I don’t know how many of the readers out there have gotten to deal with the joy of document preservation, but let me tell you, it is not a lot of fun.
The process at Columbia starts off with either the General Counsel’s (GC) office receiving a notice from a lawyer or the University deciding to enter into a lawsuit. At that point, I usually get an email from one of the 30 or so lawyers in our GC office with a list of names of the people involved in the litigation.
Sometimes, I receive a copy of a subpoena, if it is an outside party suing us; often it is a description of the case with a list of the things that need to be preserved. Here is a sample of the laundry list of items that have appeared in these holds:
The “materials” I am referring to include traditional paper documents, electronic data or other ESI (including but not limited to audio, video, email, email attachments, memos, notes, spreadsheets, and discs) and all information preserved in any other tangible form.
Lawyers refer to these materials as “documents, data, and tangible things.” “Documents, data, and tangible things” need to be interpreted in the broadest possible sense.
Now, this would not be so bad if the cases were limited in scope, meaning either a reasonable number of people or a reasonable length of time, but the reality is that they are often neither. I have gotten requests with over 200 people on the preserve list, and have cases that are still on ongoing hold from the 90’s.
At the risk of giving away some of my dirty little secrets, it is very possible that doing these requests in a large decentralized university may not be as neat and clean as at a large centralized company.
For one, it is very likely that on any list of more than 2 employees, more than one mail system will be involved. To make it even more deluxe, people at the University tend to move around a lot, so it is totally possible for many of the individuals to have their mail on more than one system.
I know that this is a problem that is unique to me (joke). Here is a bit of information you may find useful. Make sure that in any of your systems that may contain data that can be subject to a preservation notice, the process used to delete data is fully automatic.
The worst case possible would be a mail system in which you leave old mail for deleted users, and randomly delete files when you are running low on space. Murphy ’s Law will guarantee that as soon as you delete a bunch of old expired accounts, a lawsuit will come in for one of more of them, and because you manually deleted the files, you will automatically be in trouble.
One of the little gotchas in data preservation is that you are supposed to start preserving the data when you think that there may be a lawsuit, so by deleting that data, you have willfully destroyed evidence. If the data was deleted by a documented automated process, there is no harm and no foul.
Another annoying feature of this Data Discovery process is that because of the timeframes involved, it is totally possible that the mail systems used when the case started have been replaced, sometimes more than once, before the case closes.
I have many cases of tapes stored in Iron Mountain that are the backups of mail servers that have not been around for many years. The really bad news is that the hardware to read those tapes left our datacenter almost ten years ago. I’m not exactly sure what I would have to do to retrieve any of those mail files, but I am sure it will be very expensive.
I do not know what I would do to fix this process. It appears that we are living in a society that is increasingly litigious and obsessed with creating and saving data. These two (bad) habits have created a monster, and we poor IT people have to deal with the wild side effects.