[mythtv-users] Supporting schedulesdirect.org

[Michael T. Dean]

On 08/07/2007 05:03 PM, Jeff Wormsley wrote:
> George Galt wrote:
>   
>> While we would all rather have the data -- or just about anything else
>> -- for free, I don't see it as unreasonable that we pay something for
>> the information and for the convenience of having that information
>> integrated into MythTV.  
> Re-read my post, I didn't say that the data wasn't worth paying for, I 
> simply said it was the second worse of the possibilities.  As in the 
> following:
>
> Best.......: Free, easy to get data (what we used to have with Zap2It)
> Next Best..: Free, hard to get data (screen scraping)
> Next Worst.: With cost, easy to get data (SchedulesDirect)
> Worst......: No data
>
> Of course, some people would swap the two in the middle.  These people 
> probably bought Tivo at one point, or paid for MCE, being the type who 
> first think to buy rather than build (ok, probably a gross 
> generalization there, but there really is a difference in mentality 
> between those who think first about buying a solution and those who 
> think first about avoiding spending anything).

OK, so you're saying there are two types of people in the world:
cheapskates (whose time is worth less than their money) and lazy people
who pay for everything.  I really thought I was a person who chose the
best solution regardless of cost (and that's how I ended up using MythTV
and never bought a TiVo, MCE, or any other for-pay system).  I guess,
though, I'm a cheapskate (who happened to spend thousands of dollars on
his MythTV system over the last 4 years).

Now that I know that I'm a cheapskate (since I didn't pay for
TiVo/MCE/whatever), I suppose I should go get a $6/mo TiVo (which would
take me many hundreds (plural) of months to equal the cost of my Myth
systems).

>   Personally, I spend 
> enough on the hardware, I don't want to be bothered with having to pay 
> for listings, if I don't have to.  I may very well have to, and if the 
> inevitable screen scraping solutions prove to be too much of a pain, I 
> may well do so.  But I'll have to be convinced they are a pain first, 
> because I will always take the free path first (why else do I run Linux 
> and MythTV, rather than Windows Media Center?).
>   

Well, the reason I run GNU/Linux systems and MythTV is because they work
better than any other solution I've found (including Windows/Windows
MCE).  But--completely discounting hardware cost--they're significantly
more expensive than Windows/Windows MCE (because of the time I've put
into making them work better).  And, when only non-time costs (i.e.
hardware and subscription costs) are compared to a TiVo (all of which
generally have very low hardware costs), my Myth boxes are egregiously
expensive.

>> While one can certainly argue that the
>> information is out there and that one has already paid for it -- I get
>> it through my cable provider and in my daily newspaper -- but it isn't
>> in a form that can easily integrate into MythTV.  I'm happy to pay for
>> that convenience, though I'd rather not have to use PayPal, but that's
>> just personal preference.
>>     
> I have a problem with PayPal as well, although I have an account and 
> have used it on occasion.
>
> I guess my point is that TMS is making a mistake in charging for the 
> data (even if they are charging SchedulesDirect, who in turn charge us). 
>   Too many people will cry foul, and screen scrapers will return, and 
> they will be right back to where they were before Zap2It, with their 
> servers being overwhelmed by scrapers, and having their coders re-write 
> the presentation layer weekly to try to stop the scrapers from working.

Obviously someone hasn't seen the TV listings data sites that put /all/
the text of the titles/subtitles/descriptions into images
(hundreds/thousands of images per page) that can't be screen-scraped (In
fact, the page I saw on a real web site (for overseas listings data) had
2 separate "fields" per word or few words that were overlaid to make the
words readable--kind of like fields in an interlaced video.)  There may
also be Java- or Flash-based listings apps that hide the data
(potentially using technological measures to protect the copyrighted
data, such that DMCA would apply)...  Basically, if HTML (text) becomes
too much of a problem, what's to stop them from switching.  (And, the
image-based solution I've seen works in all GUI browsers (and the
population of text-based browsers is definitely small)--regardless of
plugin availability--though, admittedly, is an ugly hack.  Text to image
generators are a dime a dozen these days, and--with appropriate
attention to design--likely to be quite usable with today's
computational resources.)

So, I can't wait until I see the "screen-scrape to get image names/OCR
to get data" solutions people are forced to come up with as this "arms
race" escalates.  Talk about fragile systems...  And you're saying
that's a better solution than paying a reasonable fee for the data?

In an arms race for protecting data, you only have to be willing to go
far enough that the value of the data is less than the cost of accessing
the data.  Sure, text is still common (and kind of necessary if the site
wants to be accessible), but who's to say that in the past couple of
years during which they weren't fighting this losing battle Zap2It
hasn't put together a solution to win it once and for all (using a
non-text based approach).  IMHO, even an HTML-based screen scraper that
I have to update every couple of months has less value to me than a
subscription-based service--even if the service charges more than a TiVo
subscription costs.

>   If running Zap2It was such a PITA, then let SchedulesDirect take over 
> that function, but they should have given them a free license to the 
> data, for the reasons given above.  It would be worth it to them.

Oh, I see.  You personally are entitled to this data--that every TiVo 
and cable/satellite STB user has been paying to receive.  I didn't
realize you had this entitlement.

Mike