[mythtv-users] Guide Data Idea Doable or Not?

Rod Smith mythtv at rodsbooks.com
Fri Jun 22 19:30:00 UTC 2007


On Friday 22 June 2007 14:16, Gregg wrote:
> This may not be worth the coding, but is it possible not to house all
> the guide data on the myth box and just query an exiting guide service
> on demand and import the data into the database when required for
> recording?
>
> Example.  I want to record Lost, I hit guide on my myth box and it
> initiates a connection to some guide data site.  I do a search for
> lost and pick the episode I want. It downloads that specific data and
> places it into the database, then at some specific interval it
> searches again for that program every week to update time and so on.
> The gist is the IP will not be banned from the site for scraping all
> program data. It  just query's for the one it wants to record, and If
> the user wants to search current programs the guide pops up a modified
> web browser with preset location info.  The user could search and add
> for as far out as the data provider allows.

I see several problems with this:

1) If your Internet connection goes down, or if there are any problems
   reaching the site hosting the data, you won't be able to schedule
   anything until the problem is restored. With the current system, or
   most others under discussion, this wouldn't be an issue; at worst,
   you'll just be missing a day or two's data about two weeks out.

2) There'll be a performance hit. It's almost certain to take more time
   to search an external database than to search a local one. This will
   be particularly true for dialup users -- they'll have to wait for
   the modem to dial, connect, etc. Such users also won't be able to
   schedule programs when anybody is on the phone.

3) You'll probably be unable to search for programs using title, actors,
   etc., because few existing sites provide such facilities. If you mean
   you want to host a fully searchable database that you access remotely,
   then this problem isn't so big, but then the remote server would have
   to be powerful enough to handle everybody's searches.

4) I'm not convinced your method would result in lower bandwidth use.
   This is a question that would require testing, in the end, but I
   suspect that any but a very light user would end up downloading the
   same data multiple times.

5) Depending on how this would be implemented, there might be privacy
   implications. Do you really want some company to know that you're
   searching for "Paris Hilton's Makeover Tips"? ;-)

-- 
Rod Smith
http://www.rodsbooks.com


More information about the mythtv-users mailing list