MASHUPS IN WEB 3.0

Arto Salminen

Department of Software Systems, Tampere University of Technology, P.O. Box 553, FI-33101, Tampere, Finland

Keywords: Mashups, Web 3.0.

Abstract: Web has developed into a platform where applications live as services. This is referred to as Web 2.0. The

next version, Web 3.0, refers to using the Web in a new way in new domains. In addition to realizing se-

mantic web, Web 3.0 includes other advantageous concepts too. This paper discusses about mashups in

Web 3.0 and describes how mashups will be an integral part of it. Moreover, we will point out some re-

markable technical solutions that enable new kind of mashups and speculate about the time when these

mashups can be fully implemented and realized.

1 INTRODUCTION

The way the Web is used has gone through signifi-

cant changes. The Web 1.0 was a simple platform

for browsing static documents that were connected

with hyperlinks. Next version, Web 2.0, introduced

user created content in a significant degree as well as

collaboration between users. Communities and so-

cial networks as well as different services that en-

abled sharing videos, images and texts became

popular. Mashups, web applications that integrate

resources — i.e. the content created by users and

enterprises — over the web were developed into a

new breed of software that is widely utilized in dif-

ferent domains including mobile and desktop plat-

forms. Currently, we are experiencing a paradigm

shift towards web-based software (Taivalsaari,

Mikkonen, Anttonen and Salminen, 2011), which

consists of resources that can be located anywhere in

the world and require no installation or manual up-

dates. This has had a great effect not only on how

software is used but also on development and de-

ployment of software (Taivalsaari et al., 2011). In

other words, what used to be a document browsing

platform has become a means of communication

with short messages and live audio, a place for mu-

sic and video entertainment as well as a platform for

full fledged applications such as text and spreadsheet

editors and games. However, the evolution of the

Web has not ceased, and as technical and other bar-

riers are overcome it turns into version 3.0.

There is no clear definition for Web 3.0. Some

use the term as a synonym for semantic web

(Hendler, 2009). Others, however, think that Web

3.0 refers to new ways to use the web, and using it in

new domains (Silva, Saleh, Rahman and El Saddik,

2008). Our perception of Web 3.0 is the latter.

Similarly to Silva et al. (2008), we define Web 3.0

as “tomorrows web” that is ubiquitous and

pervasive, which, in addition to semantic web,

utilizes also other concepts such as ambient

intelligence, smart interfaces and intelligent agents.

In addition to these, in our view the Web 3.0

includes the concept of mashware, software created

as a mashup as described in a technical report by

Taivalsaari (2009). Mashware enables personaliza-

tion and on-fly customization of web applications;

ideally automatically or by the users.

As already pointed out, mashups combine

content from more than one source into an integrated

experience. This ability to aggregate content

leverages the power of the Web to support

worldwide sharing, accessing and reusing resources

from different locations or in different contexts.

Consequently, mashups, as well as mashware,

demonstrate the capability of the Web to act as

global-scale distribution channel for arbitrary

distributed applications.

In this paper we discuss about mashups utilizing

Web 3.0 concepts and technologies and present our

view on what will happen in the future in the domain

of mashups. In addition we describe forthcoming

issues that are related to software mashupping using

new web technologies. To gain better understanding

about how Web 3.0 will benefit mashup developer,

we start with a brief background study about the

189

Salminen A..

MASHUPS IN WEB 3.0.

DOI: 10.5220/0003953901890194

In Proceedings of the 8th International Conference on Web Information Systems and Technologies (WEBIST-2012), pages 189-194

ISBN: 978-989-8565-08-2

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

concept.

This paper is structured as follows. In Section 2,

we present the concepts that are included and

discussed about under term “Web 3.0”. In Section 3,

we describe mashups that are enabled by Web 3.0

technologies and present examples of applications

available today. Finally, in Section 4, we conclude

the paper.

2 WEB 3.0 CONCEPT

We do not use term “Web 3.0” as a synonym for

semantic web. Our perception of Web 3.0 contains

not only the ideas introduced as “semantic web” but

a lot more. Term “semantic web” refers to extending

web documents so that information in them has clear

meaning understandable for machines (Berners-Lee,

Hendler and Lassila, 2001; Shadbolt, Berners-Lee

and Hall, 2006). It can be used to create interoper-

able websites that make information exchange ef-

fortless. For example, semantic web has been suc-

cessful in the domain of scientific publications (Das,

Goetz, Girard and Clark, 2009). In our view, seman-

tic web is an enabler for Web 3.0 applications but

there are other aspects as well.

In the Web 3.0, the connection with the rest of

the world via Internet is pervasive. It is available at

everywhere for everyone at anytime. This has been

achieved already as mobile terminals — that have

become inexpensive enough for everyone to pur-

chase — enable us to be online at all times, without

interruption. However, as new web-enabled devices

have been introduced, web applications can reach

new fields of everyday devices. Game consoles,

televisions and set top boxes already contain web

applications and similar capabilities are spreading

into cars, book reading devices and picture frames,

for instance. Furthermore, progress of ubiquitous

technologies (Weiser, 1991) makes everyday arte-

facts connected to the Internet, thus transforming

their data into resources for applications.

Another interesting development related to per-

vasive computing is ambient intelligence, which is a

slightly different concept. Ambient media or intelli-

gence refers to systems that are unobtrusive, context

aware, personalized, adaptive and anticipatory

(Zelkha, Epstein, Birrell and Dodsworth, 1998). It

emphasizes the ability of devices to communicate

and make decisions independently without user

interaction. Ambient media has been included in

Web 3.0 concept by Silva et al. (2008). This pro-

vides another view to web as a pervasive platform

used for making daily tasks easier.

Our view to Web 3.0 includes the concept of

mashware (Taivalsaari, 2009) as well. In mashware,

the idea of mashups has been expanded into soft-

ware. Mashware is created from software compo-

nents that are retrieved from all over the web and

composed together without static linking or pre-

processing. Realizing mashware in the full extend

would allow really large-scale collaboration between

developers as application components could be

shared and reused without restrictions. Similarly,

what has been achieved with current mashup tools

(see a detailed summary in (Taivalsaari, 2009)),

mashware applications could be developed possibly

by the end-users themselves. Mashware, naturally,

requires well-defined interfaces and we believe that

a lot can be learnt from the work already done in the

domain of semantic web.

Interestingly, Google’s CEO Eric Schmidt de-

scribed ideas similar to mashware when he gave his

definition for Web 3.0 at Seoul Digital Forum in

2007. Schmidt’s definition remarked that Web 3.0

applications will be pieced together, relatively small,

able to be run on any device, fast and customizable,

distributed via social networks and using data stored

in the cloud. Certain characteristics, i.e. “pieced

together”, “fast and customizable” and “using data

stored in the cloud”, are features of mashware as

well. However, according to Taivalsaari (2009),

mashware is not limited to small applications, and

distribution through social connections. Further-

more, mashware does not require universal-scale

cross-platform compatibility even though this can be

achieved with certain technologies.

Mashups rely on web services that are accessed

through APIs. Therefore, mashups benefit greatly

from recent developments called Open Web and

Open API. The term Open Web refers to develop-

ment promoted by Open Web Foundation, which is

founded by major web organizations and aimed at

promoting specifications that are royalty free and

compatible with open licenses. This makes using

different interfaces in co-operation easier. An inter-

face following Open Web recommendations and

available to be used by different parties is referred to

as an Open API.

2.1 Enabling Technologies

Web 3.0 is driven by technological challenges that

include implementing semantic web, expanding web

browser capabilities, having reliable high bandwidth

network connections, and linking physical world

with the Web.

Semantic Web. Giving semantics for data refers to

turning it into information that can be processed

independently by machines. In the domain of the

Web this means giving well-defined structure and

meaning for data stored currently within unstruc-

WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies

190

tured and meaningless web documents. Languages

and frameworks that can be used to achieve this

already exist. RDF (Resource Description Frame-

work), for instance, can be used with OWL (Web

Ontology Language) or XML (eXtensible Markup

Language) to describe the meaning of data within

structured web documents.

In addition, there are solutions that can be used

with the HTML, language currently used to describe

most documents available in the Web. Microformats

and microdata, the former described by the micro-

formats community (http://www.microformats.org)

and the latter introduced as part of HTML5 specifi-

cation (Hickson, 2011), can be used to annotate the

content with machine-readable labels. These solu-

tions are lightweight, simple and evolutionary, in

contrast to RDF used with OWL or XML, which can

be described as revolutionary, but more complex and

difficult to understand as well.

Web Browser Capabilities. Web browser perform-

ance has been increasing at unforeseen pace during

last few years. Performance in JavaScript execution

has skyrocketed thanks to development of powerful

engines during ongoing JavaScript engine race

started in 2008 between major browser vendors. In

addition to code execution performance, also docu-

ment rendering speed has evolved.

Two developments, HTML5 and WebGL, have

significantly improved capabilities of web browser

as a next generation application platform. HTML5

specification (originally named Web Applications

1.0) determines features that are typical for desktop

applications and makes them available in browsers

implemented as native features. These features in-

clude support for drag and drop, local data storage

(offline functionality), drawing surface available for

direct access by graphics hardware as well as video

and audio playing capabilities. Even though HTML5

specification is still at the draft stage many features

have already been implemented by browser vendors

and included in stable versions of web browsers.

WebGL is a technology that enables hardware

accelerated 3D graphics in a web browser. This is

remarkable because it allows visually attractive

games and other applications to be created without

browser plug-ins. WebGL can be used for 2D graph-

ics as well, and this enables developers to create

graphics in procedural style without web browsers

document object model (DOM), which is aimed at

presenting static documents. WebGL specification

has reached its first stable version number 1.0. Be-

cause of WebGL is a very low level interface, nu-

merous higher level frameworks and libraries have

been developed to make it easier and faster to create

WebGL applications.

Cross-origin Resource Sharing. One of the signifi-

cant problems in composing mashups has been web

browsers poor capability to communicate across

domains. Because of restricting security model of

web browsers, also known as the Same Origin Pol-

icy, to be able to communicate directly with differ-

ent web services one has need to use some cumber-

some workarounds, such as dynamically inserted

script elements and JSON with padding (JSONP).

These workarounds are typically prone to security

threats. Thanks to a recent specification called

CORS (Cross-origin resource sharing) by W3C,

mashup developers will be able to make cross-

domain requests in similar fashion as the same do-

main requests. However, CORS needs to be imple-

mented by service providers and it is currently sup-

ported by only a few services.

High Bandwidth Connections. High bandwidth

Internet connections are already available for house-

holds and the price of subscribing for a high band-

width connection in western countries has decreased.

However, numerous areas exist where the prices are

still relatively high. In addition to fixed Internet

connections, mobile connections with fixed rate data

plans have rapidly become common in western

countries. According to the OECD’s latest statistics

(December 2011) there are 309 million fixed and

590 million mobile broadband subscriptions. Num-

ber of wireless subscriptions rose 26 % during last

year whereas the number of fixed subscriptions was

increased only by 5.8 %.

In spite of the increase in mobile subscriptions,

mobile Internet is still often unreliable, and it suffers

from different kinds of defects. For instance, mobile

connections typically suffer from long latency times

and issues on handover situations, which can be

major shortcomings with certain types of applica-

tions. Luckily, new cellular broadband technologies

are providing solutions for these and other technical

issues.

Linking Physical World with the Web. Linking

the physical world with web capable applications

such as mashups can add a new dimension to the

user experience of a mashup. Location-awareness

has already been proven successful in mobile games,

in which it has been found to be a very attractive

feature (Korhonen, Saarenpää and Paavilainen,

2008). In addition to player location, physical arte-

facts have been incorporated into games as well

(Reid, 2008). Similar idea would benefit mashups as

well. However, linking not only location but also

physical items into the mashup requires some infra-

structure to be built. The infrastructure, however,

does not need to be high-end technology, but simple

2D bar codes containing small amount of necessary

information of the linkage may be enough.

MASHUPSINWEB3.0

191

3 MASHUPS IN WEB 3.0

Mashups will take advantage on developing web

technologies and concepts introduced along Web

3.0. In the following, we discuss about how mashups

benefit from Web 3.0 technologies today, in near

and in distant future.

3.1 Mashups Today

Mobile Mashups. Most successful mobile mashups

today are those used for communication with multi-

ple instant messaging services. These are available

for all mobile operating systems and used quite

widely. Reasons behind this success of instant mes-

saging mashups are, first the obvious need for appli-

cations of this kind caused by rivalling instant mes-

saging service providers, and second the fact that

mobile devices suit particularly well for communi-

cating with other people. Another type of a success-

ful mobile mashup is map-based mashups. Typi-

cally, these mashups show some additional location-

related information on top of a map, for instance

some mashups show other user locations on a map.

Pervasive Mashups. Mashups can be found in eve-

ryday devices as well. Some televisions, for in-

stance, include media front-end applications that

have capabilities to present videos from multiple

web services. Furthermore, mashup for presenting

weather information is another popular application

in this kind of device.

HTML5 Mashups. Utilizing HTML5 in mashups is

already possible. With HTML5 creating mashups is

more straightforward. It enables using video and

audio elements in mashups without the need for

plug-ins. With HTML5’s WebSockets it is possible

to create real-time collaborative mashups, as they

can be used for low latency bi-directional communi-

cation without the overhead caused by HTTP, which

is used before for applications of this kind.

Using W3C’s Geolocation API it is possible to

create location-aware mashups. This enables some

context-awareness such as location dependent

searches and filtering. However, location accuracy

gained with this technology is sometimes very poor.

For instance, when travelling with a train and con-

necting to the web with the internal WLAN of the

train, Geolocation usually points to one of the sta-

tions that may be on the other side of the country.

As HTML5 specification and other new APIs are

relatively new and still at draft stage, it is necessary

to create fallback mechanisms if it is desired to be

sure that mashups relying on these techniques work

for all users. This adds complexity of mashup archi-

tectures and implementations.

3.2 Mashups in Near Future

Some Web 3.0 technologies are already available

outside research laboratories in commercial devices.

We anticipate that mashups relying on such tech-

nologies will be available in the near future. In the

following, we discuss about mashups that are built

on these technologies.

WebGL Mashups. WebGL specification enables

hardware accelerated 3D graphics in a web browser.

This technology has not been yet utilized in mash-

ups. 3D graphics could be well fitted into social

mashups which would be used to communicate with

other users in virtual spaces. Another way to use

WebGL in mashups could be creating new visualiza-

tions for services, for instance video and picture

services. Furthermore, 3D enables new ways to

present complex data. For instance, a stock mashup

that tries to compress enormous amount of informa-

tion in one screen could benefit from using WebGL

graphics to make the information easier to interpret.

Mashups Accessing Arbitrary Mobile Peripher-

als. Mashups with access to mobile device peripher-

als will become common in the near future. One

domain of such mashups is augmented reality.

Augmented reality mashup could access the camera

of the device and add information related to the

context of the user on top of the real time video. The

information would be retrieved from web services,

for instance a mashup could use a visual search

service to provide photos and details of an unidenti-

fied plant or access Wikipedia to provide additional

information about sights located around the user.

Mashups of this kind would rely on accurate infor-

mation about user location as well as device orienta-

tion.

Ambient Mashups. With ambient media, there are

possibilities for even more advanced mashups. Us-

ing a mobile terminal to have effortless access to

content relevant to the user context, and combining

this context with the resources of the Web, can be

very valuable for the user, especially if the mashup

could work autonomously without explicit user

input. Furthermore, if the mashup allows user col-

laboration, even richer user experiences can be pro-

vided.

Mashware. Another development in the near future

will be the first fully mashware applications. Mash-

ware will be combined from multiple components

according to user’s needs. For instance, a mashware

video player could be constructed from components

that are added according to which services the user

WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies

192

would like to access, what types of videos are played

and what type of device the player is executed on. If,

for instance, YouTube (http://www.youtube.com/)

and Vimeo (http://vimeo.com/) services were used,

the application would include interfaces only for

those services and exclude interfacing components

for Qik (http://qik.com/) and Yahoo Video

(http://screen.yahoo.com/). The player could add the

rendering component according to the video types so

that opening a video in Ogg Theora, H.264 and

Adobe Flash format would result in adding a corre-

sponding component to the application. Further-

more, the user interface of the player would be

added as a component so that a living room media

centre would have different look and feel than a

mobile version of the same application. These appli-

cation components could be downloaded over the

web from different services following a common

interface specification. Naturally it is possible to

provide a closed repository of components as well.

High-bandwidth Mashups. As mobile network

connections get faster it will be possible to develop

mashup applications that combine resources that

require a lot of bandwidth. This can be for instance a

high quality real time mobile television combined

with related web content.

Mashups Embedded in Everyday Devices. In

addition to televisions, other devices will have

mashup applications as well. Mashups will be in-

cluded in devices such as vehicles and game con-

soles. For instance a car navigator with a web con-

nection could show live weather information and

web camera images in addition to driving instruc-

tions. Furthermore, games can use mashups to add

changing content into the game worlds. These

mashups embedded in everyday devices will turn

more advanced as well when the full potential of

these specialized platforms is combined with web

resources in imaginative ways.

Mashups in Games. In the future, mashups will be

implemented within games as well. Web content

from different sources can elaborate games in nu-

merous ways. Dynamic content has already proven

to be successful in games (Vanhatupa, 2009). For

instance, game character’s presence could be added

with images and messages from players account on

social networking services. Game worlds could

include content from the real world, for example

news casts could be played and real weather condi-

tions could be reflected. Furthermore, games that are

relying on user context, such as location-aware and

augmented reality games, could naturally include

some web content that would complement the ex-

perience.

Mashups Utilizing Microformats. Microformats,

i.e. small details of information embedded in exist-

ing HTML documents, will be used in mashups as

well. These lightweight semantics are already widely

available, and using them in mashups is only a mat-

ter of time. Microformats are currently used by

search engines to annotate search results. In mash-

ups microformats could be used as an input when

requesting relevant content. Another option where

microformats are useful is content filtering.

3.3 Mashups in Distant Future

Some Web 3.0 technologies are not yet deployed in

public, or they have some technical issues that need

to be overcome before using them more widely. We

anticipate that it will take some time for mashups

relying on such technologies to emerge. In the fol-

lowing, we discuss about mashups that will be avail-

able in distant future as technical barriers are broken.

3D Mashware. As 3D web technologies have just

reached somewhat stable stage, it will take a while

for the first 3D mashware applications to become

existence. Adding third dimension to graphics re-

quires a new approach to user interface development

and therefore existing user interface frameworks

cannot be used. Furthermore, as the WebGL is a

rather low level interface, creating mashware de-

mands much attention to details. However, it is

likely that existing higher level libraries and frame-

works build for WebGL can be used as a stepping

stone. Yet another technical restriction is the inabil-

ity to use canvas element to render content formatted

in HTML. This makes it hard to reuse the existing

visual elements of the web that rely on Document

Object Model (DOM).

Mashups Relying on Semantic Web. The full-

fledged semantic web is still around the corner.

Utilizing semantic web’s full power in mashware

will result in mashups that are capable of including

new services and content types automatically.

Mashups could benefit from semantic web’s knowl-

edge about relations of artefacts (and users) in the

web and use it to provide information that is most

relevant to the user.

Mashups Utilizing Large-scale Physical Infra-

structure. Mashups that take advantage of physical

infrastructure build in everyday devices will eventu-

ally be available, but because of required installa-

tions of hardware, this is not possible in the near

future. However, mashups utilizing such physical

connection can be really innovative and useful. For

instance, Near-Field Communication (NFC) tags

installed into a retail store could be used as an input

for mashup providing price comparison information

as well as other information about products such as

possible use scenarios.

MASHUPSINWEB3.0

193

4 CONCLUSIONS

In this paper, we argued that mashups and mashware

are at the core of what is known as Web 3.0, the next

version of the web. Some mashups that utilize Web

3.0 technologies are already available and in wide

use. In addition to describing some examples of

today’s Web 3.0 mashups we pointed out mashups

and mashware applications that will likely be avail-

able in the near future. Furthermore, we discussed

some more advantageous ideas that will take some-

what longer to appear.

From software engineering point of view mash-

ups set an interesting challenge. Mashup design has

been previously considered as ad hoc activity with

minimal relation to software engineering practices,

architecting or disciplined development (Hartmann,

Doorley and Klemmer, 2008). However, our re-

search has been focused on describing disciplined

guidelines for mashup development (Salminen,

Nyrhinen, Mikkonen and Taivalsaari, 2010) as well

as general architecture for mashups (Mikkonen and

Salminen, 2011).

As technical issues are being solved in accelerat-

ing pace, our position is that mashups and mashware

will play an important role in how we will be using

the Internet in Web 3.0 era.

REFERENCES

Berners-Lee, T., Hendler, J. and Lassila, O. (2001, May).

The Semantic Web. Scientific American, 29-37.

Das, S., Goetz, M., Girard, L. and Clark, T. (2009). Scien-

tific Publications on Web 3.0 (pp. 107-129). Proceed-

ings of 13th International Conference on Electronic

Publishing, Milan, Italy.

Hartmann, B., Doorley, S., Klemmer, S. (2008). Hacking,

Mashing, Gluing: Understanding Opportunistic De-

sign. IEEE Pervasive Computing. 7(3), 46-54.

Hendler, J. (2009). Web 3.0 Emerging. IEEE Computer,

January, 42(1), 111-113.

Hickson, I. (2011). HTML Microdata, Editors draft. Re-

trieved March 12, 2012, from

http://dev.w3.org/html5/md/

Korhonen, H., Saarenpää, H., and Paavilainen, J. (2008).

Pervasive Mobile Games - A New Mindset for Players

and Developers. Proceedings of the 2nd International

Conference on Fun and Games (pp. 21-32). Springer-

Verlag, Berlin,Heidelberg.

Mikkonen, T. & Salminen, A. (2011). Towards a Refer-

ence Architecture for Mashups (pp. 647-656). Pro-

ceedings of the Second International Workshop on

Variability, Adaptation and Dynamism in Software

Systems and Services. Berlin, Heidelberg, Springer-

Verlag.

Reid, J., (2008). Design for Coincidence: Incorporating

Real World Artifacts in Location-based Games. Pro-

ceedings of the 3rd international conference on Digi-

tal Interactive Media in Entertainment and Arts (pp.

18-25). ACM, New York, NY, USA.

Salminen, A., Nyrhinen, F., Mikkonen, T. and Taivalsaari,

A. (2010). Developing Client-side Mashups: Experi-

ences, Guidelines and the Road Ahead (pp. 161-168).

Proceedings of MindTrek 2010 Conference. ACM,

New York, NY, USA.

Silva, J., Saleh, A., Rahman, M. and El Saddik, A. (2008).

Web 3.0: A Vision for Bridging the Gap between Real

and Virtual. Proceeding of the 1st ACM international

workshop on Communicability design and evaluation

in cultural and ecological multimedia system (pp. 9-

14). ACM, New York, NY, USA.

Shadbolt, N., Berners-Lee, T., and Hall, W. (2006). The

Semantic Web Revisited. IEEE Intelligent Systems

21(3), 96-101.

Taivalsaari, A. (2009). Mashware: The Future of Web

Applications. Sun Microsystems Laboratories Techni-

cal Report TR-2009-181.

Taivalsaari, A., Mikkonen, T., Anttonen, M., Salminen, A.

(2011). The Death of Binary Software: End-User

Software Moves to the Web (pp. 17-23). Proceedings

of the 9th International Conference on Creating, Con-

necting and Collaborating through Computing, Kyoto,

Japan.

Vanhatupa, J-M. (2009). Generative Approach for Extend-

ing Computer Role-playing Games at Run-time (pp.

177-188). Proceedings of 11th Symposium on Pro-

gramming Languages and Software Tools and 7th

Nordic Workshop on Model Driven Software Engi-

neering, Tampere, Finland.

Weiser, M. (1991). The Computer for the 21st Century.

Scientific American, 265(3), 3-11.

Zelkha, E., Epstein, B., Birrell, S., Dodsworth, C. (1998).

From Devices to "Ambient Intelligence", Digital Liv-

ing Room Conference.

WEBIST2012-8thInternationalConferenceonWebInformationSystemsandTechnologies

194