Machine Learning Authors: Liz McMillan, Janakiram MSV, Roger Strukhoff, Yeshim Deniz, Pat Romanski

Related Topics: Machine Learning , Mobile IoT, Agile Computing

Machine Learning : Article

AJAX and Mobile Web 2.0: A Service Blueprint Combining del.icio.us and flickr

"So, is this a Mobile Web 2.0 service?"

Following my article about "Mobile Web 2.0," I wanted to find a blueprint/case study of a Mobile Web 2.0 service.

This follow-up article is a bit of a gedankenexperiment – but I have drawn on the excellent work being done by Dr Marc Davis and his team at the Garage cinema research at the University of California (Berkeley).

The service I am considering here is a ‘mobile’ version of a combination of del.icio.us and flickr

As you probably know, both del.icio.us and flickr are based on tags. However, note that in a mobile context, a ‘tag’ would have a different meaning to the term on the web. People do not like to enter a lot of information on a mobile device. Thus, a tag in a mobile sense, would be explicit information entered by the user (i.e. a ‘web’ tag) but more importantly information captured implicitly when the image was captured(for example the user’s location).

The service would enable you to
a) Search related images and get more information about a ‘camera phone image’ using historical analysis of metadata (including tags) from other users. This bit works like del.icio.us i.e. searching via tags BUT with a mobile element because the ‘tag’ could include many data elements that are unique to mobility(such as location)
b) ‘Share’ your images with others (either nominated friends or the general public similar to flickr but as a mobile service)

From a user perspective, the user would be able to
a) Capture an image using a camera phone alongwith metadata related to that image
b) Gain more information about that image from an analysis of historical data (either a missing element in the image or identifying the image itself)
c) Search related images based on tags
d) Share her image with others – either nominated friends or the general public

Let’s break down the components further. We need:
a) A mobile ‘tagging’ system at the point of image capture
b) A server side processing component which receives data elements from each user. It then adds insights based on historical analysis from data gleaned from other users.
c) An ability to deliver the results to the user(these could be a list of related images based on the tag or ‘missing’ information about the image)
d) A means to capture the user’s feedback to the results
e) A means to share images with others.


Tagging an Image
It’s not easy to ‘tag’ a mobile image at the point of capturing it. In fact, in a mobile context, implicit tagging is more important than explicit tagging(An explicit tag being a tag which the user enters themselves).

At the point the image is taken from a camera phone, there are three classes of data elements we could potentially capture

a) Temporal for example the time that the image was captured
b) Spatial – The GPS location or cell id
c) Personal/Social - Username (and other personal profile information which the user chooses to share), presence, any tags that the user has entered, other people in the vicinity(perhaps identified by Bluetooth), other places of interest recently visited etc

The client component captures all the data elements and sends them to the server. It also displays the results from the server. (The garage cinema research uses a system called Mobile Media Metadata (MMM) which performs this function.)

Server Side Processing
The server aggregates metadata from all users and applies some algorithms to the data. The data could also be ‘enriched’ by data sources such as land registry data, mapping data etc. It then sends the results back to the user who can browse the results.

Finding ‘Missing Elements’ of Your Image
In many cases, it’s not easy to identify elements of the image(or in some cases, the image itself).

Consider the three images of Big Ben shown below. The third image is not very clear. It also includes two neighbouring ‘points of interest’ i.e. the River Thames and the Houses of Parliament .


Based on Metadata from other users, the ‘River Thames’ and ‘Houses of Parliament’ could be identified to the person capturing the third image. This is because - potentially other users would have captured separate images of the three points of interest and tagged them.

Thus, if the third user wanted to know ‘the river in the image’ or the ‘building in the image’ - they would be presented with a likely set of related points of interest which could include the river Thames and the house of commons. (Laughably trivial – I know – but it illustrates the point!)

Sharing Your Images
This is the ‘flickr’ component. However, ‘sharing’ in a mobile context, also includes location. This is very similar to the ‘air graffiti’ system I described in my previous article.

To recap, from my previous article, the air graffiti system is the ability to ‘pin’ digital ‘post=it notes’ at any physical point. Suppose you were at a holiday destination and you took a picture or a video of that location. You then ‘posted’ that note digitally with your comments and made it accessible to your ‘friends’. Many years later, one of your friends happened to come to that same place and as she walked to the venue, a message would pop up on her device with your notes, picture and comments.
Like flickr, ‘friends’ may be members of the general public with similar interests (i.e. like flickr ) or a closed group.

So, Is This a Mobile Web 2.0 Service?
Let’s consider some of the principles here (for a detailed explanation, please read my article Mobile Web 2.0: Web 2.0 and its impact on the mobility and digital convergence (Part one of three)

• It’s a service and not packaged software.
• It’s scaleable.
• It utilizes the ‘long tail’ i.e. input from many users as opposed to a core few.
• The service is managing a data source (it’s not just software).
• The data source gets richer as more people use the service.
• Users are trusted as ‘co-developers’ i.e. users contribute significantly.
• The service clearly harnesses ‘collective intelligence’ and by definition is ‘above the level of a single device’.
• Implicit user defaults are captured.
• Data is ‘some rights reserved’ – people are sharing their images with others.

The two aspects not covered above are:
• A rich user experience and
• A lightweight programming model

These are implementation issues and could easily be included. So, IMHO - indeed this is an example of a mobile web 2.0 service!

a) The example may sound trivial since Big Ben is a well-known location – but the same principle could apply to images of other lesser-known sites.
b) Of course, other types of data could be captured from the mobile phone for example video and sound.
c) There are no major technical bottlenecks as far as I can see(there are some commercial/privacy issues though).
d) From the above, you can see that Moblogging , in itself, is not an example of a Web 2.0 service
e) There are a whole raft of problems when it comes to the network effect and mobility. I have not discussed these here.

Garage cinema research

  • "Mobile Media Metadata for Mobile Imaging"
    Marc Davis, University of California at Berkeley and Risto Sarvas, Helsinki Institute for Information Technology
  • "From Context to Content: Leveraging Context to Infer Media Metadata"
    Marc Davis, Simon King, Nathan Good, and Risto Sarvas
    University of California at Berkeley

    Photo Credits:
    Image One
    Image Two
    Image three

USS Voyager Blueprint image : and http://www.startrek.com

For a daily dose of Ajit Jaokar, readers might like to visit


More Stories By Ajit Jaokar

Ajit Jaokar is the author of the book 'Mobile Web 2.0' and is also a member of the Web2.0 workgroup. Currently, he plays an advisory role to a number of mobile start-ups in the UK and Scandinavia. He also works with the government and trade missions of a number of countries including South Korea and Ireland. He is a regular speaker at SYS-CON events including AJAXWorld Conference & Expo.

Comments (6)

Share your thoughts on this story.

Add your comment
You must be signed in to add a comment. Sign-in | Register

In accordance with our Comment Policy, we encourage comments that are on topic, relevant and to-the-point. We will remove comments that include profanity, personal attacks, racial slurs, threats of violence, or other inappropriate material that violates our Terms and Conditions, and will block users who make repeated violations. We ask all readers to expect diversity of opinion and to treat one another with dignity and respect.

CloudEXPO Stories
Cloud-Native thinking and Serverless Computing are now the norm in financial services, manufacturing, telco, healthcare, transportation, energy, media, entertainment, retail and other consumer industries, as well as the public sector. The widespread success of cloud computing is driving the DevOps revolution in enterprise IT. Now as never before, development teams must communicate and collaborate in a dynamic, 24/7/365 environment. There is no time to wait for long development cycles that produce software that is obsolete at launch. DevOps may be disruptive, but it is essential. DevOpsSUMMIT at CloudEXPO expands the DevOps community, enable a wide sharing of knowledge, and educate delegates and technology providers alike.
In a recent survey, Sumo Logic surveyed 1,500 customers who employ cloud services such as Amazon Web Services (AWS), Microsoft Azure, and Google Cloud Platform (GCP). According to the survey, a quarter of the respondents have already deployed Docker containers and nearly as many (23 percent) are employing the AWS Lambda serverless computing framework. It's clear: serverless is here to stay. The adoption does come with some needed changes, within both application development and operations. That means serverless is also changing the way we leverage public clouds. Truth-be-told, many enterprise IT shops were so happy to get out of the management of physical servers within a data center that many limitations of the existing public IaaS clouds were forgiven. However, now that we've lived a few years with public IaaS clouds, developers and CloudOps pros are giving a huge thumbs down to the...
Kubernetes is an open source system for automating deployment, scaling, and management of containerized applications. Kubernetes was originally built by Google, leveraging years of experience with managing container workloads, and is now a Cloud Native Compute Foundation (CNCF) project. Kubernetes has been widely adopted by the community, supported on all major public and private cloud providers, and is gaining rapid adoption in enterprises. However, Kubernetes may seem intimidating and complex to learn. This is because Kubernetes is more of a toolset than a ready solution. Hence it’s essential to know when and how to apply the appropriate Kubernetes constructs.
To enable their developers, ensure SLAs and increase IT efficiency, Enterprise IT is moving towards a unified, centralized approach for managing their hybrid infrastructure. As if the journey to the cloud - private and public - was not difficult enough, the need to support modern technologies such as Containers and Serverless applications further complicates matters. This talk covers key patterns and lessons learned from large organizations for architecting your hybrid cloud in a way that: Supports self-service, "public cloud" experience for your developers that's consistent across any infrastructure. Gives Ops peace of mind with automated management of DR, scaling, provisioning, deployments, etc.
xMatters helps enterprises prevent, manage and resolve IT incidents. xMatters industry-leading Service Availability platform prevents IT issues from becoming big business problems. Large enterprises, small workgroups, and innovative DevOps teams rely on its proactive issue resolution service to maintain operational visibility and control in today's highly-fragmented IT environment. xMatters provides toolchain integrations to hundreds of IT management, security and DevOps tools. xMatters is the primary Service Availability platform trusted by leading global companies and innovative challengers including BMC Software, Credit Suisse, Danske Bank, DXC technology, Experian, Intuit, NVIDIA, Sony Network Interactive, ViaSat and Vodafone. xMatters is headquartered in San Ramon, California and has offices worldwide.