This issue's contributors cover works that are results of paradigm shifts where data-intensive systems expanded beyond relational model, disk-based performance optimizations, one-dimensional data, and ''small'' scale. Enjoy reading! While I will keep inviting members of the data management community, and neighboring communities, to contribute to this column, I also welcome unsolicited contributions. Please contact me if you are interested.

Welcome to ACM SIGMOD Record's series of interviews with distinguished members of the database community. It is my pleasure to host Angela Bonifati today. She is a Distinguished Professor of Computer Science at Lyon 1 University, Head of the Database Research Group at the CNRS Liris Research Lab, and a senior member of the Institut Universitaire de France. I'm H. V. Jagadish, and I'm glad that Professor Bonifati is with us today. So, let's dive in. Welcome, Angela!

This issue's contributors cover papers that focus on different aspects of accessing data: RDFs, approximate nearest neighbor search, and compact hash tables. Furthermore, they all highlight the impact of the papers not only on their own research and career but also for the community in general. Enjoy reading! While I will keep inviting members of the data management community, and neighboring communities, to contribute to this column, I also welcome unsolicited contributions. Please contact me if you are interested.

The unification of large language models (LLMs) and knowledge graphs (KGs) has emerged as a hot topic. At the LLM+KG'24 workshop, co-located with VLDB 2024 in Guangzhou, China, the key theme explored was important data management challenges and opportunities due to the effective interaction between LLMs and KGs. The report outlines major directions and approaches presented by various speakers during the workshop.

A data structure is history independent if its internal representation reveals nothing about the history of operations beyond what can be determined from the current contents of the data structure. History independence is typically viewed as a security or privacy guarantee, with the intent being to minimize risks incurred by a security breach or audit. Despite widespread advances in history independence, there is an important data-structural primitive that previous work has been unable to replace with an equivalent history-independent alternative—dynamic partitioning. In dynamic partitioning, we are given a dynamic set S of ordered elements and a size-parameter B, and the objective is to maintain a partition of S into ordered groups, each of size θ(B). Dynamic partitioning is important throughout computer science, with applications to B-tree rebalancing, write-optimized dictionaries, log-structured merge trees, other external-memory indexes, geometric and spatial data structures, cache-oblivious data structures, and order-maintenance data structures. The lack of a historyindependent dynamic-partitioning primitive has meant that designers of history-independent data structures have had to resort to complex alternatives. In this paper, we achieve history-independent dynamic partitioning. Our algorithm runs asymptotically optimally against an oblivious adversary, processing each insert/delete with O(1) operations in expectation and O(B logN/ loglogN) with high probability in set size N.

When a database researcher or user hears the word ''expressiveness,'' they most likely think about what kind of query some language or formalism supports. For instance, can a given version of SQL express recursion or computation of the transitive closure of a graph? For what fragment of relational calculus can we decide query containment? Can relational algebra express every relational-calculus query and vice versa? The following paper by Gatterbauer and Dunne may forever change the discourse on expressiveness in the context of relational languages by prompting the follow-up question: ''Do you mean logical expressiveness or pattern expressiveness?''

Differential privacy (DP) data synthesizers are increasingly proposed to afford public release of sensitive information, offering theoretical guarantees for privacy (and, in some cases, utility), but limited empirical evidence of utility in practical settings. Utility is typically measured as the error on representative proxy tasks, such as descriptive statistics, multivariate correlations, the accuracy of trained classifiers, or performance over a query workload. The ability for these results to generalize to practitioners' experience has been questioned in a number of settings, including the U.S. Census. In this paper, we propose an evaluation methodology for synthetic data that avoids assumptions about the representativeness of proxy tasks, instead measuring the likelihood that published conclusions would change had the authors used synthetic data, a condition we call epistemic parity. Our methodology consists of reproducing empirical conclusions of peer-reviewed papers on real, publicly available data, then re-running these experiments a second time on DP synthetic data and comparing the results.

When interactively working with data, query latency is very important. In particular when ad-hoc queries are written in an explorative manner, it is essential to quickly get feedback in order to refine and correct the query based upon result values. This interactive use case is difficult to support if the underlying data is large, as analyzing large volumes of data is inherently expensive.

Synthetic data is a vital substitute for real sensitive personal data in supporting social science research and policy studies. Extensive prior research has delved into various models for generating synthetic data, from traditional statistical approaches to cutting-edge deep-learning methods. However, selecting the most suitable one for unforeseen applications poses a significant challenge due to the varying strengths and weaknesses, dependent on factors such as the application domain, data distribution, analytical requirements, and privacy considerations.

We consider the problem of computing differentially private approximate histograms and heavy hitters in a stream of elements. In the non-private setting, this is often done using the sketch of Misra and Gries [Science of Computer Programming, 1982]. Chan, Li, Shi, and Xu [PETS 2012] describe a differentially private version of the Misra-Gries sketch, but the amount of noise it adds can be large and scales linearly with the size of the sketch; the more accurate the sketch is, the more noise this approach has to add. We present a better mechanism for releasing a Misra-Gries sketch under (ε, δ)-differential privacy. It adds noise with magnitude independent of the size of the sketch; in fact, the maximum error coming from the noise is the same as the best known in the private non-streaming setting, up to a constant factor. Our mechanism is simple and likely to be practical. In the full version of the paper we also give a simple post-processing step of the Misra-Gries sketch that does not increase the worst-case error guarantee. It is sufficient to add noise to this new sketch with less than twice the magnitude of the non-streaming setting. This improves on the previous result for "-differential privacy where the noise scales linearly to the size of the sketch.