Sapient said:
How hard can it be to program NBR anyway? It seems like something I could even have done in BASIC on an Apple IIC a million years ago when I was young. The names have got to be in some database or array or something, right? Read the names, compare to a list, record the matches.
LOL.. Yeah right that simple.. :lol: Boy if I had a nickle for every time an engineer oversimplified a task they were not responsible for I would be a rich man.
I am sure NBR is not as simple as one would think to implement for the following reasons.
1) You are shoe horning this functionality in and existing code base. This is not a ground up design and has to work within the current archetecture.
2) it is by no means as simple as compare to list and record matches. You have do deal with repeats, conflict priorities, live viewing events etc. You are also dealing with a remote as your input device that needs to map well. You also have to deal with GUIDE updates and changes.
3) This is an embedded environement and you usually don't have all the tools that you would have in a enterprised or desktop enviroment.
It may be simple from 1,000 foot level, but my guess is that there is more there than meets the eye. Personall I try not to make trivialize the amount of work someone has to do unless I know the environement hte person is working under, have a good idea of the architecture and code based, and have done the task or a similar task under those conditions before.
For reference, I am working on a Web based UI for an embedded product that certain people within my company thought would be done by Sept 30th. They were shocked by the estimates I gave and more than once I have been told by a low level embeeded software engineer or manager that it should not take that long. These are people with little or no experience actually doing this type of work.
My orginal estimate was the end of November. Due to other issues and external forces, we will not be making this date. I could get into the details why, but I would rather not. I will say that the major reasons for the slip has nothing to do with the engineers.
There are a number of engineers on here with various backgrounds and experiences. Some have worked with companies with good software process and others have worked with ones that don't. I myself have worked in both environments. I am sure we all have opinions about how long implementing NBR shoudl take given reasonable staffing. However, it is my opinion that without actually understanding the details our estimates are majore SWAGS at minimum and are most likely huge underestimates of the effort.