In Broadcast - November 2015 - 48


48

www.inbroadcast.com | Vol: 5 - Issue 11 | November 2015

InMedia

Inaction Spells
Doom For Broadcast
Archives
The clock is ticking on content libraries
around the world which have not yet been
transferred to file-based storage, writes
Christophe Kummer, NOA, Managing Director...

M

any broadcasters and media
companies are still sitting on
enormous archives held on carriers
that are obsolete or soon to become
obsolete. The timescales involved
in addressing this issue are such
that delay now will create serious
difficulties in securing content assets
for the future.
Any media organisation with more
than 100,000 video tapes to preserve
has a narrow window of opportunity
while the machines are still supported,
and there are still skilled staff to
operate them. With each year, finding
the staff, spares and maintenance will
become harder.
With Sony's announcement of the
discontinuation of IMX machines, it
makes sense to devote the remaining
working life of those machines to
securing the migration toward filebased storage. Leaving the problem
to be tackled in three or four years is
leaving it too late.
But amazingly, inaction is all too
common. Reliable estimates are that
between 40% and 50% of media
organisations using tape machines
have not yet started to digitise their
assets. This amounts to perhaps 100
million hours of content worldwide.
Migration WIndow
For
the
major
broadcast
organisations to achieve the safe
migration of their content libraries
before the machines are no longer
supported, each will require around
twenty parallel lines to migrate from
all forms of tape to file-based storage.
And it's not a problem that can be
easily solved by outsourcing: most
broadcasters want to ensure access to
their content, and shipping it offsite to

a distant third-party archiving facility
involves unacceptably long access
times to travelling content, and the
complications of insuring the assets.
By contrast, a facility on-site or
nearby makes it possible to limit
downtime to little more than the
length of the content itself. Costs are
lower in-house too, at around €10
to €15 per content-hour, while the
additional costs of sending it to a third
party may drive that up towards €30
to €40.
Content Savings
But why is there such a lack of
action? File-based production is
common enough, and the savings and
efficiencies there are well understood.
With content production costs being
anything from a few thousand to
several million Euro per carrier,
the cost of preserving that content
securely for the future is minute in
comparison.
Partly a natural tendency to
procrastinate is to blame, but there
are also debates within the archiving
world about the best approach to
migrating broadcast content libraries.
There is a common argument that
all content committed to an archive
must be broadcast ready, but this
implies the detailed verification of
the existing copies of the content,
a large proportion of which has
probably been previously migrated
from the previous original (i.e. a 1"B
or a Betacam) to a digital tape format
(IMX, DigiBeta). Without being certain
that these previous migrations had not
introduced errors, some broadcasters
are reluctant to take the next step and
migrate away from tape altogether.
But the complete process of making a

Is your video archive system obsolete?

broadcast-ready archive implies visual
inspection of all the content in realtime during the transfer - prohibitively
expensive and time-consuming.
It makes better sense to use the
remaining lifespan of the tape
machines to secure the preservation
of the content as the urgent priority,
with the emphasis on being totally sure
that the migration process toward filebased storage does not introduce any
new errors in itself.
Status Reports
This can be achieved by using an
archiving system with good low-level
status reporting at the machine
level during the migration process.
The question of errors inherited
from previous migrations can be
addressed as a separate issue, either
as part of the archiving process, or as
a subsequent operation.
The percentage of existing errors
in content libraries depends on the
care with which the library has been
curated: it could be as low as 2% in a
European broadcaster's library...or
much higher in some other parts of the
world.
For those who want to integrate
QC for broadcast-readiness into

their migration, and can support the
additional cost, it's easy enough to
do. NOA's archiving workflow allows
for inclusion of third party algorithmic
content
analysis
that
detects
previously-generated artifacts, and
triggers intervention to allow archive
staff to go back to previous copies and
attempt to make a better transfer from
them - although that's only possible if
the earlier recordings still exist.
Archive Readiness
This introduces an element of
intellectual assessment into the
archiving operation, so it demands
greater resources and more time,
perhaps doubling the cost. But
crucially it doesn't require staff
to monitor the content real-time
during the transfers, so the cost is far
lower than that method of ensuring
broadcast readiness. Alternatively,
there are restoration algorithms that
can be applied as a later process.
It's critical though to act swiftly in
the limited time available before tape
machines and spare parts such as head
drums together with the necessary
service skills are a thing of the past.
The debate about broadcast-readiness
should not be a cause for delay.
http://www.inbroadcast.com http://www.noa-archive.com/
Table of Contents for the Digital Edition of In Broadcast - November 2015
