Former academic archivist here. I can provide at least a partial answer, excepting the question about social media accounts. I may not be directly answering your question, because I'm referring to materials that are placed into research collections, rather than those that exist outside of research collections (e.g. Facebook messages...not typically something that is included).

Much of what you have asked about is already in practice for institutions that collect materials pertaining to "important individuals." The main challenge of connecting researchers to these materials is not the question of privacy or ethics, but rather cost, time, and difficulty in providing access to electronic formats.

Usually, there is a link between an institution and the individual, based on an existing relationship - someone who was a professor, or a political figure who received a degree from the school, etc. When it's determined that an archive can be formalized from an existing collection of materials, an agreement is made between the institution and the estate or owner of the materials. Sometimes that happens years in advance of actually transferring the materials to the institution.

The determination that an archive "exists" can be made by the institution, the individual, the family, the estate, government regulations, institutional policies, etc (examples upon request). The decision about the scope of the collection comes into play at this time, and that's when things like emails and hard drives are considered.

It's often at that point that someone with a background in archival practice will evaluate the materials, but after the agreement is signed, the materials are accessioned into the institution's collections and reviewed by an archivist. The collection is given a formal name, unique ID, etc. A "finding aid" will be written, which describes the nature and arrangement of the collection, and other information (access policy, historical notes, etc).

At some point, the institution may open the collection for research, since this is typically the goal for collecting the materials in the first place. Archivists will examine how to provide access to the materials by patrons. In doing so, they need to determine things like

a) whether to digitize paper materials and how to disseminate them to patrons

b) how to transfer electronic materials into accessible formats

Many institutions are currently collecting hard drives or copying hard drives as part of the accessioning process. Archivists sometimes have a rule of thumb regarding how long it might take to process a standard document box of papers - call it 10 hours. Consider for a moment the time it would take to process the contents of a hard drive, which might translate into dozens or hundreds of physical boxes. There may be a fair amount of "noise" (files that are not pertinent to the scope of the collection), files that are protected by law (student records, personal identifying information, materials that are protected from release by their copyright status), or files that need to be sanitized. Bottom line, processing these takes a LOT of time. And money.

For that reason, there are a whole bunch of hard drives sitting in the archives of the world's most prominent research institutions right now. The longer they sit, the harder it becomes to process them. I've dealt with collections that involve laser discs, zip disks, floppy disks, U-Matic video cassettes, and a host of other formats that no one on Reddit under 35 has ever touched.

Questions about privacy are already addressed in law, policy, and the agreements that archives make when acquiring collections. Collection processing acts as a check point for ensuring that materials released for research fall within the bounds of law and policy before access is granted to patrons.

However, I would be interested to hear if anyone can address the question of social media, particularly as it applies to deceased individuals.