It can. 

The NYT this morning tells us that "Big News Media Join in Push to Limit Use of Unidentified Sources."  Readers are told:

"Concerned that they may have become too free in granting anonymity to sources, news organizations including USA Today, The Washington Post, The Los Angeles Times, NBC News and The New York Times are trying to throttle back their use.
     "But some journalists worry that these efforts could hamper them from doing their jobs - coming in a hothouse atmosphere where mistrust of the news media is rampant, hordes of newly minted media critics attack every misstep on the Web, and legal cases jeopardize their ability to keep unnamed news sources confidential....
     "Last year, The New York Times adopted a more stringent approach to its treatment of confidential sources, including a provision that the identity of every unidentified source must be known to at least one editor. A committee of the paper's journalists recently recommended that the top editors put in place new editing mechanisms to ensure that current policies are enforced more fully and energetically."

We look forward to these "new editing mechanisms."

Yes, policies on unnamed sources should be made, those policies should be clear and everyone in the newsroom should know what they are.  But more often (as in "every day"), editors must know the sources -- indeed, all sources -- are for a story, how to reach those souces and how to verify what the reporter wrote, even if the reporter is out-of-pocket. 

This is not difficult if journalists recognize that a PC-based word processing application already has the tools to assist in this "Who Are The Sources" mission. (If the publication is still using something like the old Coyote terminals, sorry, we probably can't help  you.) 

The tool is the "comment" function in the word processor.  While the newsroom is making policies about sourcing, add this one: "Every paragraph of every story will end with an embedded comment.  That comment will show editors exactly how the reporter knows what he or she just wrote."  The comment might include a source's name, phone number and date-time-place of interview.  The comment might include a URL or a bibliographic citation.  It might include reference to the specific reporter's notebook.  But in the end, the comments should be sufficient that an editor can "walk the cat backward" to determine exactly how the reporter knows what he/she just wrote.  Doing so helps prevent unwarranted assumptions and errors of fact, if not interpretation.

There will be those of the Burn-Your-Notes School of libel defense who will contend this is comment thing is suicidal.  We would suggest, first, that very few stories ever become court cases.  Secondly remember that truth is the first defense in libel actions, and it is our responsibility to deliver that truth.