From: stephen@sprunk.org
On 23-Apr-15 11:42, Adam H. Kerman wrote:
> Stephen Sprunk wrote:
>> On 22-Apr-15 11:18, Adam H. Kerman wrote:
>>> Stephen Sprunk wrote:
>>>> [GSM] was developed by CEPT and later transferred to ETSI.
>>>
>>> That would be the consortium of European post offices, not a
>>> world-wide standards-making process, so your earlier statement
>>> was wrong. ...
>>
>> It's a de facto world standard, with 85% of the market and used in
>> 212 countries. That it wasn't developed by ISO and made a de jure
>> world standard is moot.
>
> It's a standard. Certain places in the world adapted it. That's all
> you can say about standards, that they are standards where
> implemented and ignored where not.
That one country consistently ignores international standards and
creates its own (often inferior) national standards does not negate the
former's existence.
>>>> GSM isn't particularly clever; the point was that everyone
>>>> (except the US) quickly standardized on GSM, so they got
>>>> economy of scale, and for commercialization, that's usually
>>>> more important than cleverness.
>
>>> That's still ridiculous. The United States had the world's
>>> largest market for cellular service at the time,
>
>> No, it didn't.
>
> Yes, it did.
Repeated assertion does not make something true.
> Cellular wasn't affordable without a business purpose.
Yes, it was; I knew plenty of people who had cellular phones, even back
in the 1980s, for personal use.
High prices means you can't _address_ the entire market, but it doesn't
change the size of the market itself, and the US market was smaller than
the European market simply due to the population difference.
Yes, the Europeans went with a different billing model. That reduced
the prices, which increased adoption, which improved economies of scale,
which reduced the prices further, etc. in a virtuous cycle.
>> NMT in Europe had a larger customer base than AMPS in the US, and
>> the problems scaling up NMT to deal with customer density were what
>> led to the development of GSM. . . .
>
> Ok; still not an "economy of scale" issue, because that doesn't
> change the customization required for network design and
> infrastructure. In fact, you've just argued against your own
> position.
No, I haven't. If you think so, you misunderstand something.
>>> so don't give us your "economy of scale" nonsense. In Europe,
>>> they formed a consortium to avoid different standards in
>>> neighboring countries,
>>
>> Europe already had a single standard with international roaming in
>> 1981; the US hasn't achieved that even 30+ years later, even
>> domestically, because we _still_ have competing standards.
>
> I coulda sworn in 1981 we had just the one standard.
Yes, but without international roaming. By the 1990s, we had _four_
standards, all of them still without international roaming because we
didn't use NMT or GSM. Heck, we don't even have good domestic roaming
today, which we _did_ have in the 1980s.
> So you're rejecting the idea that Europe went with an international
> standard that benefited them, without considering the rest of the
> world. It's not like they organized the entire world not on US
> standard into their consortium.
In the early years, the GSM consortium was all European. However, many
other countries joined as time went on, to the point it now has more
members than even the UN.
Europe is varied enough that, if something works for them, it will
almost certainly work for everyone with little, if any, change. And
they're always open to adaptations for other markets because they
understand it's better to sell one product worldwide than to have
different domestic and export products (typical for US companies).
There is a cost to adding support for another country, but that cost
diminishes as the list grows because most of the variations will have
been covered by previous ones, until you reach the point that all new
ones are effectively free; GSM passed that point _long_ ago. This is
yet another form of economy of scale.
>>> In any event, your argument doesn't work because you have to
>>> pick and choose your costs and ignore other costs. There are
>>> major infrastructure costs of the cell phone network that don't
>>> scale up, like erection of towers and equipping them and
>>> connecting them to the telephone network. Sure, individual parts
>>> benefit from mass manufacture, but a whole lot is individually
>>> customized on a per-location basis.
>>
>> Some parts benefit more from economy of scale than others, yes, but
>> that doesn't mean that it's irrelevant as a factor.
>
> Nice backpedal. Irrelevancy isn't the issue. The issue is that you
> exaggerated the benefit hugely.
It's not backpedaling. I said it mattered. You said it didn't. I
showed why it matters.
>>>> ETSI standards often get used elsewhere simply because the rest
>>>> of the world (except the US) doesn't see the point in
>>>> developing competing standards. GSM, for instance, was
>>>> deployed in Australia in 1993, not long after Europe's first
>>>> GSM network went live in 1991.
>>>
>>> Oh, bullshit. It's whichever manufacturer reaches the market
>>> first.
>>
>> No, it isn't. There are many examples of one vendor being first
>> to market and getting stomped by later entrants, particularly if
>> the later entrants join together to create an open standard and
>> thus get better economy of scale.
>
> Snarf. That just means they ganged up and got rid of a competitor
> and got the gubmit to do some of their dirty work for them. It
> doesn't mean "better".
VHS killing Betamax wasn't due to a govt mandate; it was because VHS
cost less due to economy of scale. Same with Android vs iPhone, PC vs
Mac, x86 vs RISC, Ethernet vs Token Ring, etc. The specific players
change, but the ending of the story is almost always the same: whoever
has better economy of scale wins in the end.
S
--
Stephen Sprunk "God does not play dice." --Albert Einstein
CCIE #3723 "God is an inveterate gambler, and He throws the
K5SSS dice at every possible opportunity." --Stephen Hawking
--- SoupGate/W32 v1.03
* Origin: LiveWire BBS -=*=- UseNet FTN Gateway (1:2320/1)
|
