Paper Title: The Two Cultures: Mashing up Web 2.0 and the Semantic Web
Authors: Anupriya Ankolekar, Markus Krotzsch, Thanh Tran, Denny Vrandecic
Three Critical Questions
Monday
Group 1:
Member Name:Chiranjeevi Ashok Puvvula
1. What should be done to eliminate the logical complexity associated with “Expressive Ontology’s” for answering complex query’s?
2. Usually in mash up’s different web services can be grouped to obtain a result, many incompatibilities like format of the data, type of data, data interoperability will come into picture. How can these incompatibilities be addressed for a semantic-web2.0 mash up?
3. Do we need full automation or semi automation while mapping ontologies? Full automation is needed when there is no static data, and semi automation is needed when there is some static data. Both have advantages and disadvantage. Which would be preferable?
4. Can data integration, data interoperability and data exchange be done with out a standard language? Existing languages like SPARQL, OWL support this with some additional overheads. Do we need a new language to support this?
Group 2:
Member Name: Srikanth Kodali
1. According to the author, the semantic and web technologies are to be integrated to exchange the data between them. How exactly they can integrate with each other? How they can exchange the data between the applications?
2. In author’s scenario, the blogger have to tag with the Meta data of the data source before beginning. If there are different kinds of data which is tagged with the same Meta data, how the heterogeneity is eliminated?
3. In interoperable semantic applications, the data can be transformed between the applications. How exactly the security is established between the applications?
4. Suppose if two different data sources giving contradictory information to a semantic application? How the semantic application choose among them and how the data from that source is reliable than other?
Group 3:
Member Name: Sunil Kakaraparthi
1. The author in the paper suggests that the in the incorporation of the semantic web creation by combining the existing applications becomes the semantic web data sources but how would the incorporation done that easily as it involves several issues such as complexity, security, readability, redundancy etc.,?
2. RDF exporter is used by the author for creating the blog but has problems such as using the URI’s which confuses and it is serializes as the XML which has problems for the XML and RDF. Is there an alternative to the RDF exporter which overcomes these problems?
3. The application referred by the author is a combination of number of web technologies and human contributions. What are the Web technologies does the author refer to?
4. The author has mentioned that integration of various applications with the new one (semantic web) creates various complexities as the web is already expanded a lot that it is almost impossible to restructure. So, how can the web be modified such that semantic web replace it without any complexity and efficiently?
Group 4:
Member Name:Ramya Devabhaktuni , Prashant Sukari
1.The paper stated that the “integration and mapping” involves sharing of data between different domains. During this exchange, there may be “overlap” of data or may contain heterogeneous data where overlapping is not possible. How these differences are “aligned” or “resigned”?
2.It is mentioned in the paper that “semantic data” cannot be controlled using the “HTML tags and the “digital signature”. How the key factors like confidentiality and trust be achieved by using the “Semantic data”?
3.The author mentioned that “the semantic data” are defined in terms of different data formats. But achieving these data formats involves “high programming effort” and involves manual efforts like “publishing” and “updating”. How these issues are resolved in case of “complex knowledge”?
Group 5:
Member Name: Lokesh Reddy Gokul
• The paper presents views on traversal towards Web2.0 from Semantic Web vision rather than a traversal from Web2.0 towards Semantic Web, which may not be a good sign of progress in Web evolution. What are the reasons for proposing such an approach? And is this approach really fruitful enough?
• The paper proposes the vision of deploying Semantic Web as equal to an World Wide Web and hence an evolution of semantic web deployment from an intranet level to more wide range so that it is more user friendly and user knowledge in ontology or other such technical implementation is not required. What would it take for such a deployment? And what might be the immediate effects expected?
• In proposing the Next steps which may be required for efficient Web mash-ups; authors say that it is necessary to incorporate semantics into the applications to enhance usability of ontologies. But if none of ontologies match user’s needs, this may reduce the semantic annotations. This may also increase redundancy of ontologies when user may guess that there is no ontologies that match his purpose because of lack of annotations available, he may go on to make a new one. So how are such flaws countered?
Group 6:
1) The web 2.0 provides very good user experience which is dominating factor in the present world. While the use of semantic web puts a lot of time in creation and does not have very good user experience. How these two technologies can be used combined, to get maximum advantage? 2 )The author in this paper explained about the creation of wiki blogs etc with the help of semantic web, but in general the customers who are using the blog spent very less amount of money for using it. The use of semantic web for blogs or wiki’s makes the organization spending a lot of money. How semantic web is going to solve this problem?
3) Semantic web uses the universal resource indication to point out the objects in the web while the web2.0 does not use such one. How can both of them work together?
Group 7:
Member Name: Priyanka Koneru
1. “ It is mentioned in the paper that using ‘Smoov’ it is possible to guess the location of the user using her IP address. But it is not always correct that ‘Smoov’ gives the correct IP address. But it is possible to hack the passwords, confidential files from the system in which the user is currently accessing the ‘Smoov’. So the confidentiality of the user is not maintained. Are there any alternatives that it is possible to maintain user and data confidentiality?”
2. “ The main problem of today’s web is controlling and gathering of semantic data which becomes more difficult, private and proprietary information can hardly be restricted. The author didn’t suggest any solution to this problem”.
3. “ There are many questions which aren’t answered in the paper like distribution, gathering and combining the gathered data, reusing the existing data. What are the sources of semantic data and can this data be put into practical use etc”.
4. “ For ‘integration and mapping’, data is to be sharing between different domains. How are the confidentiality issues addressed in this scenario?
5. “Inorder to implement web applications dynamically, data sources use caching. But how can chaching be implemented by the semantic web as it doesn’t have controlled data sources?”
Group 8:
Member Name:Durga Maheswari Muppalla
1. It is stated that the ontologies exchange process require the common understanding and common patterns of the data. As the requirements changes the modeling and structure also changes so how the “mapping and the integration” can be done when there are complex models which needs to be modified continuously. And moreover the ontologies often require the “extensions” and the “modifications” and there mapping is a critical process.
2. As the “data sources” are interconnected and are distributed over the web when the interconnections break or when they are lost the “cache” must be “reinvestigated” and “re implemented” based on the dynamic content in the web. Isn’t the investigation of the cache a time consuming and what if when there are interconnection between the complex models and when they acquire from the dynamic content there will be difficult because as the requirement changes the content also changes. how are these issues addressed is not discussed in the paper.
3. It is stated that the tagging using the HTML is a “potentially useful” and it is well established. How the reliability concepts and the trust are achieved when using this type of technology.
4. How the “complex content” of the ontologies can be encoded without any additional “expressivity” and moreover ontologies have many challenges how these challenges can be overcome.
Wednesday
Group 1:
Member Name: pelluri,voruganti,lattupalli
To obtain reusability author suggested using mash up concepts. But consider a situation where there exists one or more services and some message passing should be done between them. Here, both of the services may use their own terminology then message passing may be difficult. Also on using mash ups we are depending on external data source. What if that external data source is down for a while?
Author said that semantic web technologies make use of common data formats that are universal across the domains. On doing so where is the freedom for develop his application in his own way? After all, one of the promises of semantic web technology is that it should interoperate between different applications which are developed on different formats.
While dealing with interoperability issue author assumed that accepted vocabulary is available. According to me, the major challenge the semantic web technology facing is the diverse ways in representing data. How can author assume that accepted vocabulary is available? Even in small domains different developers may use different terminologies. So having an accepted vocabulary is impractical.
Group 2:
Member Name: Addagalla Satyanaga Manoj Kumar; Bobbili, Shrinath; Gopinath, Sreejith
1) The authors argue the power and advantages of semantic web and mashup applications by citing the example of a blog. The question is whether the advantages could be leveraged in business and operation-critical scenarios, and if so, whether it would be as easy and effective as described in the example.
2) From an infrastructure point of view, the arguments put forth by the authors suggest a step backward from the current paradigms of greater machine autonomy and decision-making involvement(the example cited by the authors involves significant human input and web service involvement). The critical question is whether this paradigm shift is healthy for the future of semantic web and machine learning; are we not going backward from where we are already?
3) As regards to mapping and integrating applications, this approach assumes a single agreed-upon ontology or set of ontologies. The questions is whether the mashup can scale up if programmers of some of the participating applications introduce new features and there is a need to deviate from the agreed-upon ontology.
4) The authors have suggested a new paradigm, a new approach to mashing up application. The question is whether they have actually implemented such a system and if so, do the performance and effectiveness experiences vindicate their stand?
5) The authors advocate the development of domain-specific mashups and semantic wen applications – is this not against the very doctrine of semantic web-technology; are we trying to integrate the whole current web or keep them divided into specialized domains that do not or have no scope for interaction?
Group 3:
Member Name: Chi Zhang
1. The author intends to talk about the benefit and drawbacks that the semantic web will have on the community, and he switches his view first to the combination of Web 2.0 and semantic web technology and then he turns to stress the great importance of semantic web. Other than giving a scenario of the movie weblog and difficulty of encode ontologies, he surely does not talk much on this subject. Is the user-experience not important? Or the statistical result of the community is hard to come by?
2. As we all know Web 2.0 is a successful business and technological model. The income is much higher than the cost of establishing certain websites. However, for the semantic web, the make-money mode is not really clear. And the author tend to stress more on the technological aspect instead of commercial. Yes they can compliment each other, but who will realize the system if the only benefit is to the customers?
3. The author seems to forget one important point that is real-time. Since the world is becoming smaller, tasks are more critical that tons of data should be exchanged in real time. For example if we update our ontology today, how soon it will be noticed by end users? Why this is not pointed out by the author? Since it is a very practical issue.
Group 4:
Member Name:Shaiv, karuna priya rameshwaram, anusha vunnam
1) The paper moves on talking about the scenario of web by merging the two webs by explaining the things related to cinema blogging, of course it might be appropriate in this case. Would we get the same outcome for every real world application?
2) While reusing, we come across various types of data and we have to implement caching for these data if necessary. What would be cost of implementing caching?
3) Even though the usage of lightweight ontologies are flexible,what additional features are required to be employed by the complex ontology languages in order to address the necessities of interfaces in a software?
Group 5:
Member Name: Rahul Mootha, Rahul Reddy
• A challenge in decentralized systems is tracing the origin of the data and trusting the data that is available. How far can the data be trusted using digital signatures since it may not be possible for various heterogeneous data to have digital signatures and moreover they cannot be traced ?
• The author says that there might be differences in ontologies and when shared data needs to be integrated the differences have to be aligned and mapped . How is this integration and aligning the differences performed to achieve the goal of mapping and sharing data?
• Scalibility is an important issue in Semantic Web. Dealing with dynamic websites where the data is to be updated often and large amounts of data which is distributed needs to be aggregated. How is this problem dealt?
Group 6:
Member Name:
Group 7:
Member Name:
Group 8:
Member Name:Brugu Kumar Bhargava
1) Author says that we can use technologies like XML for developing the side bar of Chrissie’s blog then why do we need semantic technology for developing the side bar?
2) Authors proposes that there is a need for exchange of the semantic data but as the ontology’s grow large and have many filtering methods it becomes really complex for exchange of semantic data. Then how can the exchanges takes place in a large ontology’s?
3) Author while explaining about the “Expressive Ontology’s” says that it’s really difficult for encoding the complex knowledge and the usage of this ontology’s is really tough. Then how can we efficiently express the ontology?
Group 9:
Member Name: Satish Bhat, Holly Vo
1. Since user can trace provider information (for data judgment or in case user is anonymous, creator preference is used in web application by default as in the example of movie blog), how to protect context creator across nation boundaries? (Ex: some countries imprison writer on some semantic context which is commonly valid in the world.)
2. What party should take care of service plug-ins over various ontologies? (Ex: for novel movie vocabulary.)
3. How to normalize data qualities from distributed sources? (Ex: how to normalize movie ratings from different personal blogs?)
Group 10:
Member Name: Sunae Shin, Hyungbae Park
1) In the scenario, they talked about personalization of Web sites. It tells someone’s blog can display the same information but different values for readers based on the location of the readers. However, this can be used to track the readers of the blog by installing some other plug-in into the blog. They didn’t mention anything about security issues for this problem.
2) When I read this paper, I think that their Web 2.0 scenario using semantic technologies is not realistic since we don’t have enough infrastructures for this scenario. However, they show the basic Semantic Web infrastructure that the scenario requires. This helps me understand why they wrote such an unrealistic scenario.
3) Actually, what they are showing from the scenario can be implemented by using XML. Not many works related semantic technologies are not involved yet in their scenario. However, they didn’t mention what are the better aspects of using semantic technologies for their scenario.
4) They introduced three things for the requirements of the Semantic Web infrastructure; creation, exchange, and reuse. I think they should add a “consistency” into the requirements because the consistency is important when we use the shared data. If we cannot assure that the data is always consistent, then the using those data is not productive.
5) Their ideas are not for the individual’s benefits. Many people can benefit from their ideas. However, before their idea is published to the public, the merits of Semantic Web technologies need to be proven through real-life application scenarios. Their ideas are really creative and seem that they can bring many benefits to the public. However, only ideas are not enough. It’s a pity that they didn’t mention how they are going to prove their ideas are productive.
