[BC] HD Radio's lack of upgrade capability

[Michael Bergman]

Hi, Cowboy,

>>So: proposals to deal with the issues above?

I believe you're not talking about over-the-air upgrades, which streamlines
the discussion a bit...This is a long post, so I'll summarize as: The real
cost of upgrades is surprisingly high, and consumers will end up buying new
products instead of upgrading, even when it is offered.  Liability must be
borne by the hardware maker if the radio must be opened to do the upgrade,
and we don't want such liability unless there's a significant profit.


> EPROM.
> If the base code algorythms reside in a socketed EPROM
> of some sort from the beginning, future upgrades are
> mechanicly simple, even if never done.
>>Cost,
> Pennies, really. ( EPROM are recycleable. )
> Like anything software, the cost is in the writing of it.
> ( which will exist regardless )

First, use flash and reprogram it; it is cheaper.  Or CD-ROM, in products
which have a compatible CD drive (not all CD receivers do have the right
kind of drive.)  However, the hardware cost of the upgrade isn't problem.
Imagine your wife/neighbor dealing with this upgrade.

They'd take it to e.g. Best Buy for the rework, right?  Most people won't do
it themselves.  A car audio shop would charge $75 / hour or some such
amount.  

Also, a hardware company doesn't create upgrade firmware for free.  We have
to pay engineers and pay for testing, etc.  Real resources have to be paid
for as indirect overhead--otherwise the accounting is flawed.

This is a point that comes up frequently: Software is not free.  I think it
is like "maintenance is an investment", easy to understand if you're doing
it, hard to convince those that aren't.  (Not to compare you with GM's,
Cowboy, just trying to make a point!)  

Next: Not 100% of consumers will take the upgrade.  Cost will be spread out
among those who do.  Assume 10% take the upgrade, and run some numbers--very
quickly, you realize that people will buy new radios instead of pay the
upgrade cost.  Upgrades may still be mathematically cheaper, but not by
enough to make it a useful business model.

So you end up with: The real cost of upgrades must be born by the consumer
who opts for it, which will cause the upgrade cost to approach the new model
cost.  So we don't do a lot of upgrades.  (I did mention previously that we
do support upgrades for some models--these are specific cases that don't
create a general rule.  The "real cost" argument is closer to the general
rule.)

Also, w.r.t. software that "will exist regardless", no, we change hardware
platforms every year for cost reduction reasons.  They're not generally
backwards compatible with last year's models.  

We do one hardware/software release per year.  You get the features you paid
for.  That's how the vast majority of products work--from toothbrushes to
Office Suite; you want the new features from the next year, you buy the new
product.  


>>and liability? 
> Customer.
> They can download the BIOS upgrade, or have it done by others.
> If done at Best Buy, or the dealer from whom originally purchased,
> the merchant simply swaps a chip.
> Either way, it's at the customers option to purchase the upgrade, or not.

We don't sell anything without a warranty.  I think the UCC and most state
laws have something on warranty requirements, regardless of our policy,
anyway.  Also, the retailer would not participate (do the install) if the
hardware maker didn't warranty the result.  

> Just a few thoughts....

Appreciate the feedback...a lot of good ideas die on the vine when faced
with the rest of the solution.  Upgrades are like that.  It seems like a
good idea, like Esperanto or a flat tax, until you walk through the full
scenario.

I wish it weren't this way--my life would have been easier if I could have
shipped bug fixes or competitive features after the fact.  But even when the
product is still in the warehouse, it is very expensive to upgrade or fix.

Mike Bergman