Quote:
Originally Posted by AdultKing
How long do you think it takes for a brand new release by a studio to appear on some tubes and torrent sites?
This is something I was wondering, so I decided to delve into the data.
On the 1st February I launched some code to query my web crawlers and other sources.
I followed 40 top studio and paysite network Twitter accounts and some PR company accounts.
Then I set about running quick crawls across a sample of tubes and torrent sites.
Each time a new release was announced a time stamp was created of the first official channel sighting of the release, then a quick crawl began over 8 tube sites and 17 torrent sites.
This data sample ended today and the results are shocking (at least to me).
The shortest time measured between an announcement of a release and it's 1080p resolution availability for download or viewing was 7 minutes and 44 seconds.
The longest time measured between an announcement of a release and it's 1080p resolution availability for download or viewing was 2 hours 19 minutes 09 seconds*
The average time measured between an announcement of a release and it's 1080p resolution availability for download or viewing was 18 minutes 31 seconds
When a studio, producer or pay site network announce a release there is almost no time between the release and its availability though torrents or some tubes.
I placed a star * next to the longest time measured because this excludes titles that were not released in the sample period. Over the sample period just a few announced titles never made it to tubes or torrent sites.
Am I out of touch to even be surprised by the speed of these releases?
|
Last time I came to post something you invited me to interact more with some GFY members in their threads so here I come I will interact with your content, first of all as you made clear in the last thread English is not my first language, but I am a little bit more fluent in the language of math, just like you recommended me to improve my spelling in English I am going to make the same recommendation for you in the language of math, your language in this category is really poor ….. but let me explain what I mean by that:
- I do not understand what is the point of this thread whatsoever, if you are trying to make an informative post your sample is extremely low and irrelevant, you are making the statistical mistake that is called a non-representative sample more commonly known as an anecdotal evidence…… if you are such an expert in big data and AI as you claimed to be in your previous thread this should be common knowledge to you, but in case it isn’t here is the source: https://en.wikipedia.org/wiki/Anecdotal_evidence
- You said you followed 40 studios, twitter accounts and some PR company accounts… how many? You understand there are 180 quadrillion web pages , meaning 10exp5 (1 with 15 zeros at the right) so even if you analysed a million of pages that can be taken as a rounding error right? Meaning that even if you analysed a million an accounts that would be still absolutely irrelevant.
- I quote you: “This data sample ended today and the results are shocking (at least to me).” So you really call this a sample? Weird that somebody who says to know about Big data does not know the data he has recollected is absolutely irrelevant and can’t even be called a sample.
- I quote you again: “The shortest time measured between an announcement of a release and it's 1080p resolution availability for download or viewing was 7 minutes and 44 seconds.
The longest time measured between an announcement of a release and it's 1080p resolution availability for download or viewing was 2 hours 19 minutes 09 seconds*
The average time measured between an announcement of a release and it's 1080p resolution availability for download or viewing was 18 minutes 31 seconds”
You understand that presenting these data without a base means nothing right? What if it took 2 seconds to be pirated 10 years before? Then there is an improvement, is this fast? Compared with what? Is it better? Compared with what? Why do you show the average time instead of the mode? Does the information you collected have a normal distribution? Because if it does not you know the measures you used make no sense correct?
- I quote you once more: “then a quick crawl began over 8 tube sites and 17 torrent sites.” Once we already defined your “sample” is completely ridiculous…. You do understand that sometimes affiliates will upload to the tubes files, videos with the same name of the releases so the people who are looking for the pirated content find this files, clicks on them and get their cookie so when they go around without finding anything and go back to the page that sells the content they get accredited with the sale correct?...and these content may not be even pirated, may be the launching video edited several times, put together to be a 2hr "movie" so people think they are getting the real thing... maybe instead of insulting as spammy the affiliate techniques and tools you should try to better understand them to not make such a nonsensical posts……
I really hope you can explain further the comments I made, this post is not informative, interesting nor correct lets stop the superficial analysis that seems to be so abundant.