Vodafone RnD has developed a new application that permits users to contribute to articles in wikis. You can view and edit articles, and also upload photos from your device. Try this application on Betavine and contribute to this open source project over at the Forge.
The Betavine team will buying everyone dinner on Saturday night, at the 3rd annual BarCampBrigton on the 6th and 7th of September.
Find out more
Vodafone R&D and “Instrumentación y Componentes (Inycom)”, an Aragon-based Technological company have just released a SIP Stack for mobile devices which performs for you the 3GPP procedures for registering, instant messaging and other services, read more here. It exposes the IMS functionality by a high level API, hiding the technology complexity and enabling you to develop 3rd party IMS applications in an easy way. You can either view the project page on Betavine or the actual developer project over at the Forge.
We've just released version 2.0 of betavine.net. In December we decided to re-write the whole platform using standard open source components. This gives us greater flexibility, allowing us to develop new site features for you to use.
Find out more
The Vodafone UK Student Campus Life competition has now come to a close. The 3 winners have been announced.
Find out more
The Nokia Student WidSets competition has now come to a close. The Widsets team have announced the three lucky winners with more details on our site.
Find out more
We are pleased to announce the first of a series steps we are taking to help promote Betavine apps to a wider audience. As of lunchtime today, there is now a link to betavine.mobi in the Applications area on Vodafone Live! in Spain. This is a fantastic opportunity that opens up Betavine to over 16 million mobile users! If this proves popular we hope to link with other countries in the near future.
The homepage displayed to users accessing the site from Vodafone Live! contains a set of 'Featured Applications'. If you would like to showcase your application via this channel, please email webmaster@betavine.net.
Vodafone Global are launching a suite of applications that work on a range of phones. Due to the way apps are hosted, and our inbuilt feedback and deployment system they have chosen Betavine to partner with.
Why not have a look at some of the apps!
Find out more
For previous news visit:
News on the Vine Archive
|
jalberto Cellectivity |
18/9/07 5:36 PM
A different route to Content Adaptation
The real problem with Vodafone's strategy is its unrealistically simplistic approach (i.e. disguise all clients no matter what, unless the site is white listed in advance).
A much more clever approach would have been to pass the request as is and then look at the response to make a decision on what to do next. There are 4 possible outcomes: 1) The provider responds with WML or XHTML-MP, supported by the device (the proxy knows what is supported and what is not); in this case pass it through with no change. 2) The provider responds with HTML or some other content type not supported by the device: apply content adaptation. 3) The provider returns a 406 error or whatever informing it does not support the client user-agent. In this case try again the request disguising the client, apply content adaptation, and remember to keep adapting for the entire user session with the site. 4) The provider returns a "200 OK" but responds with an empty page. In this case do the same as for (3) above. This would allow providers with mobile content to work fine and also allow non mobile sites to be rendered properly in a transparent manner. Is this so difficult to do? No is not. The engine already needs to cache the response in order to transform it. So the only thing here is to look at the cached response, see if it looks OK for the device and if so abort content adaptation and return original response as is. If it is not OK, then either proceed with adaptation or re-issue the request. Jose Alberto |
|
Scott |
21/9/07 4:39 PM
RE: A different route to Content Adaptation
as a reply to jalberto
Hi Jose
This sounds like a good, simple technical solution but I don't thing Vodafone will go for it. Looking back through the previous messages from Brian the following stands out. "The desktop user agent string is sent primarily to get desktop internet content vs. mobile pages, not to get desktop content vs. nothing at all." My take on this is that Vodafone, would prefer the default situation to be that their customers receive the desktop PC version of a site even if a mobile or device adapted version of the site exists. Until they move away from the concept that desktop is best then I don't see the user-agent issue being resolved. Regards Scott |
|
mrbgturner Vodafone |
25/9/07 7:44 AM
RE: A different route to Content Adaptation
as a reply to jalberto
Hi Jose and Scott
We have had a look at Jose's approach. Thanks for the time and effort in preparing it. Here is our response which you are free to come back on. "Unfortunately this approach doesn’t resolve the problem that some desktop sites would become entirely inaccessible to the WTE. When condition 1 below holds, only the WAP/XHTML page can be received, and the only way to get to the HTML page is if the content provider happened to include a link providing access to that page. Conditions 3 and 4 unfortunately don’t help that problem, and they appear to occur very rarely if ever." BR Brian |
|
nigel Admob |
25/9/07 4:22 PM
RE: A different route to Content Adaptation
as a reply to mrbgturner
mrbgturner: "Unfortunately this approach doesn’t resolve the problem that some desktop sites would become entirely inaccessible to the WTE. When condition 1 below holds, only the WAP/XHTML page can be received, and the only way to get to the HTML page is if the content provider happened to include a link providing access to that page. Conditions 3 and 4 unfortunately don’t help that problem, and they appear to occur very rarely if ever." It's patently clear now that Vodafone does not care about all the existing mobile sites afterall. Please tell me that's not true. You have to realize you are destroying the mobile ecosystem. Unless, of course, the mobile site is in your white list. How much more do we have to say to convince you that you are not helping yourself? |
|
jalberto Cellectivity |
29/9/07 10:06 AM
RE: A different route to Content Adaptation
as a reply to mrbgturner
mrbgturner: "Unfortunately this approach doesn’t resolve the problem that some desktop sites would become entirely inaccessible to the WTE. When condition 1 below holds, only the WAP/XHTML page can be received, and the only way to get to the HTML page is if the content provider happened to include a link providing access to that page. Conditions 3 and 4 unfortunately don’t help that problem, and they appear to occur very rarely if ever." BR Brian I think it is incredible that you guys think you are best at defining what is in the best interest of the relationship between a user and a content provider. If I as a content provider decide to show specific content to a user, who are you to try to bypass my decision? It is up to me the content provider to decide what content I give to the users visiting my site, not Vodafone's. Imagine what would happen if you were able to do something similar with lets say the IP addresses of your users. Just for the sake of argument, lets say you decide to route all your USA users through your UK gateway. This is an entirely feasible proposition, by the way. That would mean as a content provider I will not able to offer country specific content to those users; even worst I would be offering the wrong content. With all the legal and financial implications that may carry. Just because you control the access to the users does not mean that you entitled to do whatever you want. At this point, there are mobile sites loosing revenue because of what you have done. Ringtones are not being sold, mobile games are not being bought. Does that bother you guys at all? Jose Alberto Fernandez |
|
jalberto Cellectivity |
29/9/07 3:48 PM
RE: A different route to Content Adaptation
as a reply to jalberto
Trying to continue with constructive criticism, here is an even simpler approach to the white-listing that you require at present.
Ideally I would propose you take this to W3C as a proposed standard. Basically the idea is to treat trans-coding gateways the same way as we do web-crawlers today; allowing for the site in question to indicate whether they are welcome or not and if they do which parts of the site they are allowed to act on. A file similar to robots.txt lets call it no-transform.txt and using the same syntax as robots.txt will be put on the root directory of the site. Any path marked as "Disallowed:" means that the gateway must pass requests between the client device and the site without any transformation on any direction. The gateway must send the request as is with the original user-agent and the gateway will pass the response also as is, no matter what. Let me make things clear, we are not just talking about WEB/WAP traffic here, in today's world there are mobile online games and business applications using J2ME that need to be able to connect and to comunicate using their own protocols. Vodafone nor any other operator can assume that the only traffic on the mobile web is just human readable HTML. So how about that? Is this fear enough for you to implement? Jose Alberto Fernandez |
|
mrbgturner Vodafone |
1/10/07 10:45 AM
RE: A different route to Content Adaptation
as a reply to jalberto
Hi
This approach is similar to the function we have implemented already with the use of <No transform>. * The instructions for using No-Transform are easy. To tell the WTE to avoid modifying an item of content in any way, add the following header to the outgoing HTTP response headers: * Cache-Control: No-Transform I'll forward your input to our W3C representative. Many thanks Brian |
|
luca Admob.com |
2/10/07 8:16 AM
RE: A different route to Content Adaptation
as a reply to mrbgturner
Please do not refer to W3C and standards in your postings. You are blatantly ignoring standards and breaking the foundation of HTTP for some sick business reasons, so your referring to standards is irritating to say the least. Luca |
|
jalberto Cellectivity |
4/10/07 11:11 AM
RE: A different route to Content Adaptation
as a reply to mrbgturner
mrbgturner: This approach is similar to the function we have implemented already with the use of <No transform>. * The instructions for using No-Transform are easy. To tell the WTE to avoid modifying an item of content in any way, add the following header to the outgoing HTTP response headers: * Cache-Control: No-Transform Brian, this is not the same. No-Transform is a response header. By the time one specifies this you have already sent the wrong request with the wrong User-Agent. Even if you abort the original request and re-issue an unaltered one, a transactional site may have executed all kinds of code against the wrong request. We are not talking static pages here, we are talking transactional sites with business logic involved in the processing of the requests. The approach would have made sense, if you would have sent the unaltered User-Agent on the original request. But you are not, do you? Please understand the size of the actual problem you are causing, by changing the request sent by the device. Jose Alberto |