[Geodata] Tiger Why MTFCC instead of CFCC

[Stephen Woodbridge]

Obe, Regina wrote:
> 
> One thing I'm curious about is why they replaced CFCC with MTFCC.  
> Granted CFCC codes were not perfect, but there are much fewer MTFCC 
> ones.  I still haven't completely figured out how to reconstitute the 
> old CFCC from the MTFCC - e.g what additional properties I need to look 
> at to get more granular categorization.

I'm pretty sure you can NOT do this for all CFCCs. It might be possible 
for some depending on how the data is represented. There was a CFCC the 
indicated road way divided or sharing a rail track. You might be able to 
reconstruct them by looking at

CFCC="A18" === MTFCC="S1100" and ROADFLG="Y" and RAILFLG="Y"

But the rail line might be separated out as a separate edge where the 
geometry is identical to the road segment but has a separate MTFCC. I 
have not had time to go looking got this. for a cases where that would 
prove or disprove this. But this seems to be "Special Cases #1"

One problem is the there are not flags for tunnel, bridge. And if you 
look at the cfcc_to_mtfcc.xls you will see that most all of the railroad 
CFCC are reduced to R1011.

It also appears the a complete list of MTFCC is missing from the 
documentation. Has anyone come across this yet?

-Steve