That is why some providers may dispatch an engineer and test to eliminate the effect of internal wiring.

Vague yes, but unless the market is willing to support things like engineer only installs and visits to verify speeds we are somewhat stuck with the situation.

Of course it could be said that installing FTTH or DOCSIS would resolve the issues, but people still get slow speeds on those services just it is not down to connection speed (usually) but rather congestion and other issues.

All that said the move to quad play is a move away from a vibrant broadband market in terms of migrations and easy moving, since the quad play packages are so different between the providers.

In a pure commercial world the answer may be to refuse to provide service to those with the most problematic lines.

The article did not mentioned throughput speeds. I assume both synch speed and throughput speed are considered here?

Edited by Oliver341 (Thu 11-Jun-15 10:02:14)

I am guessing it is synch speed as to do throughput speed you would need an independent way to test it that everyone agreed gave a good indication of real world. And if the ISP has control of the speed test then they could knock out much of the contention by placing the tester strategically - if they don't have control then they will blame the speed test itself or the peering links between it and wherever the test is hosted.

A little muddy, need to see the revised code of practice for the wording used the full text is not yet available, just going on what was an embargoed release talking about changes.

Yep. One of those headlines which looks good in practice, but when it comes down to the detail, it proves harder to be workable than it seems. I would think people have more issues with throughput speeds in general than slow speeds due to line faults, which would typically manifest in additional issues like voice line noise and dsl disconnections.

And that's before REIN is even mentioned...

Edited by Oliver341 (Thu 11-Jun-15 10:14:58)

Where I step in with my daily consultants fee

An ISP can game things e.g. by hosting files and thus being chosen by testers that pick source with lowest latency. A user can game things by speed testing and producing data that is slower than the reality.

By comparing a users speed tests with others on the same ISP and maybe in the same area you can get a very good idea of the issues, and with several tests you can even see patterns and maybe point towards a local issue e.g. Wi-Fi problems.

At the end of the day, an ISP is likely to already let someone go if their speeds are bad and ISP can do nothing more, if the ISP is reasonable.

If Ofcom wanted to disrupt then it would insist on a shorter maximum length of contract for FTTC migrations to reflect the 1 month Openreach contract once you've had FTTC for more than 12 months.

Cant Ofcom just ban stupidly long contracts instead, then people can move with just a months notice anyway. Many ISPs have 1 month contracts these days. Although then i guess you run the risk of price rises if an ISP dont lock you into a 12-18 month contract

In reply to a post by bobble_bob:

Although then i guess you run the risk of price rises if an ISP dont lock you into a 12-18 month contract

It's not a risk, price rises would 100% happen. Activation fees, charges for routers, etc.

If prices rise during a long contract you can already exit, as 30 days notice has to be given.

Many ISPs with 1 month contracts - if you include the SME and the less promoted options from those like PlusNet then maybe, but when did you see a 1 month contract advertised for fixed line broadband on TV.

Ofcom has banned contracts longer than 24 months already.

Short contracts carry a risk, i.e. you do a sign-up offer so people do, but then leave for another sign-up offer. So a likely reaction from the market is a lot less sign-up offers and people may then feel less inclined to switch. There is a balancing act involved.

The move by Ofcom is good, but its not a revolution. The migration changes are late, having taken 3+ years to come into effect. Have not published our version of the new process in FAQ as been waiting for any last minute changes and to try and learn the gotcha's

Ideally you could get a sign up offer by agreeing to a contract or choose not to have the sign up offer and have a shorter contract. That is not uncommon in most other industries.

If people are willing to pay the connection fees and not have special deals then a 1 month contract should be no issue. If they want the offers then they commit.