My dissertation challenges our preconceptions of the ethnic literary tradition in the United States. Minority literature is generally read within a framework of resistance that prioritizes anti-hegemonic and anti-racist writings. I focus on a set of recalcitrant texts, written in the first part of the twentieth century, that do not fit neatly within this framework. My chapters trace an arc from Ameen Rihani's The Book of Khalid (1911), which personifies a universal citizen who refuses to be either Arab or American, to Paule Marshall's Brown Girl, Brownstones (1959), which dramatizes the appeal of white identification for upwardly mobile Barbadian immigrants. I present the first comparative analysis of Afro-Caribbean, Arab, Filipino, Latino, and South Asian immigrant writings. This archive includes familiar figures such as Claude McKay and William Carlos Williams as well as understudied writers such as Abraham Rihbany and Dalip Singh Saund. I argue that these texts feature a common character: a character who does not want to be exclusively minor, who seeks to identify as widely as possible with majoritarian formations. I propose that this archetype of mixed identification be understood as "brown." Brownness is not a racial category but a literary characterization. It represents the unwieldy paradox of the majority-identified minority subject, a paradox that criticism of ethnic literature has largely ignored. I chart the complex attachments of these brown characters, their Orientalist mediations, asymptotic Americanisms, approximations of whiteness--in sum, their disidentification with minoritization itself. By exploring these vexed desires, I help to explain the perennial attraction of brown characters not only for racialized minorities but also for narratives of U.S. exceptionalism.

The human gaze serves as a potential non-verbal cue that enhances human-computer interfaces, enabling users to engage with devices through eye movements. The ability to accurately measure and interpret gaze direction plays a critical role in various domains, including social interactions, assistive technologies, augmented reality, and psychological research to examine cognitive state.

Over the past decade, gaze estimation has emerged as a prominent area of interest within the research community. Conventional gaze estimation methods rely on specialized hardware, including high-resolution cameras, infrared light sources, and image processing units, to detect eye features like the pupil center and iris boundary. While these devices offer greater accuracy and precision, their practical use is limited by factors such as high costs, restricted head movements, and limited range of allowable distances between user and device. As an alternative to dedicated gaze-tracking hardware, several techniques have been developed to infer gaze direction directly from eye images captured by standard cameras on personal devices such as laptops, tablets, and phones.

The recent emergence of deep learning techniques has enhanced learning-based gaze estimation approaches. These appearance-based gaze estimation methods directly map eye images to gaze targets without the need for explicit detection of eye features and, therefore, have a strong capability to work in unconstrained environments. However, the effectiveness of these approaches greatly depends on having access to extensive training datasets that include a variety of eye appearances, gaze directions, head poses, lighting conditions, and other variables. In this thesis, we focus on improving the adaptability and effectiveness of webcam-based gaze estimation techniques through the application of generative modeling and representation learning.

First, we propose an easy approach for calibrating a laptop camera with a commercial gaze tracker, streamlining the process of collecting labeled gaze data to make it readily accessible for all users. This dataset can then be utilized to enhance the accuracy of appearance-based gaze estimation methods for new users and different domains.

Second, we introduce a generative redirection framework designed to manipulate gaze direction and head pose orientation in synthesized images. This framework is used to generate augmented, gaze-labeled datasets, thereby enhancing the performance of gaze estimation methods.

Third, we explore self-supervised contrastive learning to acquire equivariant gaze representations through an unlabeled multiview dataset. These gaze-specific representations are utilized for few-shot gaze estimation, enhancing the efficacy of user-specific models.

Finally, we present a spatiotemporal model for video-based gaze estimation, incorporating attention modules to enhance understanding of both local spatial and global temporal dynamics. Furthermore, we improve the performance of this model using person-specific few-shot learning through Gaussian processes.

Open-source projects are pivotal to the innovation and sustainability of software, yet their complex dynamics can be challenging to decipher. Building upon the foundational work of the DECAL group, which demonstrated the utility of socio-technical networks in analyzing the lifecycle of projects, and the APEX tool developed for Apache projects to forecast sustainability, this thesis extends the idea and introduces the Eclipse Project Explorer (EPEX) tool. Employing dual-network models—social and technical—this research explores open-source collaborations within the Eclipse ecosystem. The social network leverages data from mailing lists and other communication platforms to visualize interactions among developers, while the technical network utilizes GitHub commit data to trace file changes, illuminating the evolution of software projects. These networks are visualized over time, offering developers deeper insights into project lifecycles and interaction patterns. Furthermore, we introduce a groundbreaking ‘graduation forecast‘ model that predicts the long-term sustainability of projects based on analytical data. By applying our methodology to selected case studies within Eclipse, our findings not only validate the effectiveness of our approach but also enhance our understanding of project sustainability, providing a robust tool for strategic decision-making in software development. This thesis lays foundational work for future research into the lifecycle management of open-source software, significantly extending prior research.

Genome Sequence and the Identification of Mosquitocidal Toxin operons in Clostridium bifermentans subsp. malaysia

by

Swati Chawla

Doctor of Philosophy Graduate Program in Environmental Toxicology

University of California, Riverside, March 2015

Dr. Sarjeet S. Gill, Chairperson

The management and control of mosquito vectors of human disease currently rely primarily on chemical insecticides. However, larvicidal treatments can be effective, and if based on biological insecticides, they can also ameliorate the risk posed to human health by chemical insecticides. The aerobic bacteria Bacillus thuringiensis and Lysinibacillus sphaericus have been used for vector control for a number of decades. But development of mosquito resistance to individual toxins has been seen in laboratory and in field. More recently, the anaerobic bacterium Clostridium bifermentans subsp. malaysia (Cbm) has been reported to have high mosquitocidal activity, and a number of proteins were identified as potentially mosquitocidal. However, the cloned proteins showed no mosquitocidal activity.

I determined genome sequence of the only known anaerobic mosquitocidal bacterium, Cbm by the next generation sequencing approach. I showed that the genome consists of a 3.7-Mb chromosome, seven small plasmids and a 109-kb megaplasmid that contains the two toxin operons responsible for mosquitocidal activity. Also, additional virulence-related factors could be identified for further work.

I co-authored and published that four toxins encoded by the Cry operon, Cry16A, Cry17A, Cbm17.1, and Cbm17.2, are all required for toxicity, and these toxins collectively show remarkable selectivity for Aedes rather than Anopheles mosquitoes, even though Cbm is more toxic to Anopheles. I also tried to improve the mosquitocidal activity of Cbm Cry operon by protein engineering of surface loop residues in domain II of Cry17Aa and wanted to test on 4 different species of important human disease vectors, Aedes aegypti, Anopheles gambiae, Anopheles stephensi and Culex quinquefasciatus.

I also identified the toxins that could target Anopheles are different from those expressed by the Cry operon and investigated the possibility that the genes encoded in second operon on pClosMP, the Clostridial Mosquitocidal Protein (Cmp) operon, could be toxic to Anopheles. The Cmp operon encodes seven genes, including a 150-kDa mosquitocidal toxin (CMP1) and HA gene. I show the three genes, CMP1, CNTNH, and CHA that play an important role in botulinum neurotoxicity are not active individually against mosquitoes. We need to express the native full length 16kb Cmp operon and CMP1-CNTNH-HA constructs to answer this question.

In summary, based on data in the literature and my work, I demonstrate that Cbm harbors its mosquitocidal activity on a large 109kb plasmid that encodes for two toxin operons. The Cry operon toxins work as a complex and are active against Aedes larvae. The Cmp operon toxins activity against Aedes and Anopheles larvae remains a vast field of study to be further explored.

Abstract:This dissertation joins the contemporary debate on rewriting the history of Chinese architecture by critically (re-)examining the initial stage of discourse formation during the Republican period. Focusing on the ordinary house in the city, I excavate the popular voice on Chinese architecture from the newspapers and magazines, examine the municipal government’s effort in modernizing the city and life, as well as the professional community’s attempt to establish the authoritative figure of the modern architect to replace that of the traditional craftsman. I show that these different constituencies of Chinese architecture articulated and presented a more vibrant and complex scene of discourse and practice than we have recognized until now. Although each held different forms of power, and focused on different issues, these early interpreters of Chinese architecture nevertheless converged on their understanding of reforming the Chinese house and Chinese architecture. The municipal government’s modernizing projects in the city served to bolster the professional community’s claim to expertise that culminated in writing a new history of Chinese architecture. This history subsequently became the standard historiography, suppressing and subsuming the popular voice and relegating it to history’s dustbin for decades. By retrieving the popular discourse of Chinese architecture from this period, we not only gain a fuller understanding of the formative stage of Chinese architectural historiography, but are also provided with helpful indications pointing to a possible route to rewrite the history of Chinese architecture.

Key words: architectural history, Chinese architecture, historiography, popular discourse, modernity in China, modern China

This dissertation presents the first architectural history of American hackerspaces, a project designed to explore and understand the evolution of a countercultural form of architecture and entrepreneurial social space that has become common in postindustrial and post-recession US cities. Examining six hackerspaces of varied sizes and constituencies established in the Detroit and San Francisco metropolitan areas between 2007 and 2013, the dissertation considers how hackerspaces interface with American public life and public space. It argues that while nonprofit hackerspaces position themselves as public resources to ameliorate post-recession economic and social precarity, spatially and socially they tend to repeat the exclusionary politics of public space that permeate US urban and architectural history, where spaces that strive to be all-inclusive alternatives to state and market structures fracture along identity lines and unevenly allocate their resources. The dissertation examines, for example, how predominantly white and masculine cultures in the first US hackerspaces led to communal reckonings, increasing insularity, and offshoot hacker spaces that center women, LGBTQIA communities, and other marginalized constituents. Using fieldwork and engagement with vernacular architecture, global urbanism, and media studies literature, the dissertation explores the tension-filled transition of hacker communities from diffuse countercultural scenes to formal participants in the urban fabric. Using interdisciplinary methods to interpret original photographs, architectural documentation, and ethnographic data collected between 2018 and 2022, the dissertation demonstrates how hackerspace communities active in the San Francisco and Detroit areas inscribe digital and political utopianism in physical space. Hackerspace cultures and buildings thus offer microcosms of how aspirations of openness, diversity, and access to public resources outside the bounds of corporate culture, academia, and government programs fracture along identity lines. Centering the fugitive spatial practices of hackerspaces, the dissertation offers new directions for an inclusive architectural history of normalized precarity in post-recession and COVID pandemic-era cities in the US.