ANd here is a really depressing thought. There are many D* fans here that rave about the quality of 1440x1080i. I guess if you starve someone long enough they'll eat anything.
An Update
- Thread starter ScoBuck
- Start date
- Latest activity Latest activity:
- Replies 549
- Views 113K
- Status
I'm new here, and not to jump on the NBC bashing bandwagon, but I have to say that speaking from experience with NBC 30 OTA HD in my area, it is nothing to write home about. The difference between my fox's 720p and nbc's 1080i is like night and day. Sunday Night Football suffers greatly where I am. Its like watching the Sportscenter Live Update on ESPN.com.
Its not as simple as that. I'm willing to bet that they have a lot of proprietary software and use a lot of middleware that just converts it to be able to used in other environments, such as the website and other departments. I don't think it is as simple as having a Microsoft Dev house where everything talks to one another.
I would be concerned about testing on a vmware machine with the amount of data that would need to be tested... but thats just me.
on a smaller scale... yeah
I will have to agree and say that the HD quality from NBC is some of the worst HD I see (and I get it OTA and via Satellite)
I can almost guarantee you're right about this.
"Simulations" are limited in what they can uncover, and the more complex and intertwined the system the further short of a realistic representation such a simulation will fall.
I can almost guarantee that D*'s billing/authorization system is a three headed monster, consisting of several networked systems from diverse vendors and with diverse archetectures tenticled all the way from the desktops of the CSR's to the live satellite data stream. The idea of being able to simulate all contigencies on such a system with complete reliability makes my head spin, and I've worked on some pretty awesome systems.
I would say that the system has probably been pieced together over time with different modules added to it.
Its probably like Spaghetti.
Its probably like Spaghetti.
And that is the WORST way to have a system run. I bet the biggest problem was just getting people who knew the system and/or the language together to work on it.
great, so we are all placing our hopes and trust in leftover pasta..... where do I find an HD meatball around here?
:I'd be very disappointed if the software integrators didn't test this with real data.
Take a snapshot subset of live data in a staged environment and pass it through your system to see where it breaks.
Would they take all the data? Surely not. Would not having all the data run through in the test environment still cause issues when it gets rolled out? Sure.
But the issues that should crop up should be rather case-by-case, not bring the whole thing to a grinding halt.
One of the most important things to do on data-reliant software projects is testing with "real" data that's also recent data. Not mocked up data. Not real data, but from 5 years ago. Today's data. Guaranteed if you don't use real data, something bad will happen.
And if this is what happened...someone needs to have their employment reviewed.
Take a snapshot subset of live data in a staged environment and pass it through your system to see where it breaks.
Would they take all the data? Surely not. Would not having all the data run through in the test environment still cause issues when it gets rolled out? Sure.
But the issues that should crop up should be rather case-by-case, not bring the whole thing to a grinding halt.
One of the most important things to do on data-reliant software projects is testing with "real" data that's also recent data. Not mocked up data. Not real data, but from 5 years ago. Today's data. Guaranteed if you don't use real data, something bad will happen.
And if this is what happened...someone needs to have their employment reviewed.
Only if it was a upconverted Tanya Memme! Of course i'd rather see her in True HD though, but I'll take what I can get!
exactly, can you imagine how much less post there would be if they added one channel on the 16th, one on the 18th and 19th?? We would all be so freakin excited it wouldn't matter, and we would all be talking about the possibilities of what channel might be up tomorrow instead of bitchin' about CSRs and software...
As several people have posted, they could have easily run a test on a backup of all the data, recent data, backed up moments before the test, and still had a "surprise" when they tried to run it on the real system.
I suspect it's not a matter of the script handling the data itself, but the archetecture of the system is at the root of the problem.
The system is likely a monster of components installed from 1994 to the present, all strung togather in one-of-a-kind proprietary ways.
We know this system is tenticaled into every desktop of every CSR, the live feed of data to the satellites themselves (so it can send authorization commands to your receivers) and to D*'s web site (they have a button that you can use to "refresh" your receiver's authorization all by yourself).
Simulating the monster, that I'm sure this system is, would be simply impossible to do perfectly to every detail and nuance.
Is it possible that someone did something glaringly stupid and their employment should be reviewed as a result? Sure.
Is it possible that the problem was so subtle that only the wildest stroke of luck or genius would have caught it before it bit them in the butt? Yes, that's also possible.
They probably can't add just 1 channel if the problem is indeed related to the billing software. They have to get all that fixed, then all the channels can go live.
- Status
