SoftEther VPN Server is an open-source cross-platform multi-protocol VPN server which has two advantages over existing VPN servers. First, it supports multiple VPN protocols in a single VPN server instance. This makes it easy for an administrator to configure and manage a VPN server which supports remote access and site-to-site connection from a variety of VPN client devices. To realize that, SoftEther VPN Server includes a module called an L2 adapter to exchange messages between layer-3 VPN protocols and layer-2 VPN protocols seamlessly via common virtual L2 switches. The second advantage is that it can virtualize user management and networking, which is an essential function in multi-tenant virtual hosting. SoftEther VPN Server is portable among several operating systems. SoftEther VPN Server gained a total of 242,000 installations around the world from March 2013 to September 2014. The experimental result indicates that SoftEther VPN Server is faster than combinations of native VPN servers when exchanging messages between different VPN protocols.
We have developed a picture language system for constructing 3D models in Scheme, one version of Lisp. Understanding both procedural abstractions and data abstractions is important for all people who learn to write efficient and compact source codes. A picture language has been the most effective way to understand those concepts, because it can present structures of programs as intuitively and visually understandable 2D figures, which was originally introduced in a legendary book of computer science, SICP (Structure and Interpretation of Computer Programs). We designed and implemented the extended picture language for 3D models using JAKLD and enabled the system to export models in 3D printable format, in order to deepen learners' understanding of abstractions by analogy of spatial fractal structures of substantial models. We report the use of our system as a teaching material in a lecture for first-year undergraduate students.
One of the problems with applying a model checking method to a business is stable securement of human resources. So authors developed a model checking support tool that enables model checking method using a sequence diagram widely used as a design tool. In the tool, internal specifications and external specifications written by sequence diagrams are translated into inspection models and inspection specifications, and a counterexample as a result of verification by a model checking tool is translated into a counterexample sequence diagram and expected sequence diagram. As a result, a designer can take benefits of a model checking method without learning it. As an evaluation of the tool, author confirmed that a low reproducibility defect is detected effectively as a result of applying the method to a business model.
We assume that there are three obstacles to learn programming: 1) elements of programming languages, 2) devices of educational tools for programming and 3) user interfaces of educational tools for programming. To the best of our knowledge, there is no educational tool for programming which removes the three obstacles. We propose a new educational tool for programming, named ManekkoDance. ManekkoDance has three features: 1) it manipulates cute a teddy bear robot and a cardboard robot, 2) it works on Android smartphones and tablets and 3) it provides a new programming language based on emoticons and Japanese. We evaluated ManekkoDance using questionary investigation in the open house of National Institute of Informatics. We collected 29 answers from high school students, 13 answers from other participants and 295 answers from mostly junior high school and high school students, then we found MannekoDance motivated them to learn programming, improved impressions of programming and helped to understand procedures, loops and conditions.
The studies of Repository Mining have been actively conducted. However, it is difficult to search projects with specified languages, development scale, purposes and so on. In this paper, we propose RepositoryProbe, a dataset creation support tool fot the study of repository mining. It makes easier to search and collect the projects in project hosting service on the web, and supports the creation of datasets. In addition, it can collect the social metrics, the amount of development activities.
We developed a tool for modeling GALS systems by STPN, generating incidence matrices of the STPN, and estimating a performance index through an iteration of petri-net simulations using the matrices. To reduce the cost involved in modelling GALS systems, our tool supports to draw the STPN hierachically using reusable STPN modules which model common functionalities of GALS systems. In addition, our tool can save the effort of drawing tangled arcs between modules by bundling a set of arcs which connects modules as an arc-bus. On the other hand, matrix calculation for the petri-net simulation can be accelerated by ordering of incidence matrices. The matrices are blocked into sub-matrices with respect to each module and the sub-matrices are converted to band matrices. We focus on the fact that only a part of elements in the incidence matrices are engaged in each calculation for petri-net simulation. The calculation can be further accelerated by using reference tables which dynamically hold involved elements. We applied our tool to examples and showed that the time for estimating performance indexes can be reduced up to 99.8%.
This paper studies how a developer uses crash reports sent from users in order to encourage the users to send the crash reports. This paper could contribute to what the developers understand the fault occurrence in user environment more precisely (for developers) and what the faults that users suffer would be preferentially addressed (for users). We analyze the relationship between the frequency of the crash reports in a crash repository and the probability of linkages of the crash reports and issue reports in Bugzilla. The results using the dataset collected from Firefox show some findings such as 1) 63% of crash reports are linked to issue reports and 2) 90% of the linked crash reports are linked to issue reports up to the 1,000 crash reports.
In the software development, modeling business process is important. In constructing business process model appropriately, stakeholder's requirements should be reflected in the model. Therefore, in this research, we propose transformation approach from goal models using refinement pattern to business process models. It is denoted that rules of transformation and algorithm. Using our approach supports constructing business process models by specifying stakeholder's requirements formally using refinement patterns. We evaluate the effectiveness of our approach through applying our approach for a number of cases and using model-checking techniques.
We previously proposed a method to infer preconditions that are the weakest in a combination of predicates (quasi-weakest preconditions). This method, however, had a problem in performance due to the high cost of minimal-unsatisfiable-core (or MUC) enumeration. MUC enumeration is usually realized by using enumeration of minimal correction subsets (or MCSes) as intermediate solutions, and the MCS enumeration forms large part of MUC enumeration time. In this paper, we propose three fast algorithms for quasi-weakest precondition inference, based on two properties of MCS enumeration: (1) in our setting the size of MCSes can be fixed, and (2) they can be efficently enumerated if some of the MUCs are obtained in advance. Our performance evaluation shows all of these three algorithms were superior compared to a conventional one, with a maximum speedup of 10.7 times. We report the result of the evaluation, and discuss pros and cons between the three.