Reliance’s Ambani Lays Out Plan for Low-Cost Mobile Data in India
wsj.com
wsj.com
http://arstechnica.com/information-technology/2015/03/atts-p...
This should be a concern because such data on individual subscribers would be available to law enforcement and become a potential mechanism for highly detailed mass surveillance.
It probably goes on to suggest that this is the amount service providers expect to benefit per customer by inspecting your data. It's evident that those (including marketing agencies) who compensate the service providers in lieu of this data would benefit even more. Creepy world.
I'm a bit apprehensive of their executive saying deep packet inspectionn will be the actual gold mine [http://qz.com/771690/reliance-jio-bombshell-the-good-the-bad...]
Privacy: An unnamed Jio executive mentioned “deep packet
inspection” to Reuters, saying: “It’s called deep packet
inspection, and what you can do with the analytics of that is
mind-boggling,” he said, referring to a practice that digs into
“packets” of data created by computers for efficiency, mining
them for information. If this is happening and Jio is accessing
data packets to develop patterns of user data consumption, this
is a major privacy violation. The company deserves to be taken
to court for this, as much as the India needs a privacy law.
Welcome to India!
Can you even imagine the scale needed to process this kind of data? That's petabytes (rough estimate), every day. Maybe, it's theoretically possible but any investment in this kind of technology would be enormous.
Browsing behaviour data maybe valuable but to what scale? Even a big advertising firm would balk at spending any big bucks for this and remember, the scale needed to mine any information out of this. How much valuable business insights can this generate that wasn't possible in the past?
Maybe I am wrong but I'd be very skeptical if selling is their master plan.
Also, may be some people in India won't mind accepting this if Reliance gives LTE internet at very low cost, others might not just use it i guess.
You are right about the volumes, but wrong about it being impractical.
The volumes with a relatively small opco:
- +-7m subs
- 250gb just for the protocol classification. *
- Then you also have url logs etc
Key factors that reduce the costs and investment:
- commodity hardware (with hadoop etc)
- distributed
- query patterns
- you do not need to store every single record. The data can be aggregated up to hourly, daily, monthly the older it is
This is what we did, data was aggregated which significantly reduced storage.
Tested various options: Hive, hbase, druid
Edit: * per day
It's not like this is a uniquely Indian problem. The largest wireless provider in the US (Verizon) was caught doing basically exactly this, and other (broadband) telecom providers are trying to do the same.
That's exactly what a CONTRACT is: Giving up some rights, in return for things that you didn't deserve before.
No one will dare question it because we have idiots for politicians and we are nowhere close to having a privacy law for our citizens.
Is VPN, Whonix, Tor (in essence compartmentalization) a good combination to escape their monitoring capabilities?
My (Indian) landline ISP even writes to http pages, injecting ads for higher tier internet plans and letting me know that I've reached my data cap (after which I still have internet, but at a reduced speed).
Tor is good, but slow, and many websites block it or show captchas.
I've now gotten used to pages not opening up due to ad-blockers; I refuse to disable those. It may not be all that bad to shun sites that refuse to display content upon detecting ad-blockers.
> Tor is good, but slow, and many websites block it or show captchas.
And then you have Cloudflare.
Not sure how this is relevant; I'm talking about my ISP modifying my packets to include its own ads. The websites have no hand in this aside from the choice to use http.
While I don't like ads, I personally don't use an adblocker (just Firefox's tracking protection, which still triggers antiadblock).
That's needed, but even then any random person who intercepts the cables can read your data. Or with WPA2-PSK, anybody with the password can read your data. So, treat HTTP as in the clear; it's likely someone's watching.
> Tor is good, but slow, and many websites block it or show captchas.
Tor has high latency, but the bandwidth is not that bad. CAPTCHAs only take 20 seconds per site. Other points are true.
The average monthly indian mobile monthly revenue is b/w 150-200rs ($2-$3). The hype is that a user gets a GB of data for Rs 50 (~$1).
But if you look deeper, that's just the rate of charge/GB. The initial plan is still Rs 150 per month (for a paltry 0.3GB of data with free voice). This is better than the current market, but only marginally, not substantially or revolutionary. Heck that fact that the next big plan is Rs 500, and nothing between the basic and the next big plan shows they expect a significant number of users to jump onto the Rs 500 plan. [1]
[1] http://i.imgur.com/IknpMek.jpg
What's going to happen is the 6 other network providers will huddle in their respective HQs and come out with equivalent plans within a fortnight.
But definite kudos to Jio for stirring up the market a little.
"Now, here’s my not-so-big bet: Jio will not make its magic number of 100 million customers in one year. The person who offers me the best odds before end of Saturday on a $200 bottle of a single malt is on.
In fact, more from my smoky mirror: Jio will struggle for at least a few years before it starts making a dent in the telecom market." http://goo.gl/hH7Fai
Link: http://factordaily.com/mukesh-ambani-reliance-jio-launch-imp...