20120208 ARMA Detroit Social Media Governance

Editor's Notes

• #10: The first step is to determine whether or not something is in fact a record. Just as we know that most email messages are not records, for most organizations their Facebook fan page updates will not be records either. In other words, we have to ask the same questions about these tools that we’d ask about any other type of information:Does it document a transaction or a decision? If it does, it’s probably a record. Is it captured in another form? This is the biggest reason why most social networking sites like Facebook and Twitter wouldn’t need to be captured as records – in most cases they are being used as another transmission mechanism for information stored elsewhere. Now, just because it isn’t a record doesn’t mean it couldn’t be discoverable or a public record and subject to FOIA-type laws. Again, same considerations here as for other types of information. [twitter]Determine whether something is a record or not according to its content and context.[/twitter]
• #17: During my professional lifetime, I have seen at least 4 major enterprise IT transformations, and they seem to be occurring with increasing acceleration. When I first came into the workforce, the enterprise IT norm was centered on mainframe computers focused on batch-processed financial applications. This was the era of Burroughs and Univac and NCR and Control Data and Honeywell. This era was soon eclipsed by the rise of minicomputers.Minis were themselves eclipsed by the PC revolution, stitched together in Local Area Networks. Steroids in the form of the internet changed everything about how we connected PCs together distributed documents and information around our organizations. And then along came Google and our expectations about enterprise IT and simplicity of use morphed once again.
• #18: The challenges here are enormous. Expectations of Enterprise IT are rising. The business, still reeling from the crash of 2008, is questioning the rigidity and cost of legacy systems. The focus of IT is changing from a traditional focus on standardizing and automating back-end manual processes – a focus on CONTROL – to a focus on empowering and connecting knowledge workers and improving knowledge worker productivity and innovation. in the world of Systems of Engagement – no one on the user side cares about any of this. However, because these systems are being used by enterprises, they will inevitably be subject to the same legal and social restrictions as traditional enterprise content, and therein lies the rub. Today that rub is significantly limiting endorsement and adoption of consumer-style communication and collaboration facilities around the world, and it will continue to do so until the content management industry and its customers develop protocols and policies to address its issues.
• #23: Pretty straightforward
• #24: Official vs. unofficial includesDisclaimers (this is or is not official; disclaimer of responsibility if it isn’t)Also includes a link to his social media policy
• #25: Whether approval is required to create an account (official only)It’s also useful, as CSU does, to list all the official accounts somewhere on the website.
• #26: This includes things like:What user names are appropriate, and whether to use the organization as part of it (e.g. Dell_JeffW)Pictures – same thingBio – same thing, plus things like official account, name (and sometimes personal Twitter handle) of the person behind the account, etc. Different types of contact informationIt’s also valuable to have guidelines for what types of contacts are appropriate. An official federal government account could “friend” Barack Obama on Twitter, but probably shouldn’t friend his re-election campaign or the Democratic Party (and even if it did the Republican Party as well, it’s still problematic). Similarly, it might look a bit odd for an energy company account to “friend” a parody account like BPGlobalPR, or a competitor, or an unsavory group, etc.
• #27: Pretty straightforward here. Three main points:If third party content is allowed, it should be reviewed so people don’t upload pornography, etc. If it is reviewed, the organization may have some responsibility to remove things that are inappropriate. This should be spelled out clearly and adhered to rigorously – all goes back to transparency. If an official account “likes” something on Facebook, or retweets something on Twitter, this could be considered approval or even recommendation – and if it’s something offensive, or illegal, or otherwise inappropriate, this could cause serious issues.
• #29: The policy should outline what types of groups are appropriate and what types of groups should be out of bounds. This is especially important for official commercial accounts but could be applicable even to personal accounts where the connection could be made to the organization because of the employee’s visibility. For example, it would be inappropriate for an official in charge of elections to be a member of a Facebook group focused on reelecting one candidate or another. Moreover, there are any number of groups dedicated to patently offensive or illegal causes; having accounts associated with these types of groups could bring significant risk to the organization and its brand. ~Another related area involves conveying a perception of approval of content that might be controversial, offensive, or illegal. For example, both a Facebook “like” and retweeting content on Twitter are often perceived as approval of that content. If an official account or the personal account of a senior manager retweets a sexist joke or something that condones illegal drug use, that could also cause serious issues for the organization.
• #31: These are specific to government. It’s always a good idea to link back to the organization’s home website and vice versa so it’s clear that the account is an official one. Because of public records and sunshine laws, it’s important for the agency to be open about whether comments are allowed or monitored and whether it believes them to be covered under such legislation. And for public safety accounts in particular, such as fire departments or police, the account should note whether it’s monitored and what the “official” mechanisms are to report safety issues.
• #33: This is an example of a guideline for how to engage those that comment on your social media, and those that post or comment on third party sites. This triage chart from the American Society of Chemical Engineers is not for every user in the organization, but it can be quite useful for those responsible for monitoring and engaging comments about the organization such as public affairs.
• #34: Here’s an example of this from the Seattle Fire Dept – it clearly says “This site is not monitored. Call 911 for emergencies.” It also notes the applicability of public records laws and has a link to the main website.
• #36: The first step is to determine whether or not something is in fact a record. Just as we know that most email messages are not records, for most organizations their Facebook fan page updates will not be records either. In other words, we have to ask the same questions about these tools that we’d ask about any other type of information:Does it document a transaction or a decision? If it does, it’s probably a record. Is it captured in another form? This is the biggest reason why most social networking sites like Facebook and Twitter wouldn’t need to be captured as records – in most cases they are being used as another transmission mechanism for information stored elsewhere. Now, just because it isn’t a record doesn’t mean it couldn’t be discoverable or a public record and subject to FOIA-type laws. Again, same considerations here as for other types of information. [twitter]Determine whether something is a record or not according to its content and context.[/twitter]
• #38: Prepare for discovery. This means having the same type of data map you have in place inside the organization, but with listings of all the services you use, the accounts used there, etc. At a minimum you should list any official use of services and official accounts. It also means understanding the process for getting at that information in the event of litigation, FOIA request, etc. The time to put that process in place is before the subpoena is received. For hosted tools, such as FB or Twitter, it may mean taking periodic snapshots of what is posted to them. Right now there aren’t a lot of tools that do this; one way that can be effective is to capture the RSS feeds generated by these tools. As updates are made, they are published through the RSS feed, which can be saved locally. It might also require working with the third-party vendor in the event that some information or some updates are not available through RSS – for example, web-based email. It’s also important to note that at least for commercial solutions there is very little ability to put or enforce legal holds or to prevent a user from deleting an account, at least without a subpoena and without doing it before the user knows to delete it. [twitter]Prepare for discovery in advance, including listing official use of services and accounts.[/twitter]
• #39: As we just noted, the records management or communications policies (or both) should address the use of these tools. We’ll look at some examples of policies over the next few slides. At a minimum, the policy should address: Identity, relationship, and transparency – is the account official or unofficial?Security, confidentiality, and sensitive informationComments and responses to commentsResponding to others’ posts on commercial sitesAccuracy and ethicsMonitoring and auditing[twitter]Address these tools in the records or communications policies (or both). [/twitter]
• #40: How to capture content will depend first on one key variable: where is it stored? The vast majority of social media sites are either hosted solutions or commercial ones – that is, users’ data is not stored inside the organization’s firewall, but on some third party data center outside the control of the organization. This can present a significant issue because how and how long the data is stored is almost entirely dependent on the site’s Terms of Service. If the Terms are changed from retention for 7 years to retention for 2 weeks, or to permanently, it presents a real problem for the records program. And depending on the site this is almost certainly non-negotiable with the exception of governmental entities with the force of the law behind them. ~It’s also the case that some regulatory regimes have geographical aspects that, for example, require particular content to be stored inside a country’s boundaries, or prohibit it from being stored in certain countries’ boundaries. This can present issues for services that store information in the cloud and that might have numerous data centers spread around the world. ~Once the organization has determined what to capture, the next step then is to save that information locally. This is almost always a copy – in other words, saving social content from Facebook does *not* delete it from there. There are a number of ways to accomplish this that we will review over the rest of this module.
• #47: Finally, there are enterprise versions of every Web 2.0 application. These enterprise versions are often available to be hosted inside the firewall, meaning that security is much more robust. Access can be secured to them much more effectively. They can be integrated into the organization’s identity infrastructure – whether Active Directory or something else – such that any change, post, comment, edit, update, etc. can all be tracked and, more importantly, tracked to a specific named user. No anonymous postings here. Of course, you have to pay for an enterprise version, but what you’re really paying for is a level of peace of mind. And you still get many of the same benefits – ease of use, familiarity with the type of tool, rapid and agile collaboration across geographical and time boundaries, etc. You’re just getting a more secure and robust version of it. [twitter]Consider implementing enterprise versions. FB is FB, but internal tools might be more appropriate.[/twitter]
• #49: At this point I’d be pleased to entertain your questions.