The article makes a point about some software tools not having tamper protection built in. I know that this is a current issue regarding evidence, particularly digital evidence. However, the drive to ideally preserve evidence can be taken too far - real world practicalities must also be acknowledged and accommodated or else the evidential system, and therefore justice, will suffer in the end.
Aside: Please do not mistake my point - I am not against establishing standard operating procedures and best practices for preserving evidence, performing examinations, and the like. What I am against is establishing overly idealistic expectations that are not achievable in the real world across the myriad law enforcement and justice agencies. Put another way, I am for a reasonable balance that is biased toward continually improving the system over time.To return to the main thrust of this post, cell phone data is not just any run of the mill evidence, it is "scientific evidence", so someone acting as an examiner needs to recover and analyze the data and then present the results. The article helpfully provides a link to a draft NIST (The National Institute of Standards and Technology, a US government agency) recommendation titled Guidelines on Cell Phone Forensics.
The Wired article, at least to my reading, gives the impression that if a tool does not have built-in digital signature protection that it is somehow completely suspect. I don't think that is the case. There are ways to adjust operating procedures to accommodate this, such as MD5 hash generation software routines and proper (in the British sense of the word) evidence handling procedures. I think it is a good idea to have protection built-in, but that it is likewise a bad idea to automatically assume that if a tool that is used in an investigation doesn't have digital signature features built in that the evidence was likely tampered with. That sounds blindingly obvious when approached in this manner, but may not be so obvious to a jury or the general public.
Phone forensics is a helpful tool and can provide valuable clues that would not be otherwise available. But like all scientific evidence, it must be handled, analyzed, and presented properly, and then taken into account along with other evidence, to be of use to investigators and the court.