My Photo

About Me

  • Hello you. I'm a 38-year old MSc student, studying Advanced Computer Science at Sussex University. I'm especially interested in Internet and mobile software, sensors and pervasive computing, user interfaces, and the process of developing great software.

    Before that I spent 11 years running Future Platforms, a software company I co-founded which makes lovely things for mobile phones, and which I sold in 2011.

    I read a lot, write here, and practice Aikido and airsoft. I live in Brighton, a seaside town on the south coast of the UK, with two cats and a clown.

AdSense

Stalk Me

  • Email me:
    twhume at gmail dot com
Blog powered by Typepad

« TV causes autism - study | Main | My So-called Second Life: Michael Smith, MindCandy »

October 24, 2006

Comments

Josh Russell

hello,

sorry this one i just don't get. isn't 3G designed to be an always on, persistant connection?

i use agile messenger. it's always on, it performs well. i'm not sure of the exact mechanics of it, but i can't imagine how lots of people using this will potentially bring down networks.

and also, for this reason, why will this not threaten SMS? if poeple use it, it will threaten SMS. it doesn't matter if it's a performance problem on the towers. the only reason it won't take off is not a technical one, it's a matter of user adoption and removal of barriers to it's use. unless the operators turn it off.

maybe i'm missing something?

Tom Hume

Josh: I'm no network guy, but I think this is down to signalling rather than data transfer. When I need to transfer data from my handset (to do WAP stuff or send an MMS), it sets up a 3G data connection and transfers over that.

The issue with IM is not that it won't work - as your experience, and that of many other folks, shows, it does work for limited numbers of users. But as the number of signalling messages grows (to alert individuals of members of their buddy lists hopping off and on the network), the load becomes significant... both on individual handsets, draining battery life (which Agile Messenger does, noticeably from my experience), and overloading the radio network.

SMS doesn't involve transmitting presence information, so doesn't suffer from the same issues. That's my understanding of it anyway - and I trust the guys at operators who actually run radio networks and know more about em than me ;)

Like you, I didn't realise there were technical barriers to IM adoption - which is why this was all news to me...

SergioFalletti

I was just reading today about TIM launching a full IM service in Italy. It includes presence applied to the whole phonebook.

Let's see if it gets mass adoption in Italy... and then what impact it has on the network.

Tom Hume

Yeah, I'm trying to find more about how they're doing it but the press I've seen is short on details...

Bertil

Couldn't the presence info be updated whenever you click on a contact, prior to calling?

In details:

Consider someone passive; his info are updated anytime they change:
- if he types in a new state (e. g. "Meeting");
- and that's it.
After five minutes of not being roamed back to the network, the server sets him to "Not available": this is already done.

Consider someon who wants to contact a friend:
- he scroll his buddy list, and click on a name;
- the cell phone pings the terminal: "what about that guy?";
- server comes back with a quick bit: On/Off & a 16 character state.
Then IM, voice or whatever can happen. This wouldn't take far more info than at present---however, it wouldnot allow to see the details of your buddy list in real time. . . Do we need that? Or rather: how is that bad?

Tom Hume

Bertil: you could do it that way, but you lose a lot of the value of presence. It's not just about actively finding out the status of a contact, but having this information come to you.

The point is that once you have a sizable number of contacts (10+ or so, say), then the continual "I'm here", "I'm away" messages start to make a difference...

Anders Borg

While this discussion is going on there will be companies making money from providing IM solutions, ignoring whatever network limitations there might be.

3G networks can't be so under-powered that 100 "pages" (packets?) per user is any problem. If it is then operators have done a serious misjudgment of how Internet services behave, and what network capacity is needed.

"but I think this is down to signalling rather than data transfer"

No. Also IM presence information is transferred over IP, hence data transfer.

Tom Hume

Anders: if they're ignoring network limitations which might exist then they're going to hit problems at some point :)

I'm interested to find out more about the TIM launch of IM and 3's success, and see what's different about IM in these cases which lessens the impact it has on battery life and network performance.

I can see why current networks might be designed around supporting services which comply to current models (i.e. voice and text); and it's not just a problem at a network level, but also on the device wrt battery life. As we know, battery life has an impact on services which use, say, mobile AJAX... or long-running bluetooth connections, or long-running Java applications.

And the fact that presence information is transferred over IP isn't completely relevant here. The handset either has to maintain a permanent IP connection, or be alerted as to when it should set one up. In the former case, there's a clear drain on battery life; in the latter I can see that signalling (in the SS7 layer perhaps?) would have an impact.

Anders Borg

"if they're ignoring network limitations which might exist then they're going to hit problems at some point :)"

I agree that it's the operators' task to secure that pure IP traffic runs as efficiently as possible over mobile networks. That's what 3rd-party service providers need.

"see what's different about IM in these cases which lessens the impact it has on battery life and network performance"

Do you mean you know there's anything different?

3 uses OZ's solution, so the best would be to check with them. The site doesn't provide much details: "It adapts/translates IM protocols and features to efficiently match the capabilities of the mobile network and the OMA IMPS compliant handsets of the subscribers."

Presence info could optionally be transferred via SMS. If an operator hosts the service there could be no cost for the end-user.


willyn angelique

im on sms
when i want to sign out

The comments to this entry are closed.