Re: http://www.w3.org/2008/02/MS4D_WS/papers/unicef-w3c-presentation.html

Hi Stephane and Kai and all,

Couldn't resist jumping into the interesting debate. Have a couple of
points to make.

> So VoiceXML has nothing to do with the browsable Web. And you are
> saying that SMS has also nothing do with with the browsable Web too?

I've the feeling that we are somehow in a rathole.
it all depends on your definition of browsable Web.
voice browsers are not generic browser on which you can get any URI
existing on the web. A voice browser is attached to a specific page
(like you have a home page) but the difference is that you cannot change
the uri (no address bar compared to a visual browser) so your browsable
web is what you can reach from this home page.
This is the web but with some limitation.
It is almost the same concept for widgets
Same concept i see for SMS

There are essentially two aspects that are getting mixed up here. One is
the technical feasibility of using SMS, voice etc., while the other is
usabiity and practicality of the solution. So, I agree with Stephane that
SMS for browsing the web is probably not the way forward but would like to
clarify that it is the latter reason rather than technical feasibility
being an issue here.

Specifically, with a slightly smart application that could be made
available on a device capable of running simple apps (envisioned to be
available to underprivleged in a couple of years time), it is very much
possible to present a very usable interface to the end user for browsing
web. The app could use SMS as the data transport to fetch data to and from
the web hiding all its complexity from the user and provide some sort of
delayed browsing. Even though this would take the complexity of keying in
SMS messages out of the picture and use regular URLs, it would still not be
practically usable since we would be trying to work with an unreliable
channel to deliver near real time content suited for a reliable channel
(something similar to using UDP over IP instead of TCP over IP, for the
network oriented folks).

Having said that, we may still need to expand our notion of the Web. For
instance, the VoiceXML example has come up a couple of times in this
thread. And though the argument below that voice browsers are stuck to a
page and do not allow browsing the web is correct, the voice browser need
not be equated to a Web browser. It is still very much feasible to have a
Browser with voice as an interface that can have an 'address bar' and be
used for visiting URLs (i.e. phone numbers in this case) at will. If
interested, you could glance thru this two page poster paper
(http://www2008.org/papers/pdf/p1121-agarwalA.pdf) that aims to do just
that. In fact, we recently submitted a more detailed report on our
implementation of such a browser and if possible and if there is interest,
I shall try to make that report available soon. You can also read about a
voice browsable web at (http://
www.soi.wide.ad.jp/project/sigcomm2007/pdf/nsdr41.pdf )


>>What i meant is that you will find nobody to invest money/time/resource
in developing and setting up something that are just here to solve a
transient problem.

Completely agree here though if the 'transient phase' is more than few
years then it might still be worthwhile to invest if the returns can be
justified.

>>All forecast for now is that by 2015-2020 all phones
will have a browser, and 3g available everywhere for the price of gsm
today. So sms would not be a service platform anymore

That is correct but I have a slight disgreement to the conclusion that
seems to be implied here that a regular browser on the phone is the only
way forward. Even though the price point of the device will most likely
come down to become equal or lesser than the lowest end phones today, there
are still two reasons why we should investigate other ways to access the
web (though not specifically SMS). The first reason is a personal
observation (in India) that even today few people towards the lower end are
able to own a smart phone. However, while they can sometimes afford such a
device from a second hand market, they are unable to afford the service
cost associated with features that require extra payment - i.e. their usage
of the phone is budgeted and they are unable to afford data connection - at
least today. The larger and more important problem however, is that most of
these people (again specific to India and maybe Africa) is that many of
these are semi-literate/illiterate people and are not tech-savvy enough to
be able to use features such as browsing -even if it was affordable. And
changing the literacy and education levels of entire countries may not
happen soon.


Having said all of the above, I would also like to state that it is
probably not at all important to get into a debate as to what is web and
what is not. IMHO our goal should probably be how to make information and
services (and web being the current de facto repository of those)
accessible to the underprivileged whichever best way (or ways) that come
up. Mobile Web then becomes the pioneering effort in that space but taking
all other solutions along for the benefit of the society.

Hope some of this helps.

thanks and regards
Arun Kumar

---------------------------
Research Staff Member,
IBM India Research Laboratory,
4, Block - C, Institutional Area, Vasant Kunj,
New Delhi - 110 070, India,
Email: kkarun@in.ibm.com
Tel: +91-11-66192100 / Fax: 91-11-26138889
http://www.research.ibm.com/people/a/arun

World Wide Telecom Web (aka Spoken Web):
http://domino.research.ibm.com/comm/research_people.nsf/pages/arun_kumar.WWTW.html



                                                                           
             Stephane Boyera                                               
             <boyera@w3.org>                                               
             Sent by:                                                   To 
             public-mw4d-reque         Kai Hendry <hendry@iki.fi>          
             st@w3.org                                                  cc 
                                       public-mw4d@w3.org                  
                                                                   Subject 
             11/06/2008 04:32          Re:                                 
             PM                        http://www.w3.org/2008/02/MS4D_WS/p 
                                       apers/unicef-w3c-presentation.html  
                                                                           
                                                                           
                                                                           
                                                                           
                                                                           
                                                                           






> So VoiceXML has nothing to do with the browsable Web. And you are
> saying that SMS has also nothing do with with the browsable Web too?

I've the feeling that we are somehow in a rathole.
it all depends on your definition of browsable Web.
voice browsers are not generic browser on which you can get any URI
existing on the web. A voice browser is attached to a specific page
(like you have a home page) but the difference is that you cannot change
the uri (no address bar compared to a visual browser) so your browsable
web is what you can reach from this home page.
This is the web but with some limitation.
It is almost the same concept for widgets
Same concept i see for SMS

> Wouldn't it be far simpler if they could send:
> GET tiempo.bo

well from the user perspective you might be right for you. this would be
more powerful for you because you believe that you can find the right
uri to put in, with the right parameters.
But if you have never experienced the web, or if you cannot use anything
else to retrieve the information about this URI and its parameter, then
no it is not more powerful, neither it is easier. I would prefer sending
an sms with "temperature cochamba" and get the info rather than "GET
tiempo.bo?type=temperature&location=cochamba"

That said, nothing prevent anybody to develop such a service.
There are very few chances that this would work based on the limitation
of sms and the issues i mentionned above. for voice, i did myself wrote
such an application being able to enter with keypad any uri, and then
translate the content at the uri in voicexml. I let you imagine what
kind of output you get, and how usable it is.

> Else if people get used to some proprietary text service, how are they
> to know the source of their information when they go to the Web on a
> desktop?

errr. that's all the point. this is about delivering services in places
where there is no web-enable phone nor internet cafe or similar
telecenter. So this is not for the case that people are in front of
their desktop from time to time or with their phone.

>> this will never happen.
>> neither will happen the development of a very lightweight (text-only)
>> browser, something i believed in for a while, mostly because this would
mean
>> investing in a direction that is not the trend. The trend is 2g, 3g 4g
>> networks, and full mobile browser.
>
> I don't quite understand your argument.
>
> So SMS text responses of URI requests won't happen because people
> expect the Web to be of the full sort?

What i meant is that you will find nobody to invest money/time/resource
in developing and setting up something that are just here to solve a
transient problem. All forecast for now is that by 2015-2020 all phones
will have a browser, and 3g available everywhere for the price of gsm
today. So sms would not be a service platform anymore

Steph


--
Stephane Boyera                      stephane@w3.org
W3C                                              +33 (0) 5 61 86 13 08
BP 93                                            fax: +33 (0) 4 92 38 78 22
F-06902 Sophia Antipolis Cedex,
France

Received on Thursday, 6 November 2008 13:52:16 UTC