When I just started to learn computer programming (Some 30 years ago) one of the first issues you learn is the importance of giving good meaningful names to variables, functions and objects in general.  

 I often see in organizations complete mess considering objects names (nodes, devices, networks). Good naming conventions affects management; it makes it easier to know the functionality, position or any other important aspect of the object, this of course depends on the structure of the organization, i.e. multi-sites organization can include site reference in objects name so it is simpler to locate them. And today with the growing popularity of virtual machines this becomes even more important. Naming conventions is an important issue; some will say names with meaning are security issue, other that it interferes with the organization hardware deployment (check: http://www.techrepublic.com/article/determining-a-good-naming-convention-for-your-network/6132500 for an interesting article, , it is a decade old article, but I believe still relevant). I believe the benefits outgrown the disadvantages but it is an issue to consider.

When you use good naming conventions to name your objects you automatically get the ability to manage alerts relating to the objects. Good naming conventions means you are able to use wildcards in filtering and routing rules, resulting in fewer rules which are much more readable and easy to manage and maintain. SNS++ has the ability to support objects lists which can help do the same even if your objects’ names are with no particular format, but it is much simpler if they are. So if for an example all your database servers’ names start with dbsrv  (dbsrv‑accounting, dvsrv01, dbsrv-ny) prefix you can define a rule relating to all database server just by specifying dbsrv*.

To summarize, Notification Management is one more important factor to take into consideration when selecting name conventions to your objects.