Multiplatform Reactive Crowdsourcing based on Social Networks - WWW2013

In this post I want to report on our paper on Reactive Crowdsourcing presented at the WWW 2013 conference in Rio de Janeiro, Brasil.

Here is a quick summary of motivation and idea, together with some relevant materials:

• Presentation
• Demonstration Video
• Full Paper

Need for control

We believe that an essential aspect for building effective crowdsourcing computations is the ability of "controlling the crowd", i.e. of dynamically adapting the behaviour of the crowdsourcing systems as response to the quantity and quality of completed tasks or to the availability and reliability of performers.
This new paper focuses on a machinery and methodology for deploying configurable, cross-platform, and adaptive crowdsourcing campaigns through a model-driven approach.

Control through declarative active rules

In the paper we present an approach to crowdsourcing which provides powerful and flexible crowd controls. We model each crowdsourcing application as composition of elementary task types and we progressively transform these high level specifications into the features of a reactive execution environment that supports task planning, assignment and completion as well as performer monitoring and exclusion. Controls are specified as declarative, active rules on top of data structures which are derived from the model of the application; rules can be added, dropped or modified, thus  guaranteeing maximal exibility with limited effort. The paper applies modeling practices (as also explained in our book on model-driven software engineering).

Here is the presentation thatAlessandro Bozzon gave at WWW 2013:

Reactive crowdsourcing presentation on slideshare.

Prototype and experiments

We have a prototype platform that implements the proposed framework.  We have done extensive experiments with it. Our experimentations with different rule sets demonstrate how simple changes to the rules can substantially affect time, effort and quality involved in crowdsourcing activities.

Here is a short video demonstrating our approach through the current prototype (mainly centered on the crowdsourcing campaign configuration phase):

Paper and related activities

The paper can be downloaded for free from the WWW site through this link:
http://www2013.org/proceedings/p153.pdf


The paper is a follow-up of our WWW2012 paper on Crowdsearcher, which focused on exploiting social networks and crowdsourcing platforms for improving search.
The paper nicely combines with another recent contribution of ours, presented at EDBT 2013, on finding the right crowd of experts on social networks for addressing a specific problem.


To keep updated on my activities you can subscribe to the RSS feed of my blog or follow my twitter account (@MarcoBrambi).

OMG IFML webinars in English, Spanish and Italian

For your information, we will give a webinar dedicated to the new OMG standard IFML (a modeling language for defining the User Interaction of software applications, born as evolution and extension of WebML) and its practical use within model-driven approaches that go down to the generation of industrial applications upon web, mobile and proprietary platforms, as supported by WebRatio.  The webinar will also introduce the new open source design tool for IFML developed by WebRatio.

The Webinar will be given in three languages: English, Spanish and Italian.
The schedule of the three versions of the webinar is as follows:

Friday May 31, 2013
- at 15 CET (in English)
- at 16 COT (in Spanish)

Friday June 7, 2013 
- at 15 CET (in Italian)

You can register to the webinar here:

To keep updated on my activities you can subscribe to the RSS feed of my blog or follow my twitter account (@MarcoBrambi).

Gems on MDE and UML by Jean Bézivin (and others)

I'd like to share and write down in a more permanent way some thoughts and messages I recollected from yesterday's twitter interactions with Jean Bézivin on Model Driven Engineering (MDE) and UML.
I'm sure future generations will appreciate my writeup of these small, hidden gems that otherwise would have gone lost in the Twitter flurry :) .

All this was started by my post on Quora basically saying that Model-driven Engineering is more than UML.Jean's (serial) response is reported here:

1. Sometimes MDE is the opposite of UML.
2. UML favours visual modeling; MDE treats visual and textual modeling on par.
3. UML favours General Purpose Modeling Languages. MDE promotes Domain Specific Modeling Languages.
4. UML favours huge metamodels to be filtered; MDE promotes small metamodels to be composed.

And along this line I'd like to mention two other important contributions:

• Richard Page post on The Journal of Object Technology (JOT) blog, criticizing some wrong assumptions or messages regarding model driven engineering.
• Our own book on model-driven software engineering in practice, where we collect several leads and messages on the role, scope and purpose of MDE. 
• And also a word of caution on creation vs. usage of DSLs.

Thanks Jean for your vision!



To keep updated on my activities you can subscribe to the RSS feed of my blog or follow my twitter account (@MarcoBrambi).

At full throttle towards IFML with WebRatio

With its updated site and recent blog post from its CEO Stefano Butti, WebRatio is allowing its customers to move at full throttle towards the new OMG standard IFML, the Interaction Flow Modeling Language.

In particular, WebRatio is now granting the availability of both an opensource IFML editor with proof of concept implementation of integrated fUML and Alf runtime support and of the industrial-strength implementation supporting end-to-end development, from model definition to code generation and deployment (the WebRatio application platform). These are great facilities for future IFML developers and contributors.


To keep updated on my activities you can subscribe to the RSS feed of my blog or follow my twitter account (@MarcoBrambi).

IFML: and here comes the official logo!

The Object Management Group (OMG) has released today the new official logo of IFML, the Interaction Flow Modeling Language that became a standard in March 2013.

The logo is inspired from the well-known UML logo. This is not just a graphical similarity. The aim is to underline immediately in the name and visual representation that IFML is strictly related and integrated with all the OMG modeling language suite and within the MDA framework. Indeed, IFML works perfectly together with UML, SysML, SoaML, and BPMN as a start.. and then you can pair it with any MDA-compliant modeling language.
So please enjoy and share this new logo, and have a look at the official IFML web page.

To keep updated on my activities you can subscribe to the RSS feed of my blog or follow my twitter account (@MarcoBrambi).

Success story paper: Large-scale Model-Driven Engineering of Web User Interaction with WebML and WebRatio

Our paper "Large-scale Model-Driven Engineering of Web User Interaction: The WebML and WebRatio experience" has been published online on Elsevier's journal: Science of Computer Programming, in the special issue Success Stories in Model Driven Engineering (edited by Davide Di Ruscio, Richard Paige, Alfonso Pierantonio).

The history we report spans across a decade that has seen a dramatic  change in the way software applications are built, which can be summarized  in three fundamental factors that impacted the evolution of WebML and  WebRatio:

• The progressive consolidation of theWeb as an application development platform
• At the front-end, the multiplication of access devices and usage scenarios
• At the back-end, Business Process Models emerged as a uniform way of representing cross-organization functionality, and Service Oriented Architecture as the technical vehicle for deploying process enactment on top of heterogeneous IT infrastructures.

These change drivers put much strain on a DSL like WebML, born for capturing the  features of the Web, and produced the timeline shown below:

The paper reports on our experience with WebML and WebRatio and describes the perspective of the new IFML standard adopted by OMG. The report tells the story of our company in the MDE tool market, facing the challenges of deploying MDE solutions in large-scale industrial players, with a focus on the model-driven design of user interaction and on code generation across all the tiers of Web/SOA applications. We describe our decisions on the DSL (domain specific language) and on the features we decided to implement (or not) in the tool. 

The paper includes an overview of WebRatio and of its accompanying DSL for Web application design (WebML); it describes the parallel evolution of the WebML language and of the WebRatio development environment; it reports on the the lessons learnt from the joint design of the DSL and of its support tool; it presents a sample of customer histories and reports some quantitative measures on the WebRatio usage, together with some statistics on WebML models size and development effort. Finally, we take the occasion to reflect on the success and failure factors for MDE emerged from the WebRatio experience.

The paper is available from Elsevier and also here in our open-access preprint version.

To keep updated on my activities you can subscribe to the RSS feed of my blog or follow my twitter account (@MarcoBrambi).

IFML is now adopted and published by OMG, Wikipedia, and YouTube

Along with the continuing consolidation process of the Interaction Flow Modeling Language (IFML) within OMG, we are now starting our effort of documenting and disseminating IFML.

The three most recent resources I wish to point at are:

• Official publishing of IFML in the OMG specification catalog
• A new Wikipedia page on the language
• A new demonstration video on the ongoing implementation effort of an opensource IFML editor

OMG Publishing 

OMG has now officially published IFML in its list of specifications:

• http://www.omg.org/spec/

and the official specs are now reachable through the official URL of the standard at:

• http://www.omg.org/spec/IFML/

IFML on Wikipedia 

The Wikipedia page regarding IFML is now online at:

http://en.wikipedia.org/wiki/Interaction_Flow_Modeling_Language  

The page summarizes the focus and content of the language, as well as the benefits that developers can gain by adopting it.

Demo video

The demonstration video of the new IFML editor that we are developing is available on YouTube.
This implementation consists of an opensource IFML editor based on Eclipse, EMF/GMF and the Graphiti API. At the moment the tool is under development, but it will be released as opensource Eclipse Project as soon as it reaches a reasonable level of completeness. The tool also includes the mappings from the IFML abstract concepts to the platform- specific concepts of Java Swing, Microsoft WPF, and HTML, as described in the IFML specification.

The modeling of the IFML diagrams for the UI part can be complemented with (ex- ecutable) UML diagrams according to fUML specifications combined with Alf scripts for the backend business logic.

A sneak preview of the tool and generation features is reported in the video, also shown below:

To keep updated on my activities you can subscribe to the RSS feed of my blog or follow my twitter account (@MarcoBrambi).

Expert finding in social networks and crowdsourcing

Expert selection is an important aspect of many Web applications, e.g.,  when they aim at matching contents, tasks or advertisement based on user profiles, possibly retrieved from social networks.

This was crucial for our current research on crowdsourcing, and therefore we dedicated a specific research line to this aspect. The main idea we developed was to define a model of the user expertise that takes into account user activities and roles, but also the profiles of his friends and acquaintances.

After one year of work with two M.Sc.  thesis students, we reported our results in a paper that has been presented by Stefano Ceri at EDBT 2013 in Genova.

You can look at the presentation here:

Choosing the right crowd. Expert finding in social networks. edbt 2013 from Marco Brambilla

The paper focuses on selecting experts within the population of social networks, according to the information about the social activities of their users. We consider the following problem: given an expertise need (expressed for instance as a natural language query) and a set of social network members, who are the most knowledgeable people for addressing that need?
We considers social networks both as a source of expertise information and as a route to reach expert users, and define models and methods for evaluating people's expertise by considering their profiles and by tracing their activities in social networks.
For matching queries to social resources, we use both text analysis and semantic annotation. An extensive set of experiments shows that the analysis of social activities, social relationships, and socially shared contents helps improving the effectiveness of an expert finding system.

The full paper will be available soon on the ACM Digital Library.

To keep updated on my activities you can subscribe to the RSS feed of my blog or follow my twitter account (@MarcoBrambi).

Interaction Flow Modeling Language (IFML) adopted as a standard by OMG

It's with great pleasure and also some pride that I'm announcing that the Interaction Flow Modeling Language (IFML) has been adopted as a standard by OMG during the technical meeting held in Reston, VA this week.

This is an important result we have been pursuing for more than two years now (and ultimately have its roots in the work started almost 15 years ago).

Indeed, IFML has been promoted by WebRatio, a spinoff company of Politecnico  di Milano university, born during a European cooperative research project in a period of time (1999-2001) when the Web was changing the way in which software was implemented. The company produces a tool (called WebRatio itself) for model-driven development of enterprise Web applications, based on the WebML modeling language. And IFML comes out right from there.

The OMG standardization process is not an easy one, but IFML went through it all the way down to adoption. Initially our initiative was seen with some skepticism and suspicion by the big guns of software design, system design and system architectures in OMG. User interaction was (and still is) seen as a side or minor issue in such development scenarios. Actually, this is probably true in design of such mega-systems, where the focus is multi system integration and evolution, with most attention (an money) going into military, embedded, and crowss-government integration.
However, I dare to say that finally IFML was regarded with respect and positive attitude by all the people that got informed about it (and actually they have been quite a lot, as I managed to keep pushing it meeting after meeting with everybody I managed to talk to.

This last month has been very hectic for IFML. We had to work at the specification document, at the implementation of the tool, at the code generation prototype and at the integration with the fUML interpreter.
And finally, we received the official review comments of the OMG Architecture Board (AB). The reviews were very positive, and I was particularly please because they were coming from UniSys and IBM board members. 

The Architecture Board of the Object Management Group discusses IFML in Reston, VA.

This week the approval process went through various steps at the OMG meeting: introduction at the AB meeting on Monday; thorough discussion at the ADTF (Analysis and Design Task Force) on Wednesday, followed by the two voting phases needed for ADTF to recommend the specification for adoption; quick presentation of the IFML submission at the AB meeting on Thursday, followed by vote for adoption; and final report to the Platform Technical Committee on Friday, where the charter for the IFML Finalization Task Force has been issued.

I must say I was surprised by the enthusiasm we raised. At the ADTF the two voting phases both collected unanimous approval, with almost no criticisms. And unanimity was reached also at the adoption vote at the Architectured Board. I dare to say IFML has been one of the fastest approvals cycles that OMG ever saw. The pre-voting comments from UniSys and IBM were literally two words each: "ship it!" and "thumbs up!" respectively. Very encouraging!

The transparency proposing the IFML Finalization Taskforce
(then extended with people from SparxSystem and 88Solutions).

But this is not the end of the path. There is actually still a long way to go. We will need to set up the Beta version of the specifciation document (which will be publicly available sometime in April), and then collect comments and notes until December 2013, and release the final Version 1.0 of the standard on March 2014.
Anyway, the most important result was achieved this week and it's something we all at WebRatio and Politecnico should be proud about.

I'm not going to go through the technical content of the standard in this post, we will have time for that later on (if you want to get a sneak peak on the specification, you can see this post). This is instead the time for rejoicying and acknoledging the work of all the people that contributed to this result.
The list of contributors is huge. It starts with Piero Fraternali and Stefano Ceri, who started the WebML project (and the AutoWeb project before). Then it counts Aldo Bongio, Roberto Acerbis and Stefno Butti. And then all the people that for some time joined the WebML / WebRatio adventure.

IFML is the result and the aggregation of all the experiences that have been collected in academia and industry in these 15 years. And this includes all the work of the developers that implemented the WebRatio tool, the WebML analysts and designers that built so many large-scale industrial applications and made customers happy (and sometimes swear), and also thesis students, researchers and partners at Politecnico di Milano. I'm not going to name anyone here, because the list would be too long (I've estimated some 150 people contributing in these years).

Coming back to present, some people need to be mentioned again as direct contributors to IFML, primarily the co-submitters of the standard proposal: Model Driven Solutions (and here a big thanks goes to Ed Seidewitz), Softeam, Fujitsu, and also people at Politecnico that helped both in the writing of the proposal and in the implementation of proof of concepts, still ongoing, but also in the AtlanMod group at EMN, at Soluta.net, and others.

The final point I want to make now is that IFML is now an open specification and it's on the lookout for implementors and users.

The WebML editor from WebRatio is already compliant with the notation of IFML, being WebML a more specific case of IFML. We are also implementing an open-source ECORE-based IFML editor, featuring proof of concept generation of Java UI, integrated with fUML interpreter for the backend business logic. And we are planning of supporting some UML tool implementor to help them cover the new IFML profile. If you are interested in implementing IFML at some level (UML profile, domain-specific graphical notation, or possibly a textual notation) and want to hav it mentioned in OMG, feel free to get in touch!

Related posts:

• IFML officially published by OMG and wikipedia
• IFML revised submission 
• All the posts on IFML

To keep updated on my activities you can subscribe to the RSS feed of my blog or follow my twitter account (@MarcoBrambi).

Daniel Moody on "Designing diagramming notations that work": the physics of notations

Daniel L. Moody from University of Twente gave a talk (and a full fledged tutorial) at OMG technical meeting in Reston, VA, on the principles for designing effective diagramming notations. His talk had this overall design:

(click to enlarge)

Daniel Moody wrote a famous paper on IEEE Transactions of Software Engineering titled:

The “Physics” of Notations: Toward a Scientific Basis for Constructing Visual Notations in Software Engineering

IEEE TSE. November/December 2009 (vol. 35 no. 6). pp. 756-779

Program flow charts have been invented long time ago, in 1947.
Visual notations are effective communication tools because they rely on the highly parallel way of reasoning an processing signals of the human brain.
Therefore, they should be optimized for human mind processing. It's not something that can be designed based on opinions. There are ways of measuring and proving results.
The measure of optimization is cognitive effectiveness. Effectiveness of existing notations is hampered by:

• unclear goal of the notation
• low attention to this aspect
• lack of accepted design principles

Notations like ER diagrams expose significant flaws, but is accepted and preserved with no questions. UML has the same problem. In general, another point is that in current notations there is no motivation on why one choose one symbol or another for model elements. In other words, there is no rationale for design of notations. Basically, we are still in unselfconscious design culture (more or less the same as the way huts are built by primitive people: there are no architecture principles or books, the just build huts).

The same symbols are used again and again within the same language and across languages (e.g., think about the rectangle symbol).

There is lack of testing and principles. That's why Moody proposes a set of objective principles based on scientific observation.

Out of that, he proposes a clear manifesto for design notations:

1. At least equal effort and attention for notation and content
2. Have a design rationale
3. Decision must be based on evidence
4. Design must be based on objective principles
5. There must be explicit testing of the notations (user studies)

The purpose is to move diagramming from art to science.


To keep updated on my activities you can subscribe to the RSS feed of my blog or follow my twitter account (@MarcoBrambi).