2012-03-08

Who should pay?

Discussion point.

I am not saying who this is yet and not even saying "our favourite telco" on this occasion.

We buy a service - a broadband connection - delivered at a master socket in a premises. It is a service we sell to our customer (the end user).

If something make that service, at that service delivery point (master socket), stop working, then it should be fixed, clearly. We pay for a working service.

What if the cause of the brokenness is some radio interference, on the telco side of that master socket, not actually on the end users side of it.

A common example is a faulty plugtop power supply such as a mobile phone charger causing radio interference. We have also see faulty satellite equipment, and even poorly installed telephone lines in lifts!

What if it is someone down the road interfering with all the ADSL lines on the road?

What if it is someone in the same building complex but not actually anything to do with the end user?

Now, an engineer will get called out, and investigate, and after some time may find the cause and persuade the owner of said plugtop to replace it or at least turn it off! Problem solved.

The issue is that a lot of people get involved in making that happen. The end user spends time investigating the issue. We spend time investigating it. We get our supplier to investigate. Ultimately one or more engineers actually go there, test things, hunt down the problem. The owner of the plugtop spends time on it. That is a lot of work done by a lot of people.

So who pays for that?

Who should pay for that?

Or is it one of those things - each party doing the work has to take the hit for that work to get things fixed even if not their fault?

I only ask because we have a case of a 3rd party interference, and our supplier is insisting we pay for the engineers involved. That does not normally happen, but in this case, even though clearly "3rd party", so (obviously) not us causing it, but also not even our end user causing it, they still want us to pay. They seem to have no good reason why we pay, and not the party employing the engineers, and not our supplier, but us. Clearly we can't charge our customer for fixing the service we provide to them, we would be stuck with the bill.

Maybe I need to have our customer bill us for their time, and we add out time and bill our supplier, and our supplier adds their time and bills the telco, and the telco adds their time, and bills the person with the faulty plugtop. That would be just as logical as the way it seems to be done now. The last step means showing a tort though, and maybe that last person needs to bill the manufacturer of the faulty equipment, who then charge their insurers...

14 comments:

  1. I guess it gets more interesting when you get to the situation where the interference is from a licensed spectrum user, and is therefore not faulty equipment, but where the ADSL kit is not sufficiently immune...

    From the perspective of the company who runs the copper pair, they *could* invest lots of money in installing better balanced / more shielded / better filtered / whatever systems which better reject such interference. However, it is - in the common case - not a sensible investment. But that choice can come back to bite in the rare case where the normal installation is not sufficiently immune to such interference. Perhaps you could argue they should pay, on the grounds of it being their compromised installation?

    (An analogy would be that it's perfectly sensible to install flat-roofed houses in dry countries, but don't expect to do the same in Scotland!)

    ReplyDelete
  2. I'm not sure that everyone playing 'the blame game' will do any favours in the long term.

    Sometimes we all spend time resolving things that aren't our fault. It strikes me that this one such event.

    ReplyDelete
  3. Common sense would dictate that it would be the top level supplier which would cover the cost, as it is affecting the service which they are supplying, through no fault of the end user.

    However it is ultimately what is in the contract (ADR aside). Alternatively they could decide to terminate the contract, as it isn’t financially viable for them to spend time fixing…

    ReplyDelete
  4. Well, the problem wasn't caused in equipment at the customer's end and the problem wasn't caused in equipment at your end. The problem was due to interference in the equipment in the telco: ergo, IMHO, it should be the telco's responsibility. They could have made precautions to avoid electromagentic interference (today, with the solar storm, may be interesting) with additional shielding but neglected to do so.

    You pay the telco to get a signal from the customer's site to your equipment: if that signal doesn't come through as expected, then it's the telcos fault and their responsibility.

    If the telco want to take action against the plugtop owner/manufacturer to reclaim costs, then it is up to them - but I can't see it being yours or the customers fault.

    ReplyDelete
  5. I realise this is never going to happen in reality given the speed Ofcom move at, but I would say if the emissions from the plugtop are in breach of the regulations, it is Ofcom's problem to track them down and stop them, so BT should simply hand over to Ofcom evidence of the breach and Ofcom pursue it at their cost.

    If the emissions are not in breach (either due to being too low power or because a license is held / they are in license free bands) then clearly the telco should pay as the service is not fit for purpose if it is being affected by perfectly legal RF emissions...

    ReplyDelete
  6. There's a side reason to say "if Ofcom's radio licensing enforcement doesn't fix it, make the DSL supplier pay the cost of RFI", beyond Alex's "not fit for purpose" point: the obvious upgrade from DSL (whether ADSL or FTTC) is FTTP, which is inherently proof against long-distance RFI issues.

    If the supplier chose to upgrade DSL to optical links, then they would have absolute certainty about the location of any service affecting RFI - test in the exchange, if not there, then it must be in the customer's premises. Replace CPE with tested CPE, and then it's the customer's problem.

    ReplyDelete
  7. If you know that the problem is 3rd party interference, what further work do engineers need to do? Do you need someone with better equipment to be able to identify which house the interference is coming from? Do you need an enforcer?

    Some thoughts:

    (1) Is there a local radio amateur society which you or the customer can contact for advice? While amateurs aren't licensed to transmit for commercial purposes, they're certainly allowed to wander around looking for EMC problems using whatever equipment they have. They might even appreciate that you've found a problem which may also affect their enjoyment of radio. The regional EMC advisor might be able to offer you some high-level advice or point you in the right direction - again, as a radio amateur my argument would be that anyone polluting the spectrum would be of interest to radio enthusiasts.

    Ideally you'd get advice and instructions on equipment which the customer could be loaned to walk around his neighbourhood and find the source of interference, applying this for all possible REIN matters (an AM radio is an excellent start!);

    (2) Ofcom is technically the correct place to go to for reporting spectrum licensing violations. Unfortunately, borne of the same government philosophy, they're as enthuasiastic about doing this job of resolving spectrum usage disputes as the post-Woolf civil courts are of resolving regulated business/consumer contract disputes (hence your ADR fun).

    So any process may take time, and they'll not go in guns blazing unless a particularly big business pushes them to - Ofcom's a textbook example of regulatory capture. It's not like the old Radiocommunications Agency where an officer might turn up and confiscate equipment or issue proceedings for failing to follow the rules;

    (3) Are you sure that the interference is due to emissions contrary to some licence or excessive emissions in an unlicenced part of the spectrum? This may affect whether Ofcom tries to charge you;

    (4) Is the customer or his neighbours also experiencing problems with broadcast radio or TV reception? If so, the BBC are also responsible for investigating.

    ReplyDelete
  8. OK, this is a case where an engineer confirmed it was 3rd party interference, and managed to get the source fixed.

    We are now in negotiations with said carrier over this. We will not charge our customers if they have not done anything wrong to break their own service. We already expect customers to put in their own time helping diagnose and resolve faults they have with their service, but no more than that.

    So, on principle, we will not pay for that cost either, other than the time and effort we put in to diagnosis and handling of the fault ourselves.

    If, as it seems, our carriers are going to start charging us for something that was not our fault or our end users, then we will have to consider dropping that carrier, at least for all new connections. We may also have to consider increasing prices overall for that carrier to cover our extra costs in providing the service.

    Watch this space - nothing set in concrete yet, and we hope we can resolve the matter amicably as we sort of have with our fav telco.

    ReplyDelete
  9. A faulty plugtop eh?

    Never mind the gingerbeers, who will pay for the proctologist?

    spectrum = freedom

    ReplyDelete
  10. To play devil's advocate, why should the supplier of the service be out of pocket for sorting an issue caused by a third party? The research to locate the fault would have already eaten into any profitability and they are a business?

    Shouldn't they go back to the supplier of the faulty equipment and charge them for a fault that has bled into spectrum it is not meant to?

    This really gets awkward when the issue is an another premises and not affecting that user (say an old TV)

    ReplyDelete
  11. I would have thought it was the owner of the cable for inadequate sheilding. As another user said you pay C to get a message from A to B... If the message doesnt arrive or is mangled, C is responsble.

    ReplyDelete
  12. I hope this wasn't me as we've just had exactly this issue (external REIN) on one of our lines and after many months and many, many actual and failed SFI visits finally got it fixed.

    I'm firmly of the opinion that the provider of the L1/2 link who is charging upstream for the service should fix the issue promptly and bear the cost of doing so. If they are then able to recover some of that cost from some third party who's negligence (if any) lead to the interference then great, but that is between the two of them.

    In our case we've had intermittent REIN on my home line since Dec 2010, finally fixed at the end of Feb 2012! To be fair, for most of 2011, the line was stable-ish, but in Dec/Jan 2010/2011 and Dec/Jan/Feb 2011/2012 it was non-operational. Over that time, we have had 8 SFI visits, two of which were no-shows where you sit around in a building with no broadband for a morning waiting for the engineer to turn up and then get a call at the end of the slot to say that the engineer won't be coming because they don't have any available. This makes my blood boi, as, if they sent an engineer and I wasn't there, apparently AA would be sent a bill for £98 for wasting their time. There is apparently no compensation for wasting my time.

    On all but the last SFI visit, the engineer didn't check for REIN but merely fiddled around with things randomly in the hope of improving matters, we've probably been through about 4 D-side shifts as they tried to find lines that had a better SNR. There has been a B* van in our tiny village about once a week over the last few months as most of my neighbours have also had broadband problems. Despite wasting many many days of their time (and mine, and my neighbours, and their CPs) B*'s systems and people cannot put 2 and 2 together to actually find and fix a localised REIN problem, they just keep repeating the cycle of SFI visit followed by an attempt to close the case.
    Finally one Saturday morning, the engineer assigned to yet another SFI visit was by random chance also the REIN engineer in an adjacent area and had a 444B on his van. He quickly looked at the line history, fitted an RFI filter and went for a wander around the area. He found the approximate location of the interference which he was pretty certain was the cause of the problem but couldn't do any more as he was on an SFI visit so could only refer to the REIN helpdesk and ask for it to be sent back to him in a REIN capacity.

    Sadly it wasn't, but was sent instead to another engineer who looked at the history and just closed the task as "fixed" without actually doing anything! (this seems to be an endemic problem in B*, closing tickets without actually resolving the problem) Thankfully, first engineer had been "keeping an eye on the ticket" to see how it turned out as he felt some ownership of the problem and got it re-opened, found source of issue, got in touch with householder and got equipment removed (it was a terrestrial aerial amp on an adjacent building). Of course this whole thing meant that even after the REIN issue was identified, several weeks went by before it was actually resolved.

    Aside from the monumental inefficiency in the above tale which means that it cost the telco (and me, and AA) about an order of magnitude more than it should have cost anyone to fix this, why do I think that the telco should pay?

    Well they are selling a L1/2 service that they say works and are charging money for it. That service has pretty fundamental technical issues with things like noise immunity, which mean that really it is only just fit for purpose and needs very careful fault finding in circumstances where external noise compromises the service. Notwithstanding that, the telco offer the service to everyone and make shed loads of money from it. They should be prepared to do everything possible to make the service fit for purpose and remedy any faults caused by the technical weaknesses of the product that they have chosen to sell.

    ReplyDelete
    Replies
    1. Not you in this case. But yes, it is all very inefficient in the first place. If the telco are allowed to charge for this then they have an incentive to be inefficient as that means selling more engineer time to, well, us!

      Delete
  13. A&A and I had a problem on a line in Hutton near Bristol.
    For years I had been pushing A&A to get the line to deliver what theory said it should be capable of. It was obvious when you looked at the notches in the ADSL bins, that (mainly in the evenings) interference was knocking down the line speed. You could even calculate from the frequency of the bins that were notched out, what the likely MF radio transmitter was that was causing the RFI.

    Various SFI (supposedly) experts turned up on site and did all manner of tests, but to no avail. The level of understanding of these Openreach guys was very poor. They did not know what bins were. They had no access to the logs that A&A and BT have input to. Their argument seemed to be that the phone was working so what more did I want? !

    And then at last, a REAL engineer turned up.
    One who opened up manholes in the street and tested performance at points on the cable route. He discovered that because some fitter could not tell the difference between Blue and Brown on a grubby street cable, the last 70 metres of the route to the house was untwisted pair as our house had two A legs and the house next door had two B legs !
    Success! Even the BT engineer was pleased as he had actually solved something.

    The line speed doubled at a stroke and the interference went away for good.

    I looked in the control page log a few hours after he had departed.
    The log said .......
    No Fault Found !!
    Openreach - Don't ya just love 'em?

    ReplyDelete

Comments are moderated purely to filter out obvious spam, but it means they may not show immediately.

Missing unix/linux/posix file open option

What I would like is a file open option for "create replacement file". The idea is that this makes a new inode in the same mount p...