Hacker News Reader

Comment by Jach | Hacker News Reader

The header and people's reactions makes it seem Comcast will just do this on a whim as if it's inspecting the page you visit and deciding on the page to display the warning or not. If you read the screenshot, it's just a notice that someone filed a complaint against your IP, and Comcast is alerting you via email, maybe phone, maybe even a letter, and now your web browser. One might argue whether it's better they redirect you to a Comcast Message Page on their own domain one time. One might argue that this is a "feature" on the level of Comcast DNS servers that "helpfully" forward your bad domains to a search engine instead of giving a proper server not found response.

Don't want to receive these messages from Comcast? Don't seed your torrents.

View on HN

This is correct. It is only performed after you are sent emails, letters, phone calls, etc. We do the same for when you are about to exceed your 300GB of data. Most people don't give us a good email, don't login to check it, don't login to their comcast account, etc... This type of notification is to cover those people. We are working on better ways to do this, see: https://www.caida.org/workshops/aims/1503/slides/aims1503_ba...

This system is well documented: https://tools.ietf.org/html/rfc6108

I'll bring up the idea of an opt-out for users that DO check their mail, email, phone, comcast account, etc...

Alternatively, your company could choose not to act as a copyright cop.

Edit: Actually, reading the IETF link you posted, notifying users of a potential malware infection might be an example of how to use this technology in a non-shitty manner.

These companies aren't necessarily choosing to be copyright cops. I can't imagine that sounds very fun or beneficial to them. Systems were negotiated under legal pressure from the RIAA and MPAA.

Of course, Time Warner and Comcast are both also content creators, so they might have some motivation to concede.

https://en.wikipedia.org/wiki/Copyright_Alert_System

Yes, when I wrote custom page tracking software for a custom forum I wrote and hosted, I was parsing out User-Agent information and would actually notice malware browser add-ons, and I'd gently post a notice at the top of my web page alerting them that they might want to run some malware scans. Security vs. privacy right there. That was a long time ago, and I don't track anything any more, but it was an interesting experiment.

Using a shitty technique to do something arguably "good" does not make the shitty technique "good"

Perhaps a captive portal and redirect would be a "nicer" way of doing this. At a bare minimum, it's less intrusive, and it's an accepted practice when using a provider's internet (especially wireless). Injecting JS into a page that I've requested from a third party server would erase my trust in that provider, and I would immediately cancel my service the first time my provider performed that action.

May I ask if your comments in this thread were made in an official/"on the clock" capacity?

Nada, I'm just an engineer. I try and help where possible on Reddit/DSLReports/HN.

Thank you.

(Though I'd caution you that working off the clock is an excellent way to get overworked.)

What's wrong with good old registered mail?

So, you work for Comcast? May I ask what it is you do there?

Internet Services engineering, mainly working on measurement, instrumentation, and customer experience. My group runs our Speed Test, SpeedExperience, Next-gen access trials, our RIPE Atlas probes/anchors, etc..

Thanks for responding. As frustrated as I get with the business side of Comcast, the engineering side has to be pretty interesting.

Whichever team is behind the account system needs work. Why do I have four Comcast accounts? I've only even lived at three different places. Why can I still login and "pay my bill" which still says I owe "$39.99" on my last account? I was paying into it accidentally for months before they killed my new service because I wasn't paying into the new account.

It seems to me it's just hacks on top of hacks written by hacks.

Yes! This happened to me too! I moved to a new house, they configured service and "transferred" my account.

Three months later, my service at the new house is disconnected for non-payment. I look at my bank statement - sure enough, AutoPay happily withdrawing the amount...

... and applying it to the old house's (disconnected) cable, the account for which is now several hundred dollars in "credit" (paying $200/mo for business service).

It was (relatively) easy to get resolved, thanks largely I think to the business support folks. But still... :|

Perhaps users that don't give you a good email don't want to hear from you.

If you've sent emails, letters, and phone calls already - just stop.

It's called consent, fuckers.

They require you to ack, its part of their copyright platform. Would you rather they just turn off your internet until you call in?

Maybe try a certified letter first? Also, they don't require YOU to ack, they require SOMEONE to ack, as I don't see anywhere where they actually have the "acker" verify that they are the account holder. I wonder if there is a legal loophole that the account holder could use as they can't prove that the account holder is the one that acked.

You're missing the point. Injecting scripts into HTML delivered through your ISP for any purpose is grossly abusive on the part of Comcast.

I would call the great firewall of china injecting a script into people visiting baidu, as a means to DDOS github, a gross abuse.

just for context.

I would call that an act of war.

This comment is the only real explanation of what's going on here. I'm not sure why it's buried. It seems people are just reading the headline and breaking out the pitchforks and torches. I understand the issues of privacy here, but it also seems the context in which this is taking place is an important thing to consider, as well.

You mean don't seed infringing torrents, right?

Reply on news.ycombinator.com