was made after Yahoo! Inc. delivered multiple affidavits on the hearing of the certification motion. One affidavit was a statement by Dr. Ghose, an expert in information systems. The expert had been given access to the database in order to prepare the affidavit. Prior to the cross-examination, the plaintiffs requested the production of the database and Yahoo! refused.

Prior to certification, production of records and evidence is generally limited. For documentary discovery at this stage, the judge required that the party making the request must demonstrate that the demand is proportionate, fair, and necessary for the certification motion.

In light of the significant evidence adduced by Yahoo! Inc, the judge concluded that the request was disproportionate to the needs of the certification motion. In addition, the plaintiffs did not successfully displace the burden of proving why the database would be necessary, other than a statement that the database would be necessary for cross-examination of the witness. There was no evidence to suggest that the database may be inaccurate. As a result, the plaintiffs’ motion was denied.

Records capable of being confirmatory evidence

A relatively novel use of cell phone tower evidence was at issue in R v Saleh. Evidence that an accused’s cell phone has pinged a particular cell phone tower is not infrequently used directly as evidence placing that accused in the vicinity of an offence, in order to show guilt. In Saleh, however, the evidence was used more indirectly.

The accused was charged with and convicted of first degree murder. Both the accused and the victim were involved in the drug trade, and the Crown argued that the accused had killed the victim because of a dispute related to that. Important evidence against the accused came from two witnesses also involved in the drug trade, one of whom had also been charged with the same murder but had made a deal in exchange for his testimony. Because these were two “unsavoury witnesses”, the trial judge was required to give the jury a Vetrovec caution about their evidence. That is, the judge was required to instruct the jury that they should show the greatest care and caution before accepting the evidence of those witnesses, unless there was confirmatory evidence of their testimony. That confirmatory evidence had to be independent and had to support or confirm material parts of the witnesses’ testimony.

The accused objected on appeal that the trial judge had wrongly pointed to cell phone tower evidence as capable of being such confirmatory evidence. The location of the shooting was a secluded rural area west of Ottawa, and records relating to cell phone use and cell phone towers placed the co-accused witness and the accused in relevant areas at relevant times. The accused argued that this should not be seen as confirmatory evidence: it was not material in respect of whether he was involved in the killing. Because those records were irrelevant to the question of his role in the killing, he argued, they did not meet the materiality requirement.

The Ontario Court of Appeal found no error in the trial judge’s reference to those records as satisfying the confirmatory purposes of a Vetrovec caution. Confirmatory evidence need not directly implicate the accused: it is sufficient that it be reasonably capable of strengthening the jury’s belief that the witness is telling the truth on an important aspect of the case. In this instance the cell phone and cell tower records confirmed the attendance of the accused and the witness at the remote site contemporaneous with the killing: that made it potentially material confirmatory evidence, and so the trial judge had not erred in pointing to that possible use of the evidence.

Section 714.1 application for witness to testify by video conference denied

With its decision in R v Musseau the Newfoundland and Labrador Provincial Court provided guidance on the use of section 714.1 of the Criminal Code, which allows a witness to “give evidence by means of technology that permits the witness to testify elsewhere in Canada in the virtual presence of the parties and the court”. Finding that the preconditions for such an order had not been met the trial judge dismissed the application, but provided guidance about the use of the section nonetheless.

The trial was to occur in Corner Brook Newfoundland, but the complainant was in New Brunswick and did not wish to appear personally, and so the Crown made a section 714.1 application concerning her appearance. The judge pointed out that the Crown’s application simply referred to the complainant in the case (the spouse of the accused, who was charged with assaulting her) appearing by “video conference”, without any description of the actual technology to be used. Section 714.1 requires that the technology make it possible for the witness to be in the “virtual presence” of the parties and the court, and without such evidence he was unable to assess the application. The judge noted that:

[20]…section 714.1 does not, as does section 714.3 of the Criminal Code (evidence of a witness by audio link), specifically require the Court to consider “any potential prejudice to either of the parties caused by the fact that the witness would not be seen by them”. This is a recognition of the significant difference between virtual testimony and audio testimony. Having said this, prejudice to an accused person’s ability to make full answer and defence must always be considered.

Further, the requirement for “virtual presence” was an important and meaningful one. An accused is entitled to face their accuser: although that was not to be taken literally, appearance by video technology had to accomplish the goals behind that principle. Video link technology was sufficiently sophisticated that that goal could be accomplished through its use. Indeed, the judge acknowledged at para 40 the possibility that “the use of video appearances are so common that the virtual presence requirement can be assumed or judicially acknowledged.” Nonetheless he decided that evidence concerning the nature of the technology actually to be used was necessary, and had to be provided either in writing as part of the application or in some other fashion.

In addition it was necessary to have evidence about the location from which the witness would be testifying, such as whether it would be a courtroom or something similar: the judge needed to be able to consider “whether the witness will face the same level of solemnity offered by a courtroom and whether he or she will be as free from outside influences while testifying as she or he would be if they were to testify in person before the trial judge” (para 34).

Summarizing the usefulness of section 714.1 (and quoting his own previous decision on the same issue in another case), the judge noted at para 42:

Section 714.1 of the Criminal Code is designed, in part, to lessen the financial cost and inconvenience caused by requiring witnesses to travel to testify in person to an area in which they do not reside. In a country as large as Canada, it would be foolhardy to stymie the use of such appearances. The reality is that modern technology has made the requirement for personal presence, in many instances, unnecessary and superfluous. Our courts must not ignore the reality of modern technological developments in assessing a demand for the personal presence of a witness. 

In this case, because of inadequacies in the form and content of the affidavits, and the failure therefore to establish the requirements of the section, the trial judge denied the application.

European Court determines that EU law does not provide for (but does not prohibit) global search result delisting

The European Court of Justice (Grand Chamber) has ruled that Google does not have to apply the right to be forgotten globally. The proceeding stemmed from an imposition by the French data protection authority (CNIL) of a EUR 100,000 fine against Google for not deindexing search results across all its search properties. Google had implemented geoblocking that prevented access to certain search results within the European Union, but CNIL found this to be insufficient.

The court proceeding, on the basis of the former Data Protection Directive, found:

64 It follows that, currently, there is no obligation under EU law, for a search engine operator who grants a request for de-referencing made by a data subject, as the case may be, following an injunction from a supervisory or judicial authority of a Member State, to carry out such a de-referencing on all the versions of its search engine.

Importantly, the court also noted that while EU law does not require orders with global effect, it does not prohibit them. Member states are free to adopt national laws that require global censorship.

Agreement will provide for mutual recognition of court orders and remove existing barriers

The United States Department of Justice and the United Kingdom Home Office have announced that the two countries have signed a bilateral agreement “On Access to Electronic Data for the Purpose of Countering Serious Crime”. The Agreement is intended to be a bilateral agreement of the type anticipated under the CLOUD Act. Passed in March 2018, partially to address the litigation against Microsoft related to evidence in Ireland, the CLOUD Act authorizes the United States to enter into executive agreements with other countries that meet specific criteria related to rule of law, civil rights and privacy. Once laid before Congress and approved, the result is to lift each party’s legal barriers that prevent one country’s legal processes from being recognized in the other. Many countries have been seeking an alternative to the traditional channels of mutual legal assistance, which are seen as time consuming and cumbersome. In many cases, US-service providers are prohibited from providing the content of communications except in response to a US court order, requiring the requesting law enforcement agency to go through MLAT and sometimes putting the service provider between a rock and a hard place.

On the UK side of the equation, changes were made in UK law to permit this under the Crime (Overseas Production Orders) Act 2019, which received Royal Assent in February 2019. The Agreement will enter into force following a six-month Congressional review period mandated by the CLOUD Act, and the related review by UK’s Parliament.

Australia has already announced that it is seeking its own CLOUD Act executive agreement, and Canada is rumoured to be in similar discussions.

It should be noted that initial reporting on this was conflated with recent declarations regarding encryption and access to cleartext of communications, so that some outlets reported that this agreement would require access to unencrypted communications, which is specifically excluded from the CLOUD Act.