Recently, I took the leap from Yahoo! and added a Gmail Account.
For a long time, I thought, “What can be the difference? E-mail is e-mail.” Further, I thought people were just switching because it was the latest fad, and they wanted to be associated with the then-upcoming Google versus the troubled Yahoo!
While this may be partly true, there are some tangible advantages to Gmail. Gmail has a better interface than Yahoo!—it provides one look and feel while Yahoo! has a switching mechanism between the legacy email and a new Yahoo! mail, which is still kind of quirky. Gmail better integrates other tools like instant messaging and VOIP. Gmail offers a huge amount of storage. Gmail associates email strings so you can easily expand or click through the chain.
And finally, Gmail has a label structure for emails versus Yahoo’s folder structure. This is the one that matters most.
The label structure is superior to the folders. You can have multiple labels for an e-mail and can therefore locate items of interest much more easily by checking in any of the pertinent label categories. In contrast, in the Yahoo! folder structure, you can only store the e-mail in one folder, period. This makes it it difficult to store, connect, and discover items that cross categories.
For example, if you have e-mails on enterprise architecture topics from a particular source, you may want to label it by the topic EA and by the source it came from, so in the future you can find it by topic or by source.
Reflecting on this archiving structure from an enterprise architecture perspective, it became apparent to me that the legacy folder structure used in Yahoo! mail and the typical Microsoft Office applications such as Outlook and My Documents is built according to a typical taxonomy structure. By this I mean that here are one “parent” to multiple “children” relationships (i.e. a folder has one or more files/emails, but a file/email is constrained to only one folder).
However, in Gmail, the archiving structure is built according to an ontology structure, where there are multiple relationships between objects, so that there is a many-to-many relationship. (i.e. a label category can have multiple files/emails and files/emails can be tagged to many labels)—a much more efficient and expansive metadata structure.
So in short, the analogy goes like this--
Folder structure : Taxonomy : : Labels : Ontology
And Google wins in e-mail archiving hands down!
In enterprise architecture, the implications are enormous. For example, Microsoft, which is the defacto standard in most of our organizations, rules the way we store files in the legacy folder structure. Perhaps, the time has come for us to evolve to the superior metadata structure using labeling. This will make it far easier and more productive for the average user to search and discover information they need.
Further, metadata is at the heart of enterprise architecture, where we seek to break down the siloes in and between our organizations and make for better interoperability and information sharing. The goal is a holistic view of what’s going on in our organization and between organizations, and the only way to achieve that from an IT perspective is to label information so that it is discoverable and usable outside stereotypical stovepipes.