p2pnet news view P2P | Freedom:- A Canadian ISP has accused Bell Canada, Canada’s biggest ISP, of  deliberately holding back relevant data in its efforts to justify traffic throttling.

In November last year, it was revealed Bell was deliberately blocking the accounts of its users, blaming the action on  a tiny number of them who, it charged, were were hogging bandwidth by sharing files.

The claim was, and is, spurious and it provoked 55 smaller IPS, all of them members of CAIP (Canadian Association of Internet Providers), to file a demand with the CRTC, asking it to order BELL to stop shaping traffic.

In the US Comcast, Bell’s opposite number, has just been told by the the FCC (Federal Communications Commission) to halt its ‘traffic management’ activities.

But in Canada, in the face of a huge and still growing storm of protest from angry customers, users and ISPs alike, Bell continues to fly in the faces of the people who keep it in business, even upping the stakes.

Unbelievably, it’s now launched a new anti-competitive move that could  see its wholesale customers not only throttled, but tied down to only offering 2-gigs to 60-gigs per month.

Bell conveniently ignored vital facts

TekkSavvy, an Ontario ISP run by Rocky Gaudrault (right), has been at the sharp end of the fight since day one.

Bell recently filed its final answer to the charges with the CRTC Canadian Radio-Television and Telecommunications Commission). It highlighted an unnamed provider, trying to use incomplete data as evidence traffic shaping isn’t causing serious harm to its ISP customers.

The data came from TekSavvy, based in Chatham, and with customers in Ontario, Quebec, Alberta and British Columbia. And as a consequence, it’s now its joined the long line of companies and people who’ve filed submissions

In it, Gaudrault points out Bell conveniently ignored vital facts which, had they been included, would have helped to paint an entirely different picture, at the same time showing up the Bell claim as a travesty.

Addressed to CRTC secretary-general Robert Morin, it states »»»

At pages 8, 9 and 59 of its Final Answer, Bell filed certain information in confidence with the Commission regarding one of its GAS customers, namely TekSavvy. In response to a letter from CAIP dated 16 July 2008, this information was subsequently disclosed by Bell to TekSavvy on 17 July 2008.

In TekSavvy’s view, Bell has omitted a number of significant facts from the information that it has presented in relation to TekSavvy. As a result, TekSavvy does not exemplify the conclusions that Bell seeks to draw and the Commission would be induced into error in relying on them without the benefit of the additional facts and corrections included in the attached submission.

Here’s the rest of TekSavvy’s submission »»»

4. For the purposes of the above-noted application by the Canadian Association of Internet Providers (CAIP), TekSavvy notes that it subscribes to both Bell Canada General Tariff Item 5410 (also known as Gateway Access Service or the “GAS” tariff) and General Tariff Item 5420 (also known as High Speed Access service or the “HSA” tariff). TekSavvy uses these wholesale ADSL access services to provide Internet access services on a retail basis to end-user customers as well as on a wholesale basis to other telecommunications service providers.

5. Unlike Bell, TekSavvy does not engage in any selective traffic shaping or throttling of end-user traffic. TekSavvy believes that customers have a right to privacy in their communications and that carriers and other telecommunications service providers have a duty to uphold this right in the absence of a clear and explicit lawful authority requiring or mandating otherwise. TekSavvy further believes that it is not consistent with subsection 27(2) and section 36 of the Telecommunications Act for carriers such as Bell to intentionally degrade, block or otherwise interfere with the communications that they carry for the public, including communications consisting of GAS customer traffic.
Background and Issues

6. TekSavvy is a member of CAIP and fully supports CAIP’s Application to the Commission requesting certain declarations and a mandatory order that Bell cease and desist from “throttling” its wholesale ADSL access services.

7. Unbeknownst to TekSavvy, in Bell’s 11 July 2008 Final Answer to CAIP’s Application, Bell filed information in confidence that purported to establish certain conclusions regarding the network planning practices of its GAS customers and the overall health of the independent ISP industry using TekSavvy as an apparent example of a typical GAS customer.

8. Further to a request for disclosure filed by CAIP on 16 July 2008, Bell gave notice and disclosure to TekSavvy of the data pertaining to TekSavvy that Bell had filed in confidence at pages 8, 9 and 59 of its Final Answer.

9. Having now had an opportunity to review the information relating to Teksavvy at pages 8, 9 and 59 of Bell’s Final Answer, TekSavvy is concerned that Bell has omitted significant and relevant facts from the information that it has presented about TekSavvy. In particular, TekSavvy notes that:

(a) Contrary to Bell’s assertion at paragraph 26, the facts demonstrate that TekSavvy not only planned but actually placed orders with Bell’s Carrier Services Group (CSG) for additional capacity in a timely fashion;

(b) Because of Bell’s throttling practices, it has become impossible for TekSavvy to plan for additional capacity at its AHSSPI and in the transport and backbone portions of its network and yet Bell hypocritically appears to be accusing TekSavvy and other GAS customers of irresponsibility in planning for additional capacity;

(c) In TekSavvy’s particular case, the need for additional capacity is directly related to subscriber growth rather than increases in per-user bandwidth consumption;

(d) TekSavvy’s subscriber growth patterns is likely atypical as compared to that of other GAS customers.

TekSavvy Network Planning

10. As previously stated by CAIP,1 independent ISPs and other competitors typically plan for additional capacity when peak AHSSPI utilisation rates hover at between 50 and 60 percent of total capacity of the AHSSPI.2 At paragraph 26 of its July 17 Final Answer to CAIP, Bell accuses TekSavvy of waiting until it has “greatly surpassed” this level before requesting more capacity.

11. These accusations contradict both the facts as well as TekSavvy’s entire service order history with Bell for the past year. In actual fact, TekSavvy started planning for additional capacity when the utilisation rates observed at the AHSSPI were in the range of 50-60 percent and placed its order with Bell when utilisation rates were in the range of approximately 65 per cent. As discussed more fully below, Bell is solely responsible for the delay in fulfilling these capacity requests in a timely fashion.

12. Furthermore, Bell has significantly misrepresented the data in making this accusation. In particular, Bell presents utilisation levels on dates that are months after TekSavvy had in fact placed an order for additional capacity from Bell. Bell failed to disclose that it has consistently taken months upon months for it to install additional AHSSPI capacity after an order has been submitted by TekSavvy.

13. The following table provides a list of requests by TekSavvy for additional capacity, the utilisation levels observed at the TekSavvy AHSPPI at the date in question and the Bell order fulfilment date many, many months later. TekSavvy notes that it cannot guarantee that the reported utilisation levels at the relevant times are “peak” levels. However, they are representative of responsible planning practices on the part of TekSavvy as planning for additional capacity not only occurred in advance of the dates that orders were actually placed and fulfilled, but also well in advance of any observed full utilisation levels.
Table 1 – TekSavvy Requests for Additional Gig-E AHSSPI Capacity and Bell Order Fulfilment Date

14. Thus, while in Table 1 (page 9) of its Final Answer, Bell attempts to demonstrate that on 4 March 2008, TekSavvy was operating at approximately 90 percent capacity on three AHSSPIs, Bell inexplicably fails to disclose that TekSavvy had in fact ordered a fourth AHSSPI almost three months earlier on 10 December 2007, and a fifth AHSSPI on 18 December 2007. Conveniently, Bell chose to provide data from 4 March 2008, just a couple of weeks before the fourth AHSSPI had actually been installed on 20 March 2008. Had the fourth and fifth AHSSPIs been installed by 4 March 2008, as they should have been by that time, then TekSavvy would have been operating at approximately 53 percent capacity.
3 On 28 May 2008, a Bell representative asked if TekSavvy still needed a fifth AHSSPI, to which

TekSavvy responded positively.

15. Similarly, Table 1 shows that on 3 June 2008, TekSavvy was operating at approximately 71 percent capacity. However, had the fifth AHSSPI been installed, as requested on 18 December, 2008, then TekSavvy would have been operating at approximately 56 percent capacity.

16. Consequently, it is erroneous to claim that TekSavvy waited for traffic to reach near-capacity levels before requesting more AHSSPIs. Rather, TekSavvy has consistently planned for additional capacity and requested additional capacity far in advance. The only cause of near-capacity levels being reached at the AHSSPI is unexplained delay in order provisioning on the part of Bell.

17. TekSavvy notes that it is not the only GAS customer that has experienced lengthy delays in the provisioning of AHSSPI capacity by Bell. In fact, in the response to CAIP(CRTC)15May2008-4, CAIP noted that another GAS customer was forced to introduce traffic shaping measures itself because of the lengthy delays it had experienced in ordering AHSSPI capacity from Bell.

18. While Bell dismisses CAIP’s concerns regarding its members’ inability to appropriately plan for network and capacity as a result of throttling by asserting that “the traffic kept growing even after traffic shaping”, clearly it does not understand that it is as a direct result of Bell’s traffic shaping practices that its GAS customers, such as TekSavvy, are no longer able to properly plan for additional capacity at their AHSSPIs and in the transport and backbone (peering) portions of their own networks. TekSavvy notes in this regard that:

• It has entered into contracts for additional transport, and backbone arrangements well in advance of the time when Bell began throttling its GAS services, under which significant traffic/capacity, is now not being delivered or exchanged as a direct result of Bell’s throttling;
  
• Bell’s service provisioning intervals for additional AHSSPI capacity, which in 2007 stood at a couple of months at most, have now jumped, at least in TekSavvy’s case but as corroborated by the experience of CAIP’s other members,4 to over four, five or six months, which also impinges in a significant way upon TekSavvy’s ability to appropriately plan;
  
• With the advent of throttling and Bell’s refusal to confirm whether it is expanding or plans to expand its throttling activities, it is impossible for TekSavvy to carry out standard network planning activities at the AHSPPI and in other portions of its network based on historical traffic patterns. In fact, TekSavvy finds itself in the ridiculous situation where it now must ask Bell whether it should or should not be planning for additional capacity and, if so, how much; and
  
• As a result of Bell’s traffic shaping practices, TekSavvy is still not able to fully utilize the AHSSPIs that it has ordered and that have been placed into service by Bell. The reason why this is the case is because Bell’s throttling of GAS traffic reduces overall traffic volumes at TekSavvy’s AHSSPIs, thereby preventing these AHSSPIs from processing the full traffic loads that they were actually designed to process. TekSavvy has paid for these AHSSPIs, but because they are throttled, TekSavvy is not able to take make full use of these GAS service elements.

TekSavvy’s Growth

19. In its Final Answer, Bell attempts to demonstrate that competition from independent ISPs is thriving and relies on the data provided in Figure 17 of its Final Answer in order to make this point.

20. By way of comment, TekSavvy notes that Bell’s sanguine assessment of competition in the retail market for high speed Internet access services and, in particular, the residential segment of that market, is not consistent with the facts. As noted by CAIP in its Final Reply, the retail market for high speed Internet access services is dominated by the ILECs and the cable companies, whereas the independent ISP sector represents a tiny fraction of the overall market. In 2006, for example, the total market share all independent ISPs in the retail market for residential high speed Internet services was a mere 3.5%. In the view of TekSavvy, this is hardly indicative of a fiercely competitive market with a thriving independent ISP sector. In fact, it shows just how dominant Bell Canada and other incumbents are in this market.

21. Insofar as Bell’s references to TekSavvy’s growth are concerned, TekSavvy believes that it is not an ideal example of growth in this market sector because its growth is not typical of other independent ISPs. This is due in part to the fact that its customer base is made up of both retail end-user customers as well as wholesale customers for whom TekSavvy provides a turnkey Internet access service.

22. It is also important to note that, despite Bell’s claims to the contrary, TekSavvy’s need for greater capacity has not been as a result of an increase in average levels per user. Rather, it is because TekSavvy has managed to increase its overall base of end-users. TekSavvy notes in this regard that the number of TekSavvy end-users set out in Figure 17 (page 59) of Bell’s Final Answer is incorrect because it does not take into account certain wholesale arrangements that TekSavvy has entered into with other independent ISPs. This has the effect of over-stating TekSavvy’s average levels per user. Table 2 below sets out the actual number of TekSavvy’s end-users.
Table 2 – Number of TekSavvy End Users

23. Taking the true figure of end-users that are on-net to TekSavvy, two things are clear. First, TekSavvy’s growth patterns albeit small, may be quite different when compared to other competitors. Second, TekSavvy’s requirements for additional AHSSPI capacity are due almost entirely to subscriber growth as opposed to increased levels of usage among existing end-users.

24. In many respects, however, Bell’s submissions on TekSavvy’s business are irrelevant and detract from the real issues that are under consideration in this proceeding, including the fact that the GAS traffic of competitors represents a tiny proportion of overall traffic on Bell’s network. Given this fact as well as the absence of any evidence that competitors are the cause of any alleged congestion on Bell’s network, there is no justification whatsoever for Bell’s throttling of GAS customer traffic. In fact, Bell’s throttling measures can only be explained as a short-sighted and reflexive reaction to the very small inroads that competitors have made into a market in which Bell apparently believes that only it and perhaps the cable companies are entitled to compete.

Conclusion

25. Based on the foregoing, TekSavvy submits that it does not exemplify the conclusions that Bell seeks to draw using TekSavvy as an example. Indeed, as noted above, TekSavvy is relatively unique when compared to other independent ISPs. As a consequence, the Commission would be induced into error if it were to rely on those conclusions without taking into consideration the additional facts and corrections contained in this submission.

Not at all incidentally, in the US the EFF (Electronic Frontier Foundation) has introduced Switzerland, an application designed to allow users to test the integrity of their Net communications.

“When Comcast started tearing down BitTorrent connections, it didn’t tell anyone it was doing so,” said Free Press tech consultant Robb Topolski, who’s been in the forefront of the US fight since the beginning, in a p2pnet comment post.

“When Bell Canada started throttling wholesale customers, it also was an unannounced surprise,” he said, adding:

“Software like Switzerland will help Internet users worldwide determine if their Internet provider is delivering the real deal or if it might be secretly delivering something else.”

Definitely stay tuned.

. .Stumble It!

New York Times – xxxxxxxxxxxxxxxxxxxxxx, August xx, 2008

---

Use free p2pnet newsfeeds for your site. It’s really easy!

Subscribe to p2pnet.net | | rss feed: http://p2pnet.net/p2p.rss | | Mobile – http://p2pnet.net/index-wml.php

---

Net access blocked by government restrictions? Use Psiphon from the Citizen Lab at the University of Toronto. Go here for details. Download here.

This entry was posted on Saturday, August 2nd, 2008 at 8:15 am and is filed under Freedom, P2P. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.

HOME

Viewed 3987 times