To view this page ensure that Adobe Flash Player version
11.1.0 or greater is installed.
expectations of broadcast television. We really do
assume that we are going to get a new picture every 40
milliseconds, and we get cross if we miss even one of
them. We build playout centres to at least five nines reliability
– 99.999% up time – and if broadcast quality means
anything today, it certainly means no freezes or spinning
wheels. We may get grumpy about less than perfect,
seamless content online, but we are more likely to accept
it because we remember when internet video was really,
Television roasted rare
by Dick Hobbs
I was at a conference about workflows the
other day. One of the presentations was
by a very nice man called Eric Minoli, who
is chief engineer at TFO, the French-language
public broadcaster in the Canadian province
He illustrated his presentation with a great story. Some
friends were invited around to his house for Sunday lunch,
and they were chatting with him in the kitchen while he
prepared the joint of beef for roasting. They wondered why,
before putting it into the oven, he carefully sliced the ends
off the joint.
“I have always done it,” Eric explained. Yes, but why, his
friends persisted. “Because my mother did it,” he added.
He rang his mother, and asked why she always sliced the
end off the joint before putting it in the oven. She did not
know why: she did it because her mother had always done
it. So Eric rang his grandmother and asked her why she
always sliced the ends off the joint before roasting it, a
practice which had been passed down through the family.
“Oh, that was because the oven was not big enough,” she
said. The point of the story is that it is all too tempting to do
something the way we have always done it, without ever
thinking about the reasoning behind it. Given that we are
all involved in a huge shift in technology – from tapes to
files – this might be a good moment to at least start asking
questions about the way we have always done things.
One of the direct points Eric Minoli made in his presentation
was that we should stop thinking “TV first”. Broadcast
television, as we used to know it, is now just one output of
many. Just because a broadcast server is more expensive
than an IT server, does that make it more important?
On one level I disagree with Eric on this one. I think
because of the conditioning we have all received – as
consumers not as experts – we do have very high
98 | TV-BAY MAGAZINE: ISSUE 79 80 AUGUST 2013
On another level, though, I think he has a point.
Broadcasters are now expected to deliver to a lot of
different platforms, with each brand of phone or tablet
having a different combination of screen size, codec,
wrapper, metadata and streaming structure. Why would
you not regard the SD and HD broadcast versions as just
yet more packages to be processed and spat out of the
If I may answer my own question, it could be because
broadcasters hate the F word. They do not like the
concept of their treasured content being pushed through a
factory. But that is exactly what a modern facility is, a just-
in-time supply chain ensuring each piece of media is in the
right place, in the right format, at the right moment.
To get to this supply chain approach, though, we have to
change something fundamental we do because we have
always done it that way. We regard getting content from
ingest to delivery as a technology process, because it was
always controlled by the technology.
When VT operators were replaced by robots in Sony LMS
or Panasonic Marc (the two competing robot tape libraries
when playout automation first started), it did not change
the way playout was organised. And when the first Profile
servers appeared in playout areas, again the workflow and
the operation was precisely the same.
I do not think anyone will argue when I say that, to meet
the needs of multi-platform delivery, we have to have
largely automated systems. The big leap in logic – and
in faith – is to turn the thinking upside down and define
workflows in terms of business requirements not in what
the technology can do.
Frankly, the technology can do anything. I know that
vendors hate people saying this, but most of it is only
software. What will make the difference is how we organise
In a blissfully ideal world we would set up a list of rules
along the lines of “have a look at the rights management
database, see what we can do with this programme, then
do it.” If we also asked the system to tell us what resources
were used to achieve each output then we would know
how much it cost, and we could make real commercial
decisions on the viability of providing all these services.
That would be a big change in thinking, but without it we
are not taking full advantage of the technological revolution
we are undergoing. In simple terms, we now have a much
bigger oven: how much longer are we going to slice the
ends off the joint?