Cloud Architectures and Cultural Memory - Challenges of Information Management Beyond the Cloud - page 5

3 Access to Content: Open Data and Computational Capacity

Now let me talk a bit about how Clouds change the management and access of the
cultural record and then we will move on to what they are doing to the record itself
for the remainder of our time.
Access to digital information has always implicitly required and assumed that a
certain amount of computational power is being made available for search, navigation
and display of content. Historically when we thought about doing things like reading
journal articles online, reading a newspaper online, reading a topic online, these
access activities took place at human speed, at the speed of eyes, at the speed of the
human reader flipping virtual pages; the cost of the actual amount of computation you
had to provide to support these activities per user was (at least in recent years)
relatively small and steadily diminishing over time as we happily rode the curve of
Moore's Law, computing getting cheaper every year. (There was also some cost for
searching, for discovery and navigation, but that was also computationally reasonable,
at least for the constrained search and navigation options offered by typical sites
oriented towards reading or viewing.) The notion of being able to bring up Web-based
services that would support vast numbers of concurrent readers was pretty
manageable economically. Yes, there was cost associated with it, there was
engineering associated with it, but this sort of cost per eyeball was a tractable kind of
thing you could write off through advertising revenues if you were a commercial
entity, you could write it off within an operating budget if you were a government or
not-for-profit entity. (To be sure, streamed media - audio or video - is substantially
more challenging and more costly, though clearly not out of reach, based on the
amount of this material available on the network either free or through advertising
support.)
I think you are seeing around the world (certainly in the United States) a
substantial amount of policy emphasis on opening up access to government
information, whether it's textual, numeric, or imagery, or whatever, and also opening
up access to both writings and underlying data that are the result of government-
funded research. These developments are a key part of the whole open access
movement, and indeed even broader trends towards openness in scholarship. A lot of
the focus inside academia in the last few years has been around scholarly articles and
what this means for journal publishing, and around, particularly in recent years, the
sharing of underlying research data as well, when these are the result of federally-
funded research. And there are additional, compelling, arguments advanced in various
contexts that argue for open access to scholarly publications and scholarly data
independent of funding source. In May 2013 in the United States you saw an
executive order from the President - which doesn't seem to have gotten nearly as
much press as the discussion of open access for research - basically telling all of the
executive branch agencies in the United States that they should be planning, going
forward, for making all of their data publicly available, unless there was some good
reason not to (confidentiality and privacy, security, something like that) but the
default action would be they should be designing systems that open these data assets
up to the public. There are also very broad movements urging cultural memory