Over the past year, I’ve worked with several clients that believed (Past tense) that SharePoint would replace file shares. Given the problems that organizations face with the uncontrollable growth of unstructured data and impacts on search usability, storage and operations, its a painful topic with no “magic bullet” solution.
Several articles have appeared regarding file-shares, saying they are dead and that SharePoint is the magic solution. Also, the big M marketing engine is sending the same message stating SharePoint can replace the need for file shares and thus solve the file share problem. What the articles and big M don’t address is the working involved in dealing with the files shares you have today. What many organizations have found after running proof of concepts is the the following problems:
- SharePoint stores the files in SQL (100GB limit per site collection) – not scalable, at least not like file shares
- How to move data from the file shares to SharePoint – labor intensive and requires tools and solid records mgmt guidance to be of value
- What to do with executables (Software packages)- you need file shares for this or distribution software
- How to tag the files so the search experience is a positive one – garbage in garbage out as they say, the user search experience will be negatively impacted
- The basic question of cost – file servers are cheaper than SharePoint and SQL Server farms
- Business disruption – who does the work? The business staff? Summer students?
What I generally recommend to clients is a new approach – thinking about data classification, storage and retrieval, lifecycle etc. For most IT professionals, Information Architecture and Data Life Cyleis unknown territory and so is enterprise content management – both complex topics especially in large multinational organizations. So how do file shares relate to the aforementioned topics? Organizations must think about information using a fined grained approach as opposed to the “big bucket” approach of file shares.
For example, information on file shares might consist of client information, product information, job related information, records or several other topics (Has value to the client either immediate and or long term). This information must be moved (redistributed) to purpose built repositories such as:
- Client knowledge repository
- Product information repository
- Job Function community
- Sales and Marketing repository
- Records Management repository
The same could be said of Public Folders with the difference being applications that take advantage of functionality provided by Public Folders. In cases where applications are involved a remediation methodology and tools must be adopted to inventory, assess, re-write, test and deploy applications.
So organizations move from one big-bucket of information (that lacks a usable architecture) that isn’t really usable to an information architecture that is highly usable – assuming content management disciplines are in place.
Read my article on Windows IT Pro that covers this subject in more detail.