|
RBUA
The Residential Broadband Users’ Association
August 13th, 2002
Meeting Attendees from Rogers:
- John H. Tory, Q.C., President and CEO, Rogers Cable
- Tony Faccia, Vice President, Network and Capacity Planning, Rogers Cable
- Brad Munroe, Vice President and General Manager, Rogers High Speed Internet
- Taanta Gupta, Senior Vice President, Media Relations, Rogers Cable
- Andrew Ho, Director of Network Operations, Service Management Centre, Rogers Cable
Meeting Attendees from the Residential Broadband Users' Association:
- Christopher Weisdorf, President and Technical Director, RBUA
- Jacob (Jack) Gryn, Senior Member, RBUA
- Jason deCourcy, Senior Member, RBUA
- Fergal Warde, Senior Member, RBUA
- Peter Hope-Tindall (via conference call), Senior Member, RBUA
- Tom Pettypiece (via conference call), Senior Member, RBUA
Rogers Meeting Agenda:
Presented by John Tory, President and CEO, Rogers Cable
- Update on rollout of rate throttling from 3.0 to 1.5 Megabits per second on the downstream, and 400 Kilobits per second to 192 Kilobits per second on the upstream
|
Presented by Tony Faccia, Vice President, Network and Capacity Planning, Rogers Cable
- Plans for monthly byte caps
|
Presented by Brad Munroe, Vice President and General Manager, Rogers High Speed Internet
RBUA Meeting Agenda:
1. Surreptitious (Bandwidth / Throughput / Performance) Rate Reductions
- what are / were the exact reasons (technical, economic, etc.) for diluting our service?
- why do it without any notice - either before or after the fact - to your customers? Does this constitute a reasonable level of customer service in the view of Rogers?
- why have the vast majority of your customer support reps denied any knowledge of this action?
- how many subscribers, thus far, have complained about this action?
- how many subscribers, thus far, have cancelled their service due to this action?
- what does Rogers plan to do about those who receive subpar service under the new rate throttling regime (i.e. subscribers who can never reach 1.5 Megabits per second on the downstream and / or 192 Kilobits per second on the upstream?)
- very high, regular spikes in packet latency (i.e. >500 milliseconds) occur when reaching, or exceeding 1.2 Megabits per second on the downstream; what, if anything, is Rogers doing to remedy this deficiency?
- are / were there any issues with respect to network capacity which prompted this action?
- shouldn’t the implementation of DOCSIS - which provides 200-400% more capacity than existing architectures - make this action irrelevant and unwarranted?
- does Rogers feel that charging 12.5% extra for half the service is fair to its customers? Do you think that you are giving your customers what they want by imposing these drastic reductions in rates?
- who are the beneficiaries of this action? Customers? Management? Employees? Shareholders? Creditors?
2. Monthly Bit / Byte Capping
- again - why?
- will this be silently implemented just like the rate reductions were?
- how much of a cap on the downstream; how about on the upstream?
- what are to be the billing rates for data downloaded / uploaded in excess of the caps?
- Rogers has cited in the past that a small minority of the subscriber base (i.e. ~10%) uses a vast majority of the internet service’s resources (i.e. ~90%). How were these numbers determined? Which criteria were used to evaluate service utilization? How often do these numbers fluctuate? Why aren’t comprehensive data on this topic being disclosed to your subscribers?
- will extra public IP addresses count towards greater monthly limits?
- will the Point-to-Point Protocol over Ethernet (PPPoE) ever be used in order to track bandwidth usage, or for any other purpose?
- will all transferred data be counted towards the upstream and downstream monthly limits? Will ICMP (i.e. ping, traceroute, etc.) packets count? Will all IP packets count? Will UDP packets count; what about multicast packets? Will all TCP segments count; how about TCP segments with either of the ACK, SYN, or RST flags set? Will unsolicited packets (such as HTTP queries delivered by the Red Alert and Nimda viruses) count; how about spam e-mails and web browsing pop-ups?
- will time of day be a factor in determining the extent in which data transfers count against the monthly limits?
- will a gigabyte consist of 1000 megabytes, or 1024 megabytes?
- will “unused bandwidth” be rolled over or credited?
- will transfers internal to the Rogers network be counted the same as those external to the network?
- who will handle dispute resolution? Is Rogers to be the judge, jury and executioner? Where are the checks and balances?
3. Mass Network and Server Performance Issues
- additional reports have been received from subscribers in almost all areas experiencing overall reduced and / or unreliable network performance over the past few months
- routing has been very flaky at times, with packets taking “scenic” routes through California when their ultimate destinations are on the east coast
- regular routing on Teleglobe to domestic destinations often see packets sent from Toronto to New York and back again, adding to more hops; will this kind of routing continue with Cable and Wireless?
- routing through Sprintlink is generally poor, no matter what
- DNS resolution is often far slower than it should be
- the NNTP (i.e. news) server is usually very slow and unreliable during the day; binary retention is good, but text articles can be dropped; where are the new groups? Why does the process of adding new groups take so long?
- there is an “incorrect POP3 cleanup” problem with one of Microsoft’s e-mail clients, Outlook XP; when will this be addressed?
4. Customer Support Issues
- pass-the-buck scenarios are far too commonplace these days
- when a network outage occurs, there is no update on the VRU telephone system to indicate this and an expected ETA. Why? It would spare subscribers the agony of waiting on hold if this were continually and reliably updated
- there has not been an outage or maintenance post in the newsgroups for months. Why?
- method of e-mail support is very confusing with the use of long, cryptic reference numbers
- subscribers’ machines are still being blamed for network-related problems without sufficient prior investigation
- the knowledge of support staff has not at all improved over time; support staff do not receive sufficient training to properly conduct their duties
- serious service-related problems that affect several hundred subscribers, or even several thousand, are very rarely disclosed
- ever since our first meeting in March of 1999 we have been asking for a real-time, or near real-time (i.e. updated within 5 minutes) network status page; we continue to ask for this
- the subscriber base is still kept almost completely in the dark, period.
5. Shaw -> Rogers Cable Asset Swap and Integration
- subscribers with Motorola cablemodems continue to experience lackluster and problematic service
- the few Motorola->DOCSIS cablemodem swaps that have been brought to our attention have completely rectified all of these problems - in some cases problems that have existed for years
- when will the tens of thousands of these subscribers on Motorola cablemodems be completely migrated / swapped over to DOCSIS?
- the areas / primary hubs in question that contain all the Motorola subscribers are as follows: Agincourt, McNicoll, Markham, Newkirk, Thornhill, Unionville and Woodbridge - all in Metro Toronto
- subscribers have never once been briefed on these problems, let alone given resolution dates. Why not - especially after our repeated requests for Rogers to do this?
6. Personal Server / EUA Issues
- why are subscribers running secured services on ports 22 (SSH) and 23 (Telnet) being targeted by the abuse department?
- why are subscribers who have DNS aliases being targeted by abuse?
- subscribers illegitimately disconnected due to “abuse” are never presented with evidence justifying such punitive action. Why not?
- server definition is inconsistent, vague and highly interpretational
7. Compensation / Credit for Faulty or Nonexistent Service
- credit looks like it’s being applied to most requesting subscribers’ accounts, although this has worsened since the beginning of the year
- credit policy is still very inconsistent and unclear
- early 2000 (lack of) mail service credit still never clarified to subscribers
This agenda was made possible by the contributions found at https://www.rbua.org/board/viewthread.php?FID=9&TID;=2857 and the adjoining pages.
Estimated duration of meeting: 2 hours, 20 minutes
|
Prepared by Christopher Weisdorf
President and Technical Director,
Residential Broadband Users’ Association
|
|
