No Agree because No Understand text!!
The non-packet control channel (BCCH/CCCH without the P)
is tested and verfied, exists, also handles GPRS packets.
The packet control channel (PBCCH/PCCCH with the P)
is untested and unverified, does not exist, might handle
GPRS packets in the future, and if it does what it is
supposed to, more efficiently than the non-P channel.
Additionally, according to the R97 spec handsets which
camp, exist, on non-P cells will still use the non-P
control channel, the tested, verified (but less efficient)
The problem that:
R97 complient handsets would HAVE TO USE the untested,
unverified p-version (only if) when in a p-channel-capable-cell (which do not exist yet)
In an "old" cell these R97 handsets would happily run
on the old, non-p channel, just like today.
That is,
new = untested but/and unverified -> risk
old = tested and verified but less efficient.
----
Line by line, sentence by sentence (just like with
standards)
"I don't know for a fact that the phones will have to be
recalled for technical reasons once an operator starts
to use PBCCH/PCCCH functions."
Neither do I, nor operators, nor handset manufacturers,
for 100% sure, but everyone agrees it is very possible,
potential disaster. (some claim it is even the most
propable outcome)
One reason that uncorrectly, or even correctly implemented
handsets might "upset" the whole cell, total disaster.
Motorola claims (claimed) that "the standard is clearly
enough written to work".
Nobody else seem to agree.
---
"However, I will take your word that they may be used
without harm to the network once the operators begin to use
the functions."
Here I'm lost, although trying to figure out "they" and
"the functions", I would assume something else than
what "they" seem to refer to:
My, and many other's point, operator consensus, has been that P-control channel handsets cannot be "guaranteed"
to work the (future) P-control channel cells.
- "old" works, "new" is unverifiable.
- present standard demands that "new" is used by handsets.
"However, if the (P-) functions are not turned on for a year
to eighteen months, and a substantial number of
non-compliant (to what??) phones are present, then the
benefits of using PBCCH/PCCCH are not going to be felt by
the operator unless he recalls the phones to make them
compliant"
OK, to cut it short:
two type of GPRS handsets in the future (OK, three, but
to keep it simple)
- The "Nokia type", for the present, only using the
tested, verified "old" non-P channel.
- The "Motorola type", trying to use the "new", future,
unverifiable, more efficient P-channel, but one suitable
bug, and it causes disaster instead of more efficent functions.
The bug can be:
- in the handset implementation
- in the network implementation
or even
- in the standard, doesn't work even if all implementations
are "legally correct", because "everything was not covered"
All of this embedded in the words "untested,unverified"
as well as "on real, live networks".
Ilmarinen
Hey, I'm trying, knowing that this is tough... some of
my young engineers couldn't take the heat and went into
marketing... the question of "risk" and "robust",
maybe most importantly "what happens when things do not
work" (error recovery, reboots to a better world,etc)
I was very happy to find Mr Pihl using the word "trace",
to be able to "trace" what went wrong just before things
went wrong, in contrast to a simple "yes/no" result.
Anyway, this is an intersting case of communicating
without personal contact and blockdiagrams, pictures
and instant messaging (I've had some international
experience, and it often boils down to these things,
more misunderstandngs than mutual understandings,
bless Ericsson and Sony) |
