OT: laptop reliability survey (was: Mac look alike?)

[Avi Greenbury]

Odd wrote:

> Avi Greenbury wrote:
> > Odd wrote:
> Someone who says his personal experience is more valid than
> a  survey of 30.000 computers is most definitely not in the
> real world. That's why I say it's faith-based.

I don't think he is saying that's necessarily so. I'm not. I'm saying
that I have trouble taking that graph as the gospel truth because it is
so contradictory to my experience. I'm questioning both.

> > I don't trust that survey simply because it goes so far against my
> > personal experience.
> 
> And again, you're acting on faith. Same as religious people.

No, I am not acting on faith. I am acting on experience.

It's a graph that disagrees almost completely with my experience. Of
course I'm going to be sceptical of it, and of course I'm going to have
trouble understanding your unflinching belief in it. And, clearly,
you're having trouble understanding why I won't preach it from the
rooftops with you.

> 
> 30.000 computers.
> 

Yes, 30,000 computers. I know, You've said this already.Of these 30,000
computers, all the data I can find is a pretty graph, no raw data, not
even a definition of "malfunction". No data on the average age of these
machines, whether the malfunction was due to user-error or what the
spread of manufacturers across the sample was like. 

There's not even any indication of the profile of SquareTrade
customers. If it only covers those machines that SquareTrade saw, this
would presumably limit them to machines on which people purchased their
extended warranties. Which, one would imagine, contains a fair
proportion of the kinds of people who break things more often. These
people are likely to buy from those brands with a reputation for being
better built and they'll likely still break the thing (it'll just take
longer), so squaretrade will see more of these.

It's a survey of a big number of machines, and while the conclusions
might well be absolutely spot-on, there's no real way of telling how
applicable they are to any of the market beyond whatever the sample
set was (and we can't even be certain of that), because there's no
openness in the process and no real information provided.

Also, and I'm not meaning to infer that this means they've fiddled
the data (they've already obscured it beyond usefulness), it is provided
by a company who sells warranties. There's at least a small incentive
there for them to want people with really reliable hardware to think
it's going to break.

> > I don't think there's anything wrong with Alan's scepticism. If
> > those results are accurate, they will be corroborated by other
> > surveys. If they're not, they will be disputed by them.
> 
> You go find that other survey then.

I don't know that there is one. Which hardly helps either of us.

> > Either way, I _know_ that apple and IBM hardware has tended to fare
> > better in the environment I tend to put my laptops through than Acer
> > or Dell. This might or might not be caused by the relative quality
> > of their construction, but I'm not about to discount several years
> > of experience on the basis of a single online survey, irrespective
> > of the standing of its publisher.
> 
> Faith in your personal experience against a survey of 30.000
> machines ins worthless. It's religion. The cult of Mac.

I have two data sources that disagree. For one I have the raw data and
an incentive to be correct. For the other I have a pretty picture from
people who want advertising.

Of _course_ I'm going to side with my experience. That's what people do.

--
Avi Greenbury
http://aviswebsite.co.uk ;)
http://aviswebsite.co.uk/asking-questions