Auto date / time update

[Ext User(Jack Mitchell)]

Seems that auto date / time update has been introduced onto Telstra GSM
recently.

Anyone else noticed this?

[Ext User(Dogfart)]

On Mon, 16 Jan 2006, at 01:33:56 [GMT GMT] (12:33:56 Monday, 16 January 2006
where I live) "Jack Mitchell" wrote:

> Seems that auto date / time update has been introduced onto Telstra GSM
> recently.

Would only work if a setting you have allows for this?

[Ext User(Jack Mitchell)]

Yes, most new and recent phones have the feature - even my old Nokia 8250
does.

However the feature wasn't previously available on the network.

"Dogfart" <flatulantdingo@deadspam.com> wrote in message
news:4466628555$20060116192135@dontbotherspamming. com...
> On Mon, 16 Jan 2006, at 01:33:56 [GMT GMT] (12:33:56 Monday, 16 January
> 2006
> where I live) "Jack Mitchell" wrote:
>
>> Seems that auto date / time update has been introduced onto Telstra GSM
>> recently.
>
> Would only work if a setting you have allows for this?
>

[Ext User(will kemp)]

On Mon, 16 Jan 2006 19:21:32 +1100, Dogfart wrote:

> On Mon, 16 Jan 2006, at 01:33:56 [GMT GMT] (12:33:56 Monday, 16 January 2006
> where I live) "Jack Mitchell" wrote:
>
>> Seems that auto date / time update has been introduced onto Telstra GSM
>> recently.
>
> Would only work if a setting you have allows for this?

Yeah, you'll have to turn it on in the phone's config.

[Ext User(Dogfart)]

On Mon, 16 Jan 2006, at 19:49:42 [GMT +1100] (19:49:42 Monday, 16 January
2006 where I live) "will kemp" wrote:

>> Would only work if a setting you have allows for this?

> Yeah, you'll have to turn it on in the phone's config.

And if you have a "smart" phone (combined with diary) be aware that the
times shown in the diary are local times, but the time on the phone is
wherever you are, so conflicts may arise.

[Ext User(Simon Templar)]

Dogfart wrote:
> On Mon, 16 Jan 2006, at 01:33:56 [GMT GMT] (12:33:56 Monday, 16 January 2006
> where I live) "Jack Mitchell" wrote:
>
>>Seems that auto date / time update has been introduced onto Telstra GSM
>>recently.
>
> Would only work if a setting you have allows for this?

From what I have read this feature also only works when you change
cells or turn your phone off and back on.


--
73 de Simon, VK3XEM.

[Ext User(Jeremy Quirke)]

"Jack Mitchell" <dgddg@agaga.com> wrote in message
news:8KCyf.218044$V7.129048@news-server.bigpond.net.au...
> Seems that auto date / time update has been introduced onto Telstra GSM
> recently.
>
> Anyone else noticed this?

Just arrived in Australia and you are indeed correct - I found most of the
networks in my travels have supported this and it was extremely useful,
particularly since my Nokia's backup battery seems to have failed and
therefore it forgets the time when the battery is removed. But also useful
during the numerous time-zone transitions.

[Ext User(Jeremy Quirke)]

"Simon Templar" <usenet@vk3xem.net> wrote in message
news:43cb953b$1@news.melbourne.pipenetworks.com...
> Dogfart wrote:
>> On Mon, 16 Jan 2006, at 01:33:56 [GMT GMT] (12:33:56 Monday, 16 January
>> 2006
>> where I live) "Jack Mitchell" wrote:
>>
>>>Seems that auto date / time update has been introduced onto Telstra GSM
>>>recently.
>>
>> Would only work if a setting you have allows for this?
>
> From what I have read this feature also only works when you change cells
> or turn your phone off and back on.
>
>
> --
> 73 de Simon, VK3XEM.

Well the date/time/timezone is encoded in the MM INFORMATION message, which
may be sent by the network at any time during an RR connection, according to
GSM 04.08. It appears with M-NET it is sent at least at IMSI attach (and
probably at all location update - IMSI attach is just a form of location
update really. And this makes sense, as it allows mobile stations to cross
state borders). Interestingly, I wonder how this works at say, Tweed Heads.

[Ext User(John Henderson)]

Jeremy Quirke wrote:

> Well the date/time/timezone is encoded in the MM INFORMATION
> message, which may be sent by the network at any time during
> an RR connection, according to GSM 04.08. It appears with
> M-NET it is sent at least at IMSI attach (and probably at all
> location update - IMSI attach is just a form of location
> update really. And this makes sense, as it allows mobile
> stations to cross state borders). Interestingly, I wonder how
> this works at say, Tweed Heads.

Maybe sending MM INFORMATION is new for Telstra, but I believe
they have been sending GMM INFORMATION (the GPRS version) for
many years.

But for both these "Time Zone and Time" information elements,
GSM 04.08 says:

"The purpose of the timezone part of this information element is
to encode the offset between universal time and local
time in steps of 15 minutes.
The purpose of the time part of this information element is to
encode the universal time at which this information
element may have been sent by the network."

and

"Due to ambiguities in earlier versions of the protocol
specifications, some mobile stations may interpret
the received NITZ time as local time. This may result in
incorrect time settings in the mobile."

furthermore

"The mobile station shall not assume that the time
information is accurate."

The above implies that the phone should correct the _timezone_
(offset from GMT) only. An otherwise inaccurate clock remains
inaccurate.

Is your take on GSM 04.08 different?

John

[Ext User(Simon Templar)]

Jeremy Quirke wrote:
> Well the date/time/timezone is encoded in the MM INFORMATION message, which
> may be sent by the network at any time during an RR connection, according to
> GSM 04.08. It appears with M-NET it is sent at least at IMSI attach (and
> probably at all location update - IMSI attach is just a form of location
> update really. And this makes sense, as it allows mobile stations to cross
> state borders). Interestingly, I wonder how this works at say, Tweed Heads.

I have been told that the GSM protocol was designed with borders
specifically in mind, as in Europe some countries don't want signals
going over the border into neighbouring countries. With the time
division they can apparently selected the distance to what the cell can
go out to very accurately.

Not that I would expect Telstra or any other carrier to do that here in
Australia.



--
73 de Simon, VK3XEM.

[Ext User(will kemp)]

On Tue, 17 Jan 2006 00:18:49 +1100, Jeremy Quirke wrote:

> Well the date/time/timezone is encoded in the MM INFORMATION message, which
> may be sent by the network at any time during an RR connection, according to
> GSM 04.08. It appears with M-NET it is sent at least at IMSI attach (and
> probably at all location update - IMSI attach is just a form of location
> update really. And this makes sense, as it allows mobile stations to cross
> state borders). Interestingly, I wonder how this works at say, Tweed Heads.

With CDMA, when i go across that border - which i do every now and then -
i have to switch my phone off and then back on again to get the right time.

Will

[Ext User(Jeremy Quirke)]

"John Henderson" <jhenRemoveThis@talk21.com> wrote in message
news:431p0kF1lqqu0U1@individual.net...
> Jeremy Quirke wrote:
>
>> Well the date/time/timezone is encoded in the MM INFORMATION
>> message, which may be sent by the network at any time during
>> an RR connection, according to GSM 04.08. It appears with
>> M-NET it is sent at least at IMSI attach (and probably at all
>> location update - IMSI attach is just a form of location
>> update really. And this makes sense, as it allows mobile
>> stations to cross state borders). Interestingly, I wonder how
>> this works at say, Tweed Heads.
>
> Maybe sending MM INFORMATION is new for Telstra, but I believe
> they have been sending GMM INFORMATION (the GPRS version) for
> many years.
>
> But for both these "Time Zone and Time" information elements,
> GSM 04.08 says:
>
> "The purpose of the timezone part of this information element is
> to encode the offset between universal time and local
> time in steps of 15 minutes.
> The purpose of the time part of this information element is to
> encode the universal time at which this information
> element may have been sent by the network."
>
> and
>
> "Due to ambiguities in earlier versions of the protocol
> specifications, some mobile stations may interpret
> the received NITZ time as local time. This may result in
> incorrect time settings in the mobile."
>
> furthermore
>
> "The mobile station shall not assume that the time
> information is accurate."
>
> The above implies that the phone should correct the _timezone_
> (offset from GMT) only. An otherwise inaccurate clock remains
> inaccurate.
>
> Is your take on GSM 04.08 different?
>
> John

Yes, well, I don't know.. I'm not sure how you are interpreting it.

I believe this part: "NOTE: Due to ambiguities in earlier versions of the
protocol specifications, some mobile stations may interpret
the received NITZ time as local time. This may result in incorrect time
settings in the mobile."

is referring to some older MSes interpreting the time-specific parts of the
IE as the local time, and then the timezone IE as the difference of this
local time between universal time, when in actual fact the time-specific
parts are "UTC" (i.e. close to), and the time zone IE must be added to
correct for local time.

For this part: "This IE may be sent by the network. The mobile station
should assume that this time zone applies to the Location Area of the cell
to which the Channel Request message was sent. The mobile station shall not
assume that the time information is accurate."

is referring to the fact that the time may not be precise to UTC (i.e. right
down to the second, or even 10 seconds). I have found this to be true on
some networks I was using a couple of weeks ago (I was trying to find an
accurate source for the new year timezone, but the two networks were about
30 seconds apart, indicating at least one was wrong).

[Ext User(John Henderson)]

Jeremy Quirke wrote:

> I believe this part: "NOTE: Due to ambiguities in earlier
> versions of the protocol specifications, some mobile stations
> may interpret the received NITZ time as local time. This may
> result in incorrect time settings in the mobile."
>
> is referring to some older MSes interpreting the time-specific
> parts of the IE as the local time, and then the timezone IE as
> the difference of this local time between universal time, when
> in actual fact the time-specific parts are "UTC" (i.e. close
> to), and the time zone IE must be added to correct for local
> time.
>
> For this part: "This IE may be sent by the network. The mobile
> station should assume that this time zone applies to the
> Location Area of the cell to which the Channel Request message
> was sent. The mobile station shall not assume that the time
> information is accurate."
>
> is referring to the fact that the time may not be precise to
> UTC (i.e. right down to the second, or even 10 seconds). I
> have found this to be true on some networks I was using a
> couple of weeks ago (I was trying to find an accurate source
> for the new year timezone, but the two networks were about 30
> seconds apart, indicating at least one was wrong).

It's pretty clear that your experience means that your
interpretion is correct. The bloody thing is still ambiguous -
I took "The mobile station shall not assume that the time
information is accurate." to be an instruction to phone
manufacturers, not to end users like you and I.

One remaining question is whether Telstra is sending the
"Network time zone" information element only, or "Universal
time and time zone". My phone is now updating the timezone,
but not changing the fine detail of the time (if I set it
incorrectly by 12 minutes say). This becomes /prima/ /facie/
evidence that time itself is not being broadcast. But it might
also mean the Siemens interpreted GSM 04.08 in the same way
that I did.

John

[Ext User(Jeremy Quirke)]

"John Henderson" <jhenRemoveThis@talk21.com> wrote in message
news:432o5hF1kp25oU1@individual.net...
> Jeremy Quirke wrote:
>
>> I believe this part: "NOTE: Due to ambiguities in earlier
>> versions of the protocol specifications, some mobile stations
>> may interpret the received NITZ time as local time. This may
>> result in incorrect time settings in the mobile."
>>
>> is referring to some older MSes interpreting the time-specific
>> parts of the IE as the local time, and then the timezone IE as
>> the difference of this local time between universal time, when
>> in actual fact the time-specific parts are "UTC" (i.e. close
>> to), and the time zone IE must be added to correct for local
>> time.
>>
>> For this part: "This IE may be sent by the network. The mobile
>> station should assume that this time zone applies to the
>> Location Area of the cell to which the Channel Request message
>> was sent. The mobile station shall not assume that the time
>> information is accurate."
>>
>> is referring to the fact that the time may not be precise to
>> UTC (i.e. right down to the second, or even 10 seconds). I
>> have found this to be true on some networks I was using a
>> couple of weeks ago (I was trying to find an accurate source
>> for the new year timezone, but the two networks were about 30
>> seconds apart, indicating at least one was wrong).
>
> It's pretty clear that your experience means that your
> interpretion is correct. The bloody thing is still ambiguous -
> I took "The mobile station shall not assume that the time
> information is accurate." to be an instruction to phone
> manufacturers, not to end users like you and I.
>
> One remaining question is whether Telstra is sending the
> "Network time zone" information element only, or "Universal
> time and time zone". My phone is now updating the timezone,
> but not changing the fine detail of the time (if I set it
> incorrectly by 12 minutes say). This becomes /prima/ /facie/
> evidence that time itself is not being broadcast. But it might
> also mean the Siemens interpreted GSM 04.08 in the same way
> that I did.
>
> John

Telstra is sending the full univesal time IE which includes the time zone
IE. Furthermore, the time is correct pretty much to the second.

Perhaps your equipment only regards the timezone?

[Ext User(Paul Day)]

Jeremy Quirke <jqr@nospamausmobile.com> may have written:
> Just arrived in Australia and you are indeed correct - I found most of
> the networks in my travels have supported this and it was extremely
> useful, particularly since my Nokia's backup battery seems to have
> failed and therefore it forgets the time when the battery is removed.
> But also useful during the numerous time-zone transitions.

I've found it pretty flakey when travelling, with telcos screwing up the
time-zone advertisement so often I've disabled the feature entirely and
just change the time-zone manually as I cross the border. Need to do the
manual TZ change on a myriad of other devices anyway - what's one more?

PD

--
Paul Day
Web: http://www.bur.st/~paul/

[Ext User(Dogfart)]

On Tue, 17 Jan 2006, at 04:38:58 [GMT +1100] (04:38:58 Tuesday, 17 January
2006 where I live) "Simon Templar" wrote:

> I have been told that the GSM protocol was designed with borders
> specifically in mind, as in Europe some countries don't want signals
> going over the border into neighbouring countries. With the time
> division they can apparently selected the distance to what the cell can
> go out to very accurately.

As in Hong Kong / China.

half way across the bridge at Lo Wu, (on the border between Honkers & China)
the carrier switches to whatever country you are entering.

[Ext User(John Henderson)]

Jeremy Quirke wrote:

> Telstra is sending the full univesal time IE which includes
> the time zone IE. Furthermore, the time is correct pretty much
> to the second.
>
> Perhaps your equipment only regards the timezone?

Looks like it, Siemens S55 and C45 are both behaving that way.

Thanks,

John

[Ext User(Michael)]

"Simon Templar" <usenet@vk3xem.net> wrote in message
news:43cb953b$1@news.melbourne.pipenetworks.com...
> Dogfart wrote:
> > On Mon, 16 Jan 2006, at 01:33:56 [GMT GMT] (12:33:56 Monday, 16 January
2006
> > where I live) "Jack Mitchell" wrote:
> >
> >>Seems that auto date / time update has been introduced onto Telstra GSM
> >>recently.
> >
> > Would only work if a setting you have allows for this?
>
> From what I have read this feature also only works when you change
> cells or turn your phone off and back on.

And how many people dont change cells?
Whats your point?

[Ext User(Michael)]

"Jack Mitchell" <dgddg@agaga.com> wrote in message
news:8KCyf.218044$V7.129048@news-server.bigpond.net.au...
> Seems that auto date / time update has been introduced onto Telstra GSM
> recently.
>
> Anyone else noticed this?
>
>

About 50 times over the last week
<rolls eyes>

[Ext User(Michael)]

"Jeremy Quirke" <jqr@nospamausmobile.com> wrote in message
news:43cb9d3d$1$23561$5a62ac22@per-qv1-newsreader-01.iinet.net.au...
>
> "Simon Templar" <usenet@vk3xem.net> wrote in message
> news:43cb953b$1@news.melbourne.pipenetworks.com...
> > Dogfart wrote:
> >> On Mon, 16 Jan 2006, at 01:33:56 [GMT GMT] (12:33:56 Monday, 16 January
> >> 2006
> >> where I live) "Jack Mitchell" wrote:
> >>
> >>>Seems that auto date / time update has been introduced onto Telstra GSM
> >>>recently.
> >>
> >> Would only work if a setting you have allows for this?
> >
> > From what I have read this feature also only works when you change cells
> > or turn your phone off and back on.
> >
> >
> > --
> > 73 de Simon, VK3XEM.
>
> Well the date/time/timezone is encoded in the MM INFORMATION message,
which
> may be sent by the network at any time during an RR connection, according
to
> GSM 04.08. It appears with M-NET it is sent at least at IMSI attach (and
> probably at all location update - IMSI attach is just a form of location
> update really. And this makes sense, as it allows mobile stations to cross
> state borders). Interestingly, I wonder how this works at say, Tweed
Heads.

The same as it does with phone calls