2009-10-30

When is 40 hours not 40 hours

When it is 39 hours 36 minutes.

More importantly when it is our favourite telco measuring how long they have had a fault report and working to a 40 hour target or else pay use £1.50 compensation!!

It has happened far too often now to be a coincidence. Faults clearly with them for many days have the clock magically stop at 39 hours something.

Raising the issue has simply got no response from them.

Its got to be a con I my opinion.

2 comments:

  1. The *intention* behind the 40 hour target is a noble one - to fix faults faster.

    However, the presence of the 40 hour target - an arbitrary number pulled out of thin air; why not 35 hours, or 47 hours, or 29? - causes the problem.

    The presence of a target creates a purpose of 'meet the target', so this is what happens. The 40 hours becomes defined as 'hours while with us' not 'hours from fault being reported to fault being fixed'. Everything then gets organised to make sure the numbers that get reported meet the target, while the the thing that actually matters to the customer - how long it takes end-to-end to fix the fault - gets worse as a result.

    A better way would be to not have a target at all, and have the purpose be 'find ways to reduce the time between the fault being reported and being fixed'. This, combined with an understanding of what causes delays in fixing faults and then designing a system that only does work to meet the new purpose, would produce far better results.

    ReplyDelete
  2. Oh, I quite agree!

    Making it the time with them that counts means that they send back for silly reasons without fixing the fault. What I was talking about is cases where they have not done that but have magically stopped their clock anyway. They are the ones with the stopwatch and they are cheating.

    Even when the stopwatch works, you get an arms race. They have systems to do some tests automatically and if they cannot see a fault then send back to us automatically. That way the clock runs for as little as possible. Getting a person to look at it always takes longer.

    Of course, we do the same. We have systems to retest the reported issue when they "send back for retest" and send back to them within minutes (with detailed test report and explanation of why it is still broken). Sometimes this means that over night there are hundreds of exchanges between the systems and no real progress.

    Needless to say I take care (and pride) in making sure our systems do it right and only send back when the issue really is still there. They are a dinosaur and so do not fix things that are daft. e.g. if the test we do to check if still broken is one of their tests that we run on their systems and it says the fault lies with them - but they send back saying "all tests show not fault found" (clearly a lie). That is frustrating.

    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...