See also: IRC log
<DKA> http://docs.google.com/Doc?docid=dd3jk8v_89f6vrqk9w&hl=en
<MartinJ> http://www.w3.org/2005/MWI/BPWG/Group/Drafts/BestPractices-2.0/ED-mobile-bp2-20080303
DKA: There has been a bit of dribbling...
We start on BP2...
DKA: We need to spend some time on issues, actions and coffee
JR: 2 1/5 hours on BP2?
BS: I was making changes as we
were going yesterday. The major things not yet there are the
ETRI input and the references to other "best Practice"
documents.
... most changes are in section 5. I tried to use a low-tech
way of saying things where possible. Some needs more
wordsmithing, but would be good to go through and look at what
was there.
[Bryan shows the list of things that are indicative of what "Web Applications" are]
<jo> Changes to BP 2 since yesterday
BS: These things will need to be
reflected back as requirements in the earlier section on
objectives
... there are a bunch of things to do with efficient use of
network. I already had a section on conserve network
traffic.
... 5.5.2 has changed to be more general.
JR: The notion of HTTP
compression needs to be clarified
... HTTP says you can use gzip, so do...
CMN: Gzip has a cost in terms of memory/processing, so we need to show this is valuable or raise an issue on it.
JR: We want to say "use HTTP to note that you compressed (and how)"
DKA: We may also want to
acknowledge the work of EXI (efficient XML interchange).
Without requiring the use of EXI (which is not supported) we
should encurage its use.
... EXI has been designed to minimise cost of compression
BS: I have explanatory text that explains the mechanisms, and that it needs to be balanced against resource usage etc.
<Zakim> chaals, you wanted to say we shouldn't say that until we have implementation proof
CMN: EXI cannot be anything more than an informative reference if we don't have an implementation...
JR: BS pointed out that tokenisation is an efficient approach to compression.
BS: Or WBXML - widely used in WAP1.
DKA: The thing behind WMLC?
BS: Yes
... so do you compress at application level or transport
level?
DKA: Does that mean WBXML is supported by browsers that support WML?
<scribe> ACTION: Chaals to check if Opera supports WBXML in and/or out of its WML support [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action01]
<trackbot-ng> Created ACTION-688 - Check if Opera supports WBXML in and/or out of its WML support [on Charles McCathieNevile - due 2008-03-11].
BS: We can reference EXI?
JR: Non-normatively
... It is probably worth calling out application level and
transport level compression.
<DKA> http://www.w3.org/TR/2007/WD-exi-best-practices-20071219/
BS: can we say "and other proprietary techniques"
CMN: No, those are not relevant to the Web
JR: We need to say "if the device supports it"
CMN: We should be supporting widely-used and standard mechanisms - pro gzip because it is deployed and standard, EXI perhaps because we expect it to be good, not OBML because you haven't got a sec for it anyway...
DKA: Agree. Think we can do that by ordering the way we talk abut these. When talking about XHR, do we need to make specific mention about compression for those types of transactions?
CMN: XHR doesn't currently have a way of gzipping AFAIK, although there was a request to WebAPI at one point to build an API for this purpose.
<jo> ACTION: Chaals to check XHR compression [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action02]
<trackbot-ng> Created ACTION-689 - Check XHR compression [on Charles McCathieNevile - due 2008-03-11].
DKA: Isn't it possible to set this though XHR?
CMN: Believe not - will check.
BS: Will add a note clarifying
that it may not always be possible to compress data.
... In the content, I added some stuff to clarify when it
does/doesn't help to make transactions.
JR: There is something that could be said about having regard to the type of connections available. Question is whether you can determine that
DKA: That's something about access to the device context at application level. You can infer it, but you don't know yet.
JR: There is probably something we can look at.
BS: The objective in OMA is to create a set of properties including bearer awareness, that should be available through DCCI
DKA: BP1 left the representation to the reader as an exercise. There are issues like cost of roaming that will continue - could be useful to suggest that you allow users to set the amount of access by some user-provided information
JR: It's a nice idea. It was watered down in BP1 on the basis that it wasn't actually practised, ergo couldn't be a best practice.
DKA: There is a preference in iPhone that says "don't use data when roaming". I wonder if we could give guidance to application developers to replicate that function?
JR: Think we should try to put some pointer in.
<jo> ack
BS: The networks that are being
used is information available to content providers, via cntext
information that is knowable.
... e.g. we can figure out how fast the network is running and
we forward that to the content provider
JR: So it seems there is some
information already...
... so there should be something that goes in there.
BS: [reads some example text]
BS: Push is widely deployed in networks. As a way to do event-based delivery instead of polling you could use it to minimise network traffic
CMN: It's in WAP browsers
BS: Almost every phone browser is a WAP browser, and they implemented push
CMN: Push isn't in "Web" browsers - full internet as opposed to WAP browsers
JR: There are two cases here... on the one hand WAP push and on the other hand application binding to incoming SMS to generate event based behavior
BS: There is a problem of education. There is no generic way to bind SMS to an application, and that would be a proprietary implementation detail.
JR: ..."if the device supports it".
BS: The only standard method I know of is MIDP registry.
JR: To my mind there is a distinction between WAP push and bindings on information pushed to an application. Maybe more information is needed...
BS: We haven't said that MIDP-based browsers are not in scope...
CMN: No. But I think it is clear that MIDP-based stuff is not readily within the scope of Web stuff - it is a particualr platform, in the same way that ActiveX relies ona aprticular platform, rather than being a general Web technology.
JR: Think we need some more research to look at this... would someone like to take an Action?
<jo> s/ona aprticular/on a particular/
[pregnant silence]
<jo> ACTION: JR to raise Issue as to availability of binding to incoming SMS from script [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action03]
<trackbot-ng> Created ACTION-690 - Raise Issue as to availability of binding to incoming SMS from script [on Jo Rabin - due 2008-03-11].
BS: Overall intention of this is
to bring awareness to the differentiated delivery methods
brought about by push technology. I think it is in our interest
here to promote push technology however it is implemented - it
does have a number of advantages.
... we can consider other possible approaches.
BS: Added a bunch of text here
JR: Need to avoid repeating BP1.
I was dubious about Adam's point about nested selectors,
overuse of generic class and so on.
... In respectof the technique Adam suggested, how widely
recognised is it?
... it is a disaster for maintainability if your content
changes shape, and requires support for CSS 2...
MJ: About half of Volantis code is about optimising CSS - we consider it implemented best practice.
JR: Curious about how you deal with changes to structure and the impacton CSS
MJ: We do some analysis, but we
are essentially generating the CSS anyway. The benefits tend to
outweigh the costs.
... but I do take the point.
JR: Seems that some element of
the benefit needs to be balanced against the cost of
maintainability.
... As long as you are at th top of the class structure, things
are fine.
DKA: Sounds like we ought to include "something"...
JR: Yes, we seem to be coming to recommend this, but with caveats.
BS: Should optimisation be done during delivery as opposed to up-front? It seems that reduces the cost....
JR: Yes, I think that is a good technique to use.
BS: There is also the use of markup, rather than script
CMN: Think that this si a seperate BP about not playing with the DOM more than necessary
JR: This is a specific technique for CSS, and something about dynamic changes to HTML. Leet's split these out
RF: Didn't Aron have a counter-case?
CMN: I think there is an edge-case (table processing?) that needs to be noted in a caveat
BS: Seemed to relate to both
network impact and overall size.
... impactis data use and service latency
JR: Should be balanced against the possibility of caching script files across pages.
MJ: There is a danger reading the text taht people decide to put the script files into page content, destroying the ability to cache the scripts
BS: Caching diminishes the ovrhead problem this is trying to deal with, right?
CMN: Don't we have this in BP 1?
JR: We did this in reference to CSS.
DKA: I think we are going to run up against a lot of things where we are elaborating BP1. I think that's fine.
CMN: So we should put specific pointers back to things we are elaborating from BP1.
JR: We need to discuss
terminology. "main page" is not terminology consistent with
BP1.
... Where we refer to resources, etc.
BS: I have been trying to use common language.
JR: There is a decision to be made. Consistency with our existing teminology has value.
DKA: It is important to get the
terminology right. We produced a document that was too hard to
read, and we should be looking to do better.
... We really need to keep the target audiencein mind, and not
maintain terminology from BP1 as sacred.
JR: If we have back references this needs to be clarified
<Zakim> DKA, you wanted to make a note about readibility
BS: In using the terminology like application, pulling focus away from presentation in a page as the model, we are inherently expecting the developer to stretch their though. As long as we explain up front what we mean, developers should understand and we should be able to use langauge as straightforward as possible.
JR: think we need to raise an issue on terminology.
<jo> ACTION: Dan to raise issue and start discussion on main page, external resources and so on [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action04]
<trackbot-ng> Created ACTION-691 - Raise issue and start discussion on main page, external resources and so on [on Daniel Appelquist - due 2008-03-11].
CMN: agree with Dan that clearer language is better than consistency with BP1. But we should point back to the ideas that are the same, and bring that out.
FD: Think it would be good to have more examples - bits of code that developers can look at.
JR: Now, rub some more salt
in...
... That was what the techniques wiki was meant to do. It is a
great idea, but you need the samples first
DKA: We thought building a Wiki would mean we would get the code. We could take a strctured approach and action people more directly.
JR: All a great idea. Show me the
content.
... Should cross-reference BP1 WHITE_SPACE
BS: Slotted stuff that could go into various boxes into presentation
DKA: We should make time in the agenda to look at the stuff from Jonathan.
[BREAK]
<rob> scribe: rob
<JonathanJ> http://lists.w3.org/Archives/Member/member-bpwg/2008Mar/0003.html
<scribe> scribenick: rob
Sunghan: about user-device
interaction and experience
... IP access from any other devices, as well as Web
<francois> Sunghan input
Sunghan: eg using PC to send
message to web that's received and replied to by a phone
... "seamless" means service mobility between different
networks
Dan: how can we take this input
and distill it into statements of Best Practice?
... ie techniques to ensure seamless access to content?
Bryan: BP1 recognised thematic
consistency
... already contains "multi-screen" environment.
<Zakim> chaals, you wanted to note that it raises one web as part of scope
Bryan: but switching between them is perhaps new
Chaals: missing from current scope is "kind of things in BP2 scope are things that will work on the web" [as well as on the mobile]
Jo: to get into BP2 need "do
this... don't do that..." statements
... do you mean "make your content work in all contexts" or
"design your content with all user interfaces in mind but don't
ignore focus of content in different environments"?
Sunghan: I've scoped the problem,
not proposed the solution techniques
... ie users will use PCs and mobiles together - not one or the
other
Jo: MWI steering council is asking why a seperate mobile apps WG? shouldn't we talk about how different interfaces work together?
<Zakim> chaals, you wanted to note that something this implies is "allow people to identify themselves as being on different devices"
Jo: so this is a good hint to address how users make use of different interfaces co-operatively
Chaals: allow users to identify
themselves as the same person on different devices?
... eg identity on Opera mini on a phone = Opera on a PC
Bryan: user with services on
desktop PC and mobile is commonplace
... a BP is that these views should synchronise in as timely a
manner as possible
Dan: "thematic consistency"
became an improtant BP1 principle
... maybe this can be a guiding principle for BP2?
Jo: so can we turn it into
actionable statements?
... highlighting BP1's "3.1 One Web" statement
... hints relationship betweem mobile-desktop-other screens
Sunghan: this BP1 paragraph highlights mobile-desktop relationship. BP2 could go further and consider more user interfaces.
Jo: need to constrain ourselves
to mobile, eg we're not inventing new sync techniques
... can we action Sunghan to develop some actionable
statements?
Bryan: eg "these views should synchronise in as timely a manner as possible..."
<jo> ACTION: seunyung to provide some example BP statements based on the presentation he gave at Korea F2F [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action05]
<trackbot-ng> Sorry, couldn't find user - seunyung
Jo: Dan & I will ask in UWA WG as well
<jo> ACTION: seunyun to provide some example BP statements based on the presentation he gave at Korea F2F [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action06]
<trackbot-ng> Sorry, couldn't find user - seunyun
<francois> ACTION: sunghan to provide some example BP statements based on the presentation he gave at Korea F2F [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action07]
<trackbot-ng> Created ACTION-692 - Provide some example BP statements based on the presentation he gave at Korea F2F [on Sunghan Kim - due 2008-03-11].
<Seungyun> Sunghan is not on the IRC
Dan: did we finish the discussion on URI protocol schemes yesterday?
<jo> ACTION: Dan to raise issue with Dave Raggett in UWA and see if they will take forward from where we leave off [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action08]
<trackbot-ng> Created ACTION-693 - Raise issue with Dave Raggett in UWA and see if they will take forward from where we leave off [on Daniel Appelquist - due 2008-03-11].
Bryan: yes, WRT tel: scheme BP2 doc is where we'll pick up now...
Jo: thanks Sunghan for the contribution
Bryan: actionable statements like "include the phone number as text in the link so you know who you're about to call"
Dan: and "use tel: or wtai: URI scheme as an easy way to make phone calls"
Jo: my view is a 2-step process:
(1) write it in our language and then (2) express it in
content-provider language
... so rather see it as a high-level statement for now
Dan: surely we can narrow it down right now?
Bryan: maybe I went too far to
specific actionable statements in this edit
... perhaps a generic statement plus some specific examples (eg
phone calls, send a message, ...)
<Zakim> DKA, you wanted to suggest "actionable information" as a replacement for "things"
<chaals> ["Use links like mailto:, tel: etc to help users perform relevant actions"]
<jo> "Remember that hyperlinks can be used to initiate device specific actions"
<chaals> ["Use link types like... "]
Dan: suggest "actionable information" as a replacement for "things" - we're not talking about hyperlinks that go somewhere else but about links that pop-up some additional action
Bryan: I'll propose some text to tweak
<chaals> ["Use link types like mailto:, tel: etc to help users perform relevant actions"]
Jo: sould go back to doc review now
Dan: are Jose's inputs incorporated?
Bryan: yes
Jo: ok, remember ideas are not copyright but the text is.
<francois> Apple iPhone resources and tools for developing web apps
Dan: advice about the viewport and aspect ratio is useful
Bryan: does Apple have IPR on this?
Chaals: we're not sure
Jo: caution is advisable but
Apple's guidelines are public advice, there's no license to
sign
... would prefer not to have to go to Apple for permission to
publish BP2
Dan: advice about touch-screen might be useful
Rob: eg consider people will often use their thumbs, so "mouse" pointing isn't accurate
Dan: how about "think windowless"?
Bryan: BP2 has a bullet list of presentation and interaction issues that captures a lot of this
<JonathanJ> I think iphone's "windowsless" is good idea.
Bryan: we can expand that list with stuff from here
Jo: can we get more general or do
we need specific actionalble techniques?
... if Bryan's already got the bullet list we'll continue on
that line and see how it goes
Bryan: video encoding for bearer is relevant
Dan: and preparing alternatives in advance suitable for different bearers
Jo: caveat that you might not know about bearer
<jo> [break for lunch]
<JonathanJ> my contribution : http://lists.w3.org/Archives/Member/member-bpwg/2008Mar/0003.html
<SeanP> scribe: SeanP
<scribe> scribenick: SeanP
DKA: Spend a couple more minutes examining Apple doc.
Jo: Need to look at Jonathan's contribution.
<jo> ACTION: Dan to review apple document and summarise the parts that might be applicable to BP2 [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action09]
<trackbot-ng> Created ACTION-694 - Review apple document and summarise the parts that might be applicable to BP2 [on Daniel Appelquist - due 2008-03-11].
<jo> Jonathan's Doc
<JonathanJ> please open attatched file : http://lists.w3.org/Archives/Member/member-bpwg/2008Mar/att-0003/mwi-200803-DDC-v1.0.doc
Jonathan: Proposal for ADC for
BP2.
... DDC is minimal delivery context for Mobile Web.
... [Goes through DDC requirements]
... K-DDC is delivery context for K-MWBP.
... MW2F K-Mobile OK is similar to W3C Mobile OK.
... [Explains diagram that compares W3C and MW2F in the
document]
... We are developing K Mobile Best Practices 1.5
<scribe> ...New advanced delivery context in Korea is KDDC 1.5
<scribe> ...New features were needed for KDDC
UNKNOWN_SPEAKER: [Describes table
comparing DDC and KDDC 1.5]
... Some of the new features: HTML 4.01, EUC-KR, PNG, 50K max
page size, CSS 2.1, JavaScript 3, XHR, SSL, DOM stuff
DKA: We said before that we
didn't want an ADC; doesn't mean that this isn't useful. Still
need to have an ADC in mind when creating the new BPs.
... Reason we said we didn't want an ADC is we didn't want to
open ourselves up to criticism since it will become
obsolete.
Rob: Is this DDC 1.5 finished?
Jonathan: is finished
Rob: BPWG can nod to one that already exists.
Jo: We have had a problem with
the perception of the DDC. We could have the same criticism of
the ADC.
... Point of DDC is the minimum delivery context for reasonable
experience on the Web...Designed to avoid criticism.
... Omission of PNG was a mistake.
... If you know nothing else about the device, assume the
DDC.
... BP1 also says exploit device characteristics.
... BP2 will explain how to exploit device capabilities.
... Each one of the BP2's will say if this feature exists,
exploit it in this way.
... DDC is not about a point in time, etc. BP2 doesn't need
that idea because you need to find out what the device is to
exploit the capabilities.
... What dependencies does the K BP1.5 have on the ADC?
DKA: What is in K BP 1.5 that
goes beyond the BP 1.0 that we could put in BP2 that would be
based on the ADC?
... Are there BPs related to CSS 2 or JavaScript 3?
Jonathan: We need more advanced features.
<JonathanJ> Seungyun
Jo: The assumption of BP2 is that
you know what the device is.
... You should have a way of find out what the capabilities of
the device are.
...Ex: for the tel: URI we will find out whether the device
supports it before using it.
... Wasn't an easy decision to drop ADC, but once it was done
it made sense.
Bryan: There is nothing in the
DDC that says that HTTPS is not supported.
... We have a DC context variability section in BP2.
Seungyun: Is there any
relationship between DDC and BP2?
... Won't have ADC in future?
... What kind of DDC in BP2?
Jo: We haven't got consensus yet
about whether we should update the DDC for BP 2.0.
... Personal view is that we shouldn't update it.
...Exception: PNG could be added.
DKA: I don't think that BP2 is
written in the context of the DDC. It is all about knowing
device capabilities.
... Do you have statements in the K BP 1.5 written in support
of, say, XHR?
... Is there something we can pull out of that doc and put into
BP2?
<jo> ACTION: Jonathan to extract BP statements from K MWBP 1.5 document for consideration in BP 2.0 [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action10]
<trackbot-ng> Created ACTION-695 - Extract BP statements from K MWBP 1.5 document for consideration in BP 2.0 [on Jonathan Jeon - due 2008-03-11].
<MartinJ> Agenda: http://docs.google.com/View?docid=dd3jk8v_89f6vrqk9w
<JonathanJ> Helsinki file : http://www.cs.helsinki.fi/u/pervila/Gradu/index.html
DKA: Anyone want to present University of Helsinki Master's Thesis and Frost Ajax library?
Francois: Will summarize.
... Would be interesting in having the guy who wrote it
summarize it to the group.
<DKA> http://www.cs.helsinki.fi/u/pervila/Gradu/index.html
<jo> ACTION: Daoust to summarise the U Helsinki masters thesis on Mobile Ajax performance with a view to including some aspects into BP 2 [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action11]
<trackbot-ng> Created ACTION-696 - Summarise the U Helsinki masters thesis on Mobile Ajax performance with a view to including some aspects into BP 2 [on François Daoust - due 2008-03-11].
<DKA> http://www.pavingways.com/frost-ajax-library
DKA: Frost Ajax library is for
constrained browsers. Javascript and server side component that
sends appropriate JS based on the type of the device.
... Would be interesting in seeing if we could promote that
idea.
... Developer has been active creating comments although not a
group member.
Bryan: Is there a summary report on this activity?
DKA: Someone needs to summarize
what can be pulled from this library for use by BP2?
... This pattern of using server side device detection combined
with a modular JS library...need to determine if this could be
a useful technique in BP2.
<JonathanJ> Summarize article: http://ajaxian.com/archives/measuring-the-state-of-mobile-ajax-performance
Bryan: They have some detailed information that has been put into WURFL and they use that to select JS?
Martin: ... We use a similar technique to determine which JS to send to the device.
Bryan: Does this use a test to find out the JS capabilities?
DKA: No, just uses the type of the device.
Jo: Runs an automated test on the
browser.
... It's run once; don't need to do it every time.
DKA: Sounds even more relevant--we could document what those tests are. We could say use this script to find the characteristics.
Chaals: This is a very simple test. It does 3 tests. Not sure how much time it is worth looking at this.
<chaals> [whee! Opera Mini passes all its tests :) ]
Jo: Agree with Chaals. We should have something about level of support and simple tests. Need to capture that it is a best practice to do this.
Bryan: We already have a placeholder for this. Have a section on JS reflection.
Jo: The second point is to customize the download of the JS based on the device.
DKA: I think there is still an action here.
<jo> ACTION: Dan to summarize the points he can glean from examination of the frost library [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action12]
<trackbot-ng> Created ACTION-697 - Summarize the points he can glean from examination of the frost library [on Daniel Appelquist - due 2008-03-11].
<DKA> Agenda updated: http://docs.google.com/Doc?docid=dd3jk8v_89f6vrqk9w&hl=en
<francois> Scribe: francois
<scribe> ScribeNick: francois
Jo: agenda is to talk about
mobileOK, but before that, I'd like to come back to Jonathan's
input on ADC
... The thing is we need to update DDC for BP2
... for instance, support for PNG may be assumed
... The suggestion is BP2 contains a revised version of DDC and
reviewed BPs of BP1
Bryan: The presence of DDC without saying that it does not limit the best practices of BP2 might lead to confusion
Jo: Yes, we need to be clear that
it's the minimal delivery context, not the target but the
baseline
... If you know nothing about the target, then assume (revised)
DDC
... Other than PNG, I don't really think DDC needs
changing
... I'll raise an issue on that
<jo> ACTION: JR to raise issue of revising DDC and to raise discussion of the revised definition being retroactive to BP1 [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action13]
<trackbot-ng> Created ACTION-698 - Raise issue of revising DDC and to raise discussion of the revised definition being retroactive to BP1 [on Jo Rabin - due 2008-03-11].
Jo: In the light of discussion,
it occurred to me that there is a set of properties that you
need to rely on when reading and writing the BP2 doc
... I think it's useful to list these DDR
properties
<jo> ACTION: Bryan to insert an Appendix listing the Device properties that BP2 is dependent upon [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action14]
<trackbot-ng> Created ACTION-699 - Insert an Appendix listing the Device properties that BP2 is dependent upon [on Bryan Sullivan - due 2008-03-11].
Jo: Also, I think it's clear we
don't need an ADC, but what would be great is to have a list of
tiers/classes
... so that when you develop an application, you may target
these different classes
... It seems that this is established practice. I've heard
people saying "I have 4-5 classes"
DKA: yes, I'm interested to see where that goes
Jo: I'm hoping that having a best practice around specification in that way addresses concern from persons such as Jonathan
Bryan: to summarize, you would recommend to have a BP to recommend classification, but not list the definition of the classes, right?
Jo: yes
<jo> ACTION: Bryan to introduce a BP on classification of devices into High, Mid, Low etc on a per application basis - with an extended non-normative example, pethaps [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action15]
<trackbot-ng> Created ACTION-700 - Introduce a BP on classification of devices into High, Mid, Low etc on a per application basis - with an extended non-normative example, pethaps [on Bryan Sullivan - due 2008-03-11].
Bryan: I don't have yet any inputs on:
1. toolkit developers may use
scribe: we'll have to address the impact of toolkits
2. how are we going to address the techniques and practices for non-browser applications
Jo: I don't think we need to talk
about the execution environment
... the environment is not especially relevant to mention
Jo: who knows what it is?
Audience: [smiles]
Bryan: There will be a public value to know that a site is mobileOK, and so we should find a way to mark pages as presumably mobileOK
Dan: mobileOK Scheme is the name we gave to a set of documents, and encompasses mobileOK, tests, the checker
Jo: plus it's a set of usage
rules
... we have failed, as a group, to make a start on the
doc
... We really have to do something for this in my view.
... What does the teams think?
... Do we need a mobileOK scheme document?
... Do we need another doc that says how to use mobileOK, which
icon to use, when to use it, ...?
Dan: Yes, we need it, otherwise mobileOK won't be used in public
chaals: Does it need to be a TR doc though?
Dan: no, it doesn't.
Seungyun: from a Korean's
perspective, we really need that scheme.
... We hope W3C will address that doc
Jo: the basic problem is we don't
have an editor for the doc.
... and so without editor, no doc.
... resolution to find an editor for the doc!
<jo> PROPOSED RESOLUTION: Chaals to edit mobileOK scheme document
<jo> +1
+1
<DKA> +1
<SeanP> +1
<Seungyun> +1
<jo> RESOLUTION: Chaals to edit mobileOK scheme document
Jo: I think we should cover a basic content list here and let you work on this.
chaals: I believe there is a working draft
<chaals> source for inspiration
Jo: If I remember correctly, this
is the pre-split version
... what can we extract from this?
chaals: what is mobileOK
... how does mobileOK relate to best practices?
Jo: where do I find license
info
... it also needs to discuss the checker
... there needs to be some fairly tight wording in here
... it needs to discuss that the checker is non normative
although it is a ref implementation
... it should answer the trustmark question
... dom has an on-going action to check with legal team
Dan: I think there should be a kind of usage scenario for content providers for instance
chaals: yes. How does mobileOK relate to me?
Jo: right. What are the benefits.
<jo> ACTION: Dan to write a usage scenario for mobileOK scheme [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action16]
<trackbot-ng> Created ACTION-701 - Write a usage scenario for mobileOK scheme [on Daniel Appelquist - due 2008-03-11].
Jo: when do you think you can have a first editor's draft?
chaals: [thinking hard]. Not tonight. Not tomorrow.
Jo: by next Thursday?
chaals: OK
<jo> ACTION: chaals to produce first editors draft of mobileOK scheme in 1 week [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action17]
<trackbot-ng> Created ACTION-702 - Produce first editors draft of mobileOK scheme in 1 week [on Charles McCathieNevile - due 2008-03-11].
Dan: is there another issue that
we should take on upon which is about mobileOK Pro and mobileOK
Basic
... whether or not mobileOK is synonym to mobileOK Basic
chaals: I would say over my dead
body. But that is too easy to arrange
... I think it would be a shame
... I'm not sure it's worth discussing. It will be associated
with whatever people use most.
Jo: Are mobileOK usage rules packed with scheme?
chaals: the scheme should document them, but I don't think it's up to me to write them
Jo: what I think is worth discussing: we have a fundamental decision to make as to whether it's going to be meaningful as a trustmark or just a wishful label
chaals: the HTML label is
slightly more reliable than the WCAG label, which is just a
joke in the sense that there is no way to check conformance
automatically
... in the last century, W3C had a system where you could
report abuse of trustmarks
... It stops to be a trustmark, and starts to be a badge that
you may wear anywhere
... Two possibilities: W3C sets precise rules and tries to
enforce the use of the trustmark
... or not
... These are considered issues beyond the scope of the working
group
... We should basically say: this is what we're thinking, these
are some ways to ensure the trustmark is used correctly, and
have copyright rules to ensure that it has to be removed when
rules are not followed
Bryan: Trust is something that is
based somehow on intent, and there needs to be some recognition
of the good will.
... It's never going to be 100% followed
... we need to be more flexible
<Zakim> DKA, you wanted to ask about usage rules for K-MobileOK.
Dan: I would like to ask Jonathan
or Seungyun what usage rules were developed with
K-MobileOK
... do you have a set of rules as to when people can claim they
are mobileOK?
Seungyun: so far, we don't have
specific rules for K-mobileOK
... because we're waiting for W3C!
... We need some explicit rules from W3C in order to deploy
that in Korea. We only have some requirements for the
moment.
Dan: It sounds to me that there is an aspiration for a mobileOK button.
Jo: I really like this idea. We
might even resolve on that.
... One of the things that I think is important: mobileOK means
"you want to be mobile friendly" as Bryan crystalized correctly
before
<jo> PROPOSED RESOLUTION: There will be an aspirational level represented by a badge of mobileOK called "mobileOK Checked" that links to validator/mobile with default URI of referringpage
<jo> PROPOSED RESOLUTION: There will be an aspirational badge for mobileOK called "mobileOK Checked" that links to validator/mobile with default URI of referring page
<rob> +1
<DKA> +1
<jo> RESOLUTION: There will be an aspirational badge for mobileOK called "mobileOK Checked" that links to validator/mobile with default URI of referring page
<JonathanJ> this is a mobileok test page in Korea : http://test.mobileok.or.kr/
Jo: next point is the distinction
between the claim of the trustmark and the use of the visual
representation
... previous discussion said that the label was the claim
... and the visual representation is just informal
... from a technical point of view, it's important. From the
appelquistian simple point of view, it's too complex
chaals: we should accept the fact
that putting the badge is actually claiming that you are
mobileOK.
... I would suggest that we can do is that when you have a
badge, then you have to link it to a POWDER claim.
... so if I get to the badge and cannot find the POWDER claim,
then you're breaking the rules.
Jo: Again, back to the point about the fact that the visual representation may appear on content that is not mobileOK because of thematic consistency
francois: too complex?
chaals: if you want to claim
something is mobileOK, then you need to have a POWDER
claim
... if you put a badge in your page, you need to link it to the
POWDER claim
... should the badge link directly to the POWDER doc? Quite
possibly not, because that's not really useful from a browser's
point of view.
... so maybe it's linking to a page that contains a link to the
POWDER statement.
... If you serve 400000 pages adapted to delivery contexts that
are not DDC and not mobileOK but deliver mobileOK pages to DDC,
there's no reason why you can't add the mobileOK badge to all
of your pages.
<DKA> Scribe: Jo
<DKA> ScribeNick: Jo
<DKA> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/open
<DKA> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/pendingreview
jo: Ed said leave pending
jo: suggest reassign to chaals
[reassigned to chaals]
close ACTION-606
<trackbot-ng> ACTION-606 Detail reload re: section 2.1.5 original representation availability closed
close ACTION-607
<trackbot-ng> ACTION-607 Detail what he means by "reload" request on mailing list closed
jo: on me, leave pending
ACTION-621?
<trackbot-ng> ACTION-621 -- Dominique Hazaël-Massieux to check if W3C has a liaison with OpenAjax -- due 2008-01-24 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/621
ACTION-631?
<trackbot-ng> ACTION-631 -- François Daoust to check with Dom about test cases for ISSUE-234 -- due 2008-01-31 -- PENDINGREVIEW
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/631
close ACTION-631
<trackbot-ng> ACTION-631 Check with Dom about test cases for ISSUE-234 closed
ACTION-638?
<trackbot-ng> ACTION-638 -- Kai Scheppe to raise an issue on ISSUE: Does the TF need to create device which emulates the DDC for testing? -- due 2008-02-12 -- PENDINGREVIEW
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/638
[leave open]
ACTION-589?
<trackbot-ng> ACTION-589 -- Daniel Appelquist to look for one or more likely candidates to adopt techniques and make arrangements ref copyright and attribution -- due 2007-11-12 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/589
jo: have mailed offering dotMobi to take this over as part of dev.mobi, no response from Dom as yet
close ACTION-589
<trackbot-ng> ACTION-589 Look for one or more likely candidates to adopt techniques and make arrangements ref copyright and attribution closed
[DKA opens ISSUE-239]
ACTION-594?
<trackbot-ng> ACTION-594 -- Daniel Appelquist to coordinate mobileOK Basic advancement, probably starting with a teleconf -- due 2007-12-21 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/594
close ACTION-594
<trackbot-ng> ACTION-594 Coordinate mobileOK Basic advancement, probably starting with a teleconf closed
ACTION-603?
<trackbot-ng> ACTION-603 -- François Daoust to find out how to liaise with HTTP NG work -- due 2008-01-29 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/603
fd: if we really need to add
something to HTTP we will try to do that but that's not the way
they want to go, their scope is rewriting not creating new
stuff
... if anything, we need new stuff
Close ACTION-603
<trackbot-ng> ACTION-603 Find out how to liaise with HTTP NG work closed
ACTION-605?
<trackbot-ng> ACTION-605 -- Magnus Lönnroth to suggest some text for 2.1.2 -- due 2007-12-11 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/605
ACTION-605?
<trackbot-ng> ACTION-605 -- Magnus Lönnroth to suggest some text for 2.1.2 -- due 2007-12-11 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/605
close ACTION-605
<trackbot-ng> ACTION-605 Suggest some text for 2.1.2 closed
ACTION-613?
<trackbot-ng> ACTION-613 -- Jo Rabin to start collecting mobileOK web pages for mobileOK CR -- due 2008-01-10 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/613
jo: ongoing
ACTION-618?
<trackbot-ng> ACTION-618 -- Edward Mitukiewicz to review Scope of BP1 to see what it tells us about scope of BP2 -- due 2008-01-17 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/618
[leave open]
ACTION-619?
<trackbot-ng> ACTION-619 -- Alan Chuter to update the comments list to public-bpwg-comments -- due 2008-01-24 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/619
dka: did this happen?
fd: yes
close ACTION-619
<trackbot-ng> ACTION-619 Update the comments list to public-bpwg-comments closed
ACTION-621?
<trackbot-ng> ACTION-621 -- Dominique Hazaël-Massieux to check if W3C has a liaison with OpenAjax -- due 2008-01-24 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/621
Dka: ?
fd: ?
[leave open]
ACTION-625?
<trackbot-ng> ACTION-625 -- François Daoust to initiate discuss on the exception wording ref dangerous content -- due 2008-01-29 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/625
[still open]
ACTION-629?
<trackbot-ng> ACTION-629 -- Ignacio Marin to will ask group about having it the week before the Expo starts -- due 2008-01-31 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/629
close ACTION-629
<trackbot-ng> ACTION-629 Will ask group about having it the week before the Expo starts closed
ACTION-632?
<trackbot-ng> ACTION-632 -- Bryan Sullivan to propose some recommendation on user-agent detection from a proxy and browser's (format) point of view -- due 2008-02-05 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/632
BS: still open
ACTION-633?
<trackbot-ng> ACTION-633 -- Andrew Swainston to write a clear draft on @@allow-https-rewrite and the need for the end-user to be aware of the situation -- due 2008-02-05 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/633
[change to pending review]
ACTION-634?
<trackbot-ng> ACTION-634 -- François Daoust to write a note to say something about Cache-Control: no-transform and WAP gateways -- due 2008-02-05 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/634
[change to pending review]
ACTION-637?
<trackbot-ng> ACTION-637 -- Alan Chuter to check on which WCAG 1.0 checkpoints were dropped in 2.0 due to untestability. -- due 2008-02-12 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/637
[no response from Alan]
ACTION-640?
<trackbot-ng> ACTION-640 -- Phil Archer to draft test suite document to complement Test Document - such a draft may or may not be completed depending on its usefulness in the Test Document creation process -- due 2008-02-12 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/640
[skipping Pro Test docs as we don't know
close ACTION-657
<trackbot-ng> ACTION-657 Post a questionnaire re June f2f by Feb.21 closed
ACTION-660?
<trackbot-ng> ACTION-660 -- Bryan Sullivan to raise specific points of discussion on Public List -- due 2008-02-21 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/660
close ACTION-660
<trackbot-ng> ACTION-660 Raise specific points of discussion on Public List closed
ACTION-663?
<trackbot-ng> ACTION-663 -- François Daoust to set up a poll for BPWG due in one hour -- due 2008-02-28 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/663
close ACTION-663
<trackbot-ng> ACTION-663 Set up a poll for BPWG due in one hour closed
ACTION-664?
<trackbot-ng> ACTION-664 -- Yeliz Yesilada to provide some examples to put into the document - specifically on STYLE_SHEET_SUPPORT -- due 2008-02-28 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/664
<scribe> [pending review]
ACTION-665?
<trackbot-ng> ACTION-665 -- Alan Chuter to talk to Jeffs about what support they can provide on examples -- due 2008-02-28 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/665
[leave open]
ACTION-666?
<trackbot-ng> ACTION-666 -- Aaron Kemp to draft section 2.6 listing user control options that SHOULD be supported -- due 2008-03-04 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/666
[change to pending review]
close ACTION-667
<trackbot-ng> ACTION-667 Make 2.7 and 2.8 sub sections of 2.6 closed
close ACTION-668
<trackbot-ng> ACTION-668 Raise an ISSUE on labelling using POWDER describing transformation options on sites closed
close ACTION-670
<trackbot-ng> ACTION-670 Remove sect 3.1 and transfer semantics to the present 3.2 closed
close ACTION-671
<trackbot-ng> ACTION-671 Update wording of sect 3.2 p 2 to clarify that the intent is not to respond with a transformed copy closed
close ACTION-672
<trackbot-ng> ACTION-672 Adjust text in 3,2 per the previous note in the minutes closed
ACTION-673?
<trackbot-ng> ACTION-673 -- Aaron Kemp to see if he can get some figures that scope the problem of bogus 200 responses -- due 2008-03-04 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/673
[leave open]
close ACTION-674
<trackbot-ng> ACTION-674 Produce new draft based on the many actions he has taken during this call :-) before BP meeting on THursday closed
ACTION-677
ACTION-677?
<trackbot-ng> ACTION-677 -- Daniel Appelquist to create an issue to start bringing together potential test cases. -- due 2008-03-06 -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/actions/677
[leave open]
<DKA> Scribe: Bryan
<DKA> Scribenick: Bryan
Issue-222?
<trackbot-ng> ISSUE-222 -- TAG Finding on Alternative Representations -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/issues/222
Jo: this is talking about
alternative representations; there is a can of worms to be
opened here. The suggestions shown are interesting and we can
adopt them as BPs, but
... the problem is that no one does this, except for some
specific examples.
... as an initial step we can make recommendations, e.g. based
upon proposed texton the list from last year.
<JonathanJ> http://www.w3.org/2005/MWI/BPWG/Group/track/issues/open
Jo: i.e. since there are so many
optional representations, using distinct URI schemes seems
impractical.
... other points were made in the email, e.g. re 2.1.1 point 5
in the TAG finding, the reason it is problematic is while it
talks about HTML, we are interested in more than HTML,
e.g.
... Images and links for them.
... re redirection, we are averse to redirection but may
consider the 300 response as a way to do it, e.g. as discussed
in RFC 2295/2296.
... redirection is generally to be avoided since mobile sites
don't have static versions of their resources.
... re point E, its hard to know whether a URI points to a
specific representation or a resource with multiple
representations. The linkages that are suggested require that
there be a way to distinguish this.
... CT TF needs to consider resources linked as described here.
Google has mentioned this as a recommended practice.
... The question is what do we expect from TAG in response to
our note.
Dan: they could help us solve the problem.
Jo: they may ask us what is our answer instead
Dan: we already have some answers in the CT guidelines doc
Jo: we have enough to do already without invoking a discussion with TAG that might not bear fruit.
Dan: if we issue a document that contradicts we will get comments
Francois: Jo's points are valid,
and we should give them to the TAG.
... initial feeling is that we are both wrong, content
adaptation in the future will not be solved by HTTP links for
alternate representations
... the negotiation will be more complex than supported by the
link approach
Jo: from our likely recommendation that POWDER be used to describe resources, we need a link header for that purpose, and the meta information there would be helpful for CT
Francois: they could say that the web page returned (the ML) will address the appropriate links to available representations
Jo: will draft a note around
these 5 points for review and to initiate a dialog with
TAG
... will do this in the next couple of weeks
<jo> ACTION: Jo to draft a communication with the TAG based on ISSUE-222 [recorded in http://www.w3.org/2008/03/04-bpwg-minutes.html#action18]
<trackbot-ng> Created ACTION-703 - Draft a communication with the TAG based on ISSUE-222 [on Jo Rabin - due 2008-03-11].
Issue-223?
<trackbot-ng> ISSUE-223 -- Various Items to Consider for the CT Guidelines -- OPEN
<trackbot-ng> http://www.w3.org/2005/MWI/BPWG/Group/track/issues/223
Jo: for the shopping list, we may
be able to avoid the issue through the indirect intentions
being expressed through T&C's, that the CP's intent should
be respected, but
... since our job is to promote mobile web awareness, we should
assume that CP's intend to provide the mobile representation.
But the user may express a preference for the desktop
... version by using a user-agent switcher or via other
means.
Dan: since you can't do all desktop things on the mobile site, a user may want to use the deskop for those missing things
Jo: although we never said it, we
implied that its a good idea to match the user's context but
they should allow a return to desktop view since they may have
misunderstood the context.
... So we should say that CP's should provide a desktop view,
and give the choice to the user.
... user's preferences may normally be overridden because the
CP knows better, but the user can effect an higher priority
override when desired
... we need also to put a placeholder in BP2 that CP's should
make assumptions but should allow users to override the
assumptions
Francois: Aaron's contribution goes in that direction, e.g. user priority should be given
Sean: in other areas there are some precedents, e.g. CSS, where the CP stylesheet normally overrides the user, but the user can ultimately overrride the CP
<DKA> PROPOSED RESOLUTION: In matters of presentation, Content Providers' preference should take preference over user's preference, but user should be able to exert a high-priority override over the content provider's prefernece if desired.
Jo: that is a good example, and we could reference it
<francois> +1
<DKA> PROPOSED RESOLUTION: Regarding the CT Guidelines document: in matters of presentation, Content Providers' preference should take preference over user's preference, but user should be able to exert a high-priority override over the content provider's prefernece if desired.
<SeanP> +1
<rob> +1
+1
<jo> +1
<DKA> RESOLUTION: Regarding the CT Guidelines document: in matters of presentation, the Content Provider's preference should take preference over user's preference, but user should be able to exert a high-priority override over the content provider's prefernece if desired.
Jo: on point 2, how should the
user signal their choices, they can't. A new HTTP header is
required, or an application artifact needs to be created.
... the question is thus out of scope since it's new
technology
<DKA> PROPOSED RESOLUTION: Regarding the CT Guidelines document: the question of how the user signals their choice is out of scope.
Bryan: so we can also add a statement that the user should be given an option at the application layer, a link to switch modes
<DKA> RESOLUTION: Regarding the CT Guidelines document: the question of how the user signals their choice is out of scope as a signaling question, but in scope as an application or user interaction question and we recommend that both CT Proxies and origin servers provide user interactions to effect this.
Jo: re issue 3, this is done, we
have resolved that they should present original headers
... on point 4, this is all out of scope, as an area for
product differentiation
... on point 5, we have a workable compromise but need to
express it clearly
Francois: this links back to the "dangerous" question
Jo: the answer is multipart; first, user choice if only thru T&C's; there may also be apparently malformed content that is required, e.g. for non-browsers.
Dan: is the question here that we need two shades of transform control
<DKA> PROPOSED RESOLUTION: lksakllkasd
Jo: that is addressed through the inclusion of a POWDER declaration identifying the intent
<DKA> PROPOSED RESOLUTION: Regarding the CT Guidelines document: in the matter of overriding no-transform... a. it's a user choice (which may be delegated to a service provider) b. with appropriate CT detection of non-browsers apparantly malformed content will be left alone.
+1
<jo> +1
<DKA> +1
<francois> +1
<SeanP> +1
<MartinJ> +1
<DKA> RESOLUTION: Regarding the CT Guidelines document: in the matter of overriding no-transform... a. it's a user choice (which may be delegated to a service provider) b. with appropriate CT detection of non-browsers apparantly malformed content will be left alone.
<DKA> PROPOSED RESOLUTION: Regarding the CT Guidelines document: in the matter of overriding "no-transform but tidy allowed"... we're waiting for POWDER.
<DKA> PROPOSED RESOLUTION: Regarding the CT Guidelines document: in the matter of overriding "no-transform but tidy allowed"... the content provider preferences will be expressed via POWDER.
<jo> +1
<DKA> RESOLUTION: Regarding the CT Guidelines document: in the matter of overriding "no-transform but tidy allowed"... the content provider preferences will be expressed via POWDER.
Jo: re point 6, BPWG says content
should be tested, but this can't be claimed without testing via
CT proxies. Even with test houses, it would be helpful to
provide a more realizable means for CP's to comply.
... the facilities should be provided by proxy operators.
Chaals: Opera Mini provides an onine tool to verify how something will work thru Opera Mini.
Jo: suggests Operators of CT proxies should provide test facilities.
Bryan: and not CT product providers?
<DKA> PROPOSED RESOLUTION: Operators of content transforming proxies should provide test facilities for the benefit of content providers.
Dan: the proxies are setup per the business rules of the CT proxy operator.
Martin: believes this makes sense in some cases, for customized products it may not be practical for CP's to test thru all CT proxy operators.
Chaals: unconvinced that this has a place in the CT spec
<DKA> PROPOSED RESOLUTION: Regarding the CT Guidelines document: Operators of content transforming proxies should provide test facilities for the benefit of content providers.
Jo: we can't say that you must test without giving a means or directions how to do it
<jo> +1
<francois> +1
<DKA> RESOLUTION: Regarding the CT Guidelines document: Operators of content transforming proxies should provide test facilities for the benefit of content providers.
<DKA> PROPOSED RESOLUTION: the group expresses its thanks to the hosts of this f2f, ETRI and the Mobile Web 2.0 Forum
<MartinJ> +1
<rob> +1
<jo> +1
<SeanP> +1
<DKA> RESOLUTION: the group expresses its thanks to the hosts of this f2f, ETRI and the Mobile Web 2.0 Forum
<Seungyun> suggestion : let's take a picture together !!