In my
previous post I talked about the "recordness" of electronic records as it relates to aggregation of content. Another area that is often raised as an issue deals with the appearance of a record. In this post I will examine some of those issues.
In the electronic world it has long been the case that documents might look slightly different depending on which version of Word was used to create it vs. open it. And we have had some challenges in rendering content across platforms for decades - Word vs. WordPerfect, Office for Windows vs. Office for Mac, etc. But you knew to some extent what application and version created it, and if you had that application and version you'd see the same thing.
Then came email and the ability to create RTF and HTML versions of messages, some with embedded graphics, and even some with embedded code and the problem of viewing the content slightly differently became worse. Users who used simpler email clients, or text-only clients, might see a message substantially different in appearance from someone who used the latest version of the Outlook or Notes client.
Enter the World Wide Web. After I got out of the Marines I spent some time as a starving actor/pc tech/Web developer. Every browser rendered differently, and in many cases ever version of the browser, such that every website had to have a script running to determine what browser the user was running and send them to /index.html, /index.htm, /NN30index.html, /IE50index.htm, etc. Sometimes the differences were subtle - a pixel off here, an indent or lack thereof there. Sometimes they were not so subtle, though - today, woe betide the iPhone user who tries to access a Flash-heavy site (because iPhone doesn't support Flash).
As content authoring and management applications became more sophisticated, presentation and display became even more complex. Today it is relatively simple to manage content in such a fashion as to be able to apply security not just to the document, but to parts of the document, redaction layers, and even metadata fields. For example, in an electronic health record application, a doctor might have access to the entire content of the document, while the phlebotomist might only have access to some of the document and some of the metadata fields, and the accounting staff would have access to almost none of it. The document is the same, but what each user sees is dramatically different.
What this means is that today the organization may have little control over how information is displayed except in very gross terms. No amount of site optimization and user agent checking is going to account for every single possible iteration of how content is consumed - in fact, a growing proportion of data consumption is done without a user interface, through an API by another application. In the brave new world of mashups, and smartphones, who is to say how data will be displayed - and what does that mean to the "recordness" of the record?
Some organizations have taken the approach of capturing a screenshot of what the user sees, but I think this is problematic for a number of reasons, not least of which is keeping on top of the myriad combinations of browser, device, screen resolution, etc. And for those organizations that print the captured screenshot and file it, there are even more problems.
A related approach is often recommended for those applications that are primarily combinations of database and template. In this approach the organization retains the database itself and every version of every template for some lengthy period of time "just in case".
But I think the solution is a bit simpler. Unless the appearance makes a concrete difference in how the record is understood, the content should be considered the record immaterial of how it is displayed. This is consistent with how most organizations deal with their myriad file formats but in the absence of legal guidance and with the existing high cost of discovery it makes sense to call this out in the organization's information management policy. It would also be helpful for someone like the Sedona Conference to make this recommendation.
This sounds simplistic: just keep doing what you're doing and you'll be OK. That's close to what I'm advocating, but not exactly. By being specific about the notion that some files will display differently in different browsers, devices, etc. the organization is being a bit more proactive. This means that the organization would not feel obligated to capture screenshots as long as there is another way to capture the various records such as by putting them in an ERMS.
#ElectronicRecordsManagement