Some questions answered
So last week The Legal Genealogist wrote about the search warrant issued by a local judge in the State of Florida for access to the GEDmatch database in a criminal case.
And, at that time, basically all that was known was that a warrant had been issued; a set of Q&A exchanges showed just how much we didn’t know.1
In the interim, a link to a copy of the warrant — heavily redacted — was posted on Twitter.2
So we now know a bit more and can revisit a few of those questions and answers:3
Q. What crime was committed that gave rise to this investigation?
A. Sexual battery, armed kidnapping, armed burglary, described in the warrant as “serial rape.”
Q. When did the crime occur?
A. We still don’t know.
Q. What traditional police methods were used before turning to the private data of people who tested solely for genealogical research?
A. We don’t know everything but we do know that the police checked the Florida databases for the Department of Corrections and Department of Law Enforcement, CODIS and databases of cooperating international agencies, and found no matches.
Q. Had the police accessed the GEDmatch database before its users were allowed to choose whether to opt in or opt out?
Q. What was the basis for the warrant issued by the Florida judge?
A. We don’t know the probable cause as to the crimes — that part of the warrant was redacted. As far as the detail given about the GEDmatch data, the affidavit in support of the warrant said there was matching data that police had seen before GEDmatch changed to an opt-in system. The judge was told that there were nine matches at the second or third cousin level and that the chances of solving the case by genetic genealogy were only medium because “the number of unique, potentially informative matches is small.” We don’t know how many of those matches had opted in, if any.
Q. What facts were given to the judge to decide whether to issue the warrant?
A. Among others, that “Members of the (police) genetic genealogy team had access to this report prior to May 19, 2019 and witnessed the report. This report is now blocked by GEDmatch. The data is present in the system. The public has access to the report, but law enforcement does not.” That the data sought was “open for public viewing, but not open for law enforcement viewing…” That “GEDmatch is extremely popular, highly regarded, and provides tools your Affiant can’t get elsewhere. All of the above information was available to your Affiant until May 19, 2019. It is unfortunate, but this same information is available to the public without restrictions…” That the police officer “believes that the GEDmatch data will contain information that will aid in establishing the nature of the relationship between the defendant and family members. Your Affiant believes that GEDmatch could contain sufficient reliable, competent, and admissible evidence in order to prove beyond a reasonable doubt, the participation of the defendant in the aforementioned (crimes).”
Q. Was the judge aware of the privacy settings of the users impacted?
A. Mostly by inference. Since the officers had seen the matches before the change and they were not there after the change, the kits must not have opted in.
Q. Was the warrant specific, asking for information as to specific users? Or general, asking for access to all kits no matter what?
A. The warrant sought information about all one-to-one matches to the law enforcement kit including GEDmatch kit number, email address, real name and alias, date and time stamp of profile creation date, most recent log-in, registered mobile phone number, many to one report, “Predict your eye colour” report, all ethnicity reports, all GEDCOMs, all matches, chip version, gender of tester, haplogroup(s), autosomal data (total cM, largest segment, generation), X-DNA (total cM, largest segment), matching segment report for all matches, triangulation report, relationship predictor, name (name and alias) and email of those disclosed in the process. And though the warrant asks for all one-to-one data, it’s not been publicly confirmed whether that included research kits or only those public kits not opted in.
There’s still much that isn’t known, of course. Would the judge have chosen otherwise if specifically told that the testers had chosen not to opt in? Or if the police hadn’t seen the database before it was changed to opt-in? Or if the judge had been told more accurately that the public can’t see GEDmatch matches or data — only the people who choose to contribute their own data in the hopes of matching other for family history purposes? We don’t know.
So we still don’t know the full ramifications of the warrant for other cases, and particularly for cases where there are no close matches among the opted-in users or for research kits not opted in to any public matching.
But all of these factors should be taken into consideration in choosing whether to participate — or continue to participate — at GEDmatch.
Cite/link to this post: Judy G. Russell, “More on that warrant…,” The Legal Genealogist (https://www.legalgenealogist.com/blog : posted 17 Nov 2019).
- See Judy G. Russell, “About that warrant…,” The Legal Genealogist, posted 10 Nov 2019 (https://www.legalgenealogist.com/blog : accessed 17 Nov 2019). ↩
- See PDF copy, search warrant, Orange County, Florida, 14 June 2019, uploaded to DocumentCloud by Twitter posted Shreyas Gandlur (https://assets.documentcloud.org/ : accessed 17 Nov 2019). ↩
- All quotes are from the warrant. ↩