http://www.ideals.illinois.edu:80 The DSpace digital repository system captures, stores, indexes, preserves, and distributes digital research material. http://hdl.handle.net/2142/30852 Issues in Humanities Data Curation Muñoz, Trevor Renear, Allen H. As both the materials and the analytical practices of humanities research become increasingly digital, the challenge of sustaining meaningful access to the outputs of humanities research becomes more urgent. Continuing technological change and new institutional pressures require a sustainable commitment to curate humanities data throughout its entire lifecycle from creation to re-use and long-term preservation. Success will depend on coordinated, collaborative efforts and arrangements amongst scholars, librarians, administrators, funders, and their organizations. This document is intended to provide background and provoke discussions about the skills, professional roles, training, and institutional support needed for curation of humanities research materials. Each section contains questions for further exploration and debate that we hope will provide participants with opportunities to share their own experiences and knowledge. Data Curation Humanities Education Curriculum Data Workforce Needs http://hdl.handle.net/2142/30851 Composite Refactorings Vakilian, Mohsen Chen, Nicholas Negara, Stas Zilouchian Moghaddam, Roshanak Johnson, Ralph E. Refactoring tools were invented more than a decade ago and modern IDEs support many refactorings. However, studies have shown that programmers often refactor manually even though it is tedious and error-prone to perform complex refactorings. Many complex refactorings are composed of smaller ones. To design usable tools that match programmers' practices, we need to better understand how programmers compose refactorings to make bigger changes. This paper presents the first empirical study on the composition of refactorings. We identified the frequent patterns of composing refactorings from the data that the Eclipse foundation collected from 195,105 users. Then, we examined the patterns in a smaller but richer set of refactoring usage data that we collected from 2,296 programming hours of 30 programmers. We designed an alternative composition-based interface for automated refactorings and distributed a survey to compare it with the existing wizard-based interface. The results of our survey from 100 respondents show that our approach is likely to increase the utilization of automated refactorings and improve the productivity of programmers. Refactoring Restructuring Composition Empirical http://hdl.handle.net/2142/30850 Growth and magnetoelastic behavior of b-axis-oriented dysprosium Ritley, Kenneth Allan The rare earth metals have the largest known magnetostrictions, so that their magnetic properties and structural properties are strongly coupled. This coupling is ideally suited for investigation with thin films, in which single crystal films of high quality can be prepared on substrates which cause their lattice parameters to differ from bulk values (strain) and which constrain the strictions which may occur (clamping). The present research describes a new MBE growth procedure to grow single crystal thin films and superlattices of rare earths with the hcp (llOO) b-axis normal to the growth plane. To explore the role of strain, clamping and symmetry-breaking in modifying the magnetic ordering temperatures, ultrathin films of strained Dy have been prepared between YxLu1-x alloy buffer layers, so that strain can be tuned from compressive (x=O) to tensile (x=l) and chosen values in between. The magnetic ordering temperatures, critical fields, and magnetization behavior vary systematically with strain. This behavior is interpreted in terms of fundamental magnetoelastic interactions and shape anisotropy. beta-axis growth behavior magnetoelastic behavior rare earth metals thin films superlattices http://hdl.handle.net/2142/30849 Wavelets and wavelet optimized-finite differences for electronic structure calculations Rao, Vivek To perform electronic structure calculations on inhomogenous systems, it is desirable to use methods which adapt to the problem by using a spatially-varying level of resolution. A wavelet basis can efficiently represent a function which is rapidly varying in certain regions of space. One-dimensional calculations using Daubechies wavelets as an adaptive basis have been performed, showing that a compressed wavelet basis can determine the eigenvalues of a system with high accuracy using relatively few functions. Using the Wavelet Optimized Finite Difference Method (WOFD) [1], wavelets can also be used to determine a grid for finite difference calculations. In one dimension, starting with a guess for the wavefunction on a coarse grid with few points, an accurate solution on a nouniform grid can be evolved. In three dimensions, self-consistent total energy calculations employing density functional theory using real-space grids have been performed on atomis, molecules, and quantum dots. Calculations employing WOFD grids are much more accurate than calculations using uniform grids of the same size. Comparisons are made with other adaptive methods. wavelets Daubechies wavelets Wavelet Optimized Finite Difference Model http://hdl.handle.net/2142/30848 Correlation lengths, disorder, and dynamics of vortex matter as heard through the noise Rabin, Michael William I have conducted a series of experiments measuring flux-flow voltage noise in the type II superconductor 2H-NbSe₂ concentrating on the dynamics and possible phase transitions of vortex matter associated with the peak-effect. I emphasize five conclusions: (1) Traditional models of vortex dynamics relying on purely dynamical correlations, rigid bundles, or simple two-state channel flow are invalid in the peak regime of2H-NbSe₂. (2) The noise reflects vortex configurations present in the pinned state, and our methods probe the static correlations among the vortices in the pinned state. (3) The broad-band noise is sensitive to extremely small AC magnetic field perturbations (at ~ 100 Hz); δH/H~ 10⁻⁵ can reduce the noise power at 1 Hz by a factor of2. This sensitivity of the broad-band noise reflects a loss of configurational memory, which is directly related to a correlation length (Ln) in the static vortex solid. Near the onset of the peak (at Hpl), Ln rapidly reaches an apparent sample-size limit as H decreases, suggesting a continuous phase transition or length-scale crossover in the vortex solid associated with the onset of the peak effect. (4) In the narrow range of fields around the noise maximizing field, the fluctuation kinetics are completely different from those at lower and higher H, indicating the inhomogeneous nucleation of a new phase. (5) The remarkable behavior under various types of AC driving implies low-dimensional, deterministic, periodic dynamics during periods when the Lorentz force is zero. For this work, I used the automated, digital synthesis of periodic cancellation signals based on the synchronized sampling of the periodic residual voltage. I invented this technique to automate the exploration of parameter space associated with AC noise measurement. correlation lengths disorder dynamics vortex matter noise type II superconductor 2H-NbSe₂ http://hdl.handle.net/2142/30847 Physical models of protein folding within the energy landscape theory Plotkin, Steven S. The energy landscape theory of protein fol9.ing is based on a statistical description of a protein's complex potential energy surface. Individual folding events are sampled from an ensemble of possible routes on the landscape. The theory suggests that the landscape of a protein can be described as that of a partially random heteropolymer with a rugged, yet funnelled landscape towards the native structure. The quantitative aspects of folding are developed using tools from the statistical mechanics of disordered systems, polymers, and phase transitions of finite systems. The theory provides an interpretation of results from fast folding experiments as well as minimalist models used in computer simulations. protein folding energy landscape theory potential energy surface partially random heteropolymer http://hdl.handle.net/2142/30846 The prevailing model of innovation Vojak, Bruce A. The front and back ends of innovation: The “Fuzzy Front End” and the “Stage-Gate® Process” are perspectives that dominate how we think about innovation. innovation epistemology http://hdl.handle.net/2142/30845 DCEP Final Report Centuries of Knowledge: Graduate School of Library and Information Science Data Curation Education Program Cragin, Melissa H. Renear, Allen H. Senseney, Megan F. The Centuries of Knowledge grant was designed to increase educational and research capacity in data curation at the Graduate School of Library and Information Science (GSLIS) at the University of Illinois at Urbana-Champaign. We developed the Data Curation Education Program, a specialization within our Master of Science degree program, graduating 38 students to date. New courses developed for the specialization include Foundations of Data Curation, a survey course on the emerging field, and Digital Preservation. We developed the Summer Institute on Data Curation for practicing information professionals, facilitating the development of a community of practice across U.S. and Canadian academic and research organizations. Our outreach and service activities have led to a range of new partnerships that have resulted in student fieldwork opportunities, as well as new collaborative research and education activities resulting in 4 successful grant proposals. data curation data curation education http://hdl.handle.net/2142/30844 Keep Calm and Carry On: A Training Model for Integrated Disaster Response for Cultural Collections Lapinsky, Jessica W. Disaster planning in cultural institutions is an important area of collections management that is often overlooked, but vital to the continued operations of an institution when an emergency strikes. Current planning literature emphasizes the need for establishing cooperative networks between cultural institutions and emergency personnel, and standardized response structures for dealing with large scale disasters, such as tornados, floods, and hurricanes. Large institutions, such as the Smithsonian Institution (SI) and the University of Illinois at Urbana-Champaign (UIUC), are evaluating their current collection response plans and training methods to develop a comprehensively integrated response strategy. This strategy will ensure the long-term preservation of their library, archives, and museum collections in the event of a large-scale disaster. A significant first step for the development of integrated disaster response strategies is training all stakeholders in the use of the Federal Emergency Management Agency’s (FEMA) Incident Command System (ICS). The Incident Command System is a flexible, all hazards response structure that is utilized by first responders, government agencies, and private organizations like the Salvation Army and Red Cross. As part of the efforts to develop an integrated disaster response at the University of Illinois at Urbana-Champaign, I created and implemented three workshops for local public library employees, university students studying library, museum, and archives courses, and the general public. The workshops included an introduction to the Incident Command System, salvage techniques for collections, and vital records protection. These workshops explore the different strategic ways collections repositories should respond to large-scale disasters. They can serve as a training model for other large and small institutions seeking to create new disaster response strategies. disaster planning preservation Incident Command System disaster response http://hdl.handle.net/2142/30843 T *GreenHDFS: A Data-Centric, Cyber-Physical Cooling Energy Costs Reduction Approach for Big Data Analytics Cloud Rini T Kaushik Explosion in Big Data has led to a surge in extremely large-scale Big Data Analytic platforms, resulting in burgeoning energy costs. Big Data compute model mandates strong data-locality for computation performance due to Big Data inertia and bandwidth constraints, and moves computations to data. State-of-the-art cooling energy management techniques rely on thermal-aware computational job placement/migration and are inherently data-agnostic in nature. T* GreenHDFS takes a novel, data-centric approach to reduce cooling energy costs. On the physical-side, T *GreenHDFS is cognizant of the thermal-profile in the data centers. On the cyber-side, T *GreenHDFS is aware of the differences in the data-semantics (i.e., computational job profile, job rate, and life spans) of the Big Data placed in the compute cloud. Based on this knowledge, and coupled with its predictive data models, T* GreenHDFS does proactive, cyber-physical, thermal-aware data placement, which implicitly results in thermal-aware job placement in the Big Data Analytics cloud compute model. Evaluation results with one-month long real-world Big Data analytics production traces from Yahoo! show up to 59% reduction in the cooling energy costs with T* GreenHDFS while performing 9x better than the state-of-the-art data-agnostic cooling techniques in the Big Data environment. Energy Cloud Computing Data-Intensive Computing Hadoop HDFS Cooling Cooling Energy Costs Big Data HDFS GreenHDFS Green Computing Cyber-Physical Energy Management thermal Temperature Thermal-Aware Data Placement