Mark Newsome

Bio: Mark Newsome is an academic researcher from Oregon State University. The author has contributed to research in topics: Web application & Web query classification. The author has an hindex of 4, co-authored 5 publications receiving 33 citations.

HyperSQL: web-based query interfaces for biological databases

Abstract: HyperSQL is an interoperability layer that enables database administrators to rapidly construct browser-based query interfaces to remote Sybase databases. Current browsers (i.e., Netscape, Mosaic, Internet Explorer) do not easily interoperate with databases without extensive CGI (Common Gateway Interface) programming. HyperSQL can be used to create forms and hypertext-based database interfaces for non-computer experts (e.g., scientists, business users). Such interfaces permit the user to query databases by filling out query forms selected from menus. No knowledge of SQL is required because the interface automatically composes SQL from user input. Database results are automatically formatted as graphics and hypertext, including clickable links which can issue additional queries for browsing through related data, bring up other Web pages, or access remote search engines. Query interfaces are constructed by inserting a small set of HyperSQL descriptors and HTML formatting into text files. No compilation is necessary because commands are interpreted and carried out by the special gateway, positioned between the remote databases and the Web browser. Feedback from developers who have used the initial release of HyperSQL has been encouraging. At present, query interfaces have been successfully implemented for three major NSF-sponsored biological databases: Microbial Germplasm Database, Mycological Types Collection, and Vascular Plants Types Collection.

Visual execution of assembly language programs

Abstract: Students learning computer organization and assembly language programming often have difficulty understanding the relationship between hardware functions and software instructions. This paper describes how a software tool can help students understand the organization and use of the Motorola 68000 processor family. XSIM68K employs a graphical interface to enhance a traditional “black box” simulator so that students can observe first-hand the effect of a running program on registers, memory, and II0 devices. Multiple data representations allow the student to choose the most appropriate format for data entry and display, which reduces confusion and eliminates conversion mistakes.

Microbial Germplasm Database: Facilitating Access to Microbial Germplasm through Electronic Networks

Abstract: Microbial germpiasm is essential to agriculture and global ecosystems. The hundreds of thousands of organisms maintained in research collections cannot and should not be maintained in centralized collections. Access to them can be facilitated by a universally accessible database containing information on the location and characteristics of these organisms. Microbial Germpiasm Database has been developed using the world wide web and custom software to provide easy intuitive access to this information. An overview of that access is provided in this paper.

QueryDesigner: Simplifying Access to Remote Scientific Databases

Abstract: Many large-scale scientific data collections are stored in relational databases that use command-line SQL as the query interface. For years, scientists and other occasional users have had a difficult time accessing their data because SQL is terse and unforgiving of errors. With the advent of the Web browser, forms-based interfaces have been layered on top of relational databases, shielding the end-user from the complexity of SQL. While these interfaces are easy to use, building them requires programming experience. The increasing demand for additional database interfaces is often not met due to declining personnel budgets and a shortage of programming staff. Our solution was to build a user-oriented tool called QueryDesigner that enables noncomputing experts to explore, construct, and personalize query interfaces from a Web browser. QueryDesigner is a Java applet that eliminates the requirement that the interface designer understand SQL, HTML, or any specialized language. QueryDesigner was built in collaboration with scientists who actually work with large-scale biological databases. It was designed to allow the scientist to easily compose forms-based interfaces that support the way they search databases. QueryDesigner helps the user understand unfamiliar databases by acquiring information on database organization and presenting it as interactive diagram. The applet transparently composes SQL queries, issues requests to a remote database, receives and formats the results. Scientists Have Special Requirements Unfortunately, scientists — who are often not computer experts — do not have an easy time retrieving information stored in SQL databases. To formulate an SQL query, the user must first look up the names of tables and fields, which requires the use of a proprietary (non-standard) tool (e.g., Oracle's sqlp u ). Moreover, SQL itself is terse, error prone, and unfriendly. It is easy to misspell a keyword or data name, omit a “join” clause, or construct a query that returns unintended results [Sme95]. There is also a lack of meaningful feedback during the query process, despite the fact that it is unlikely a casual user will be able to formulate a query correctly on the first attempt [JV85]. The user bears the burden of determining what went wrong, based on cryptic messages that reveal little about the source of the problem. We were able to capitalize on what we learned about formsand hypertext-based query interfaces for biological researchers during a previous interdisciplinary project [HNPM95] where we analyzed how users interact with biological databases [New96]. Scientists in the biological disciplines typically begin searches with only a partial idea of where they are going and what kind of results might be available. It is crucial for them to be able to explore the data and investigate the relationships between database entities, without having to formulate complex SQL queries. With incomplete criteria, a search is likely to return a large result set; most database software simply returns all results in a rapidly scrolling display. Scientists need to be able to determine the size of the results before they are returned, in order to know whether the query needs refining. Support is needed for retrieving results in incremental blocks, paging through them, as well as saving the data in spreadsheet format. Moreover, keyword search is essential in scientific databases because they often contain unstructured data such as field notes or historical annotations. Another common characteristic of biological data is the inclusion of a large number of attributes; no single organism is likely to contain values for every attribute, so database records contain a large number of empty (or null) data fields. This makes it difficult for users to find pertinent information, since current database management software provides no means of suppressing empty (“NULL”) output. We also found that a number of unsuccessful queries were caused by users' misspelling long scientific names, orthographic and typographic errors in the database itself, or the use of discipline-specific naming variants. When users must type values into empty blanks on a form, specifying search criteria becomes a guessing game. When no results are returned from a query, users are left wondering whether there really is no data available, or if their input was incorrectly specified. How Query Designer is Used The QueryDesigner applet (see http://www.nacse.org/qd ) employs a graphical interface that permits the end-user to either personalize or create a forms-based query interface to a remote database. The tool eliminates the need for the end-user to learn SQL or proprietary scripting languages. Instead, the user formulates queries implicitly, by specifying on an interactive E-R diagram which data fields should serve as inputs and outputs. QueryDesigner checks to ensure that the joins are legitimate and that all database elements are joined, preventing several of the most common semantic errors. Once the interface elements have been specified, the tool generates an initial query interface, using default rules for placing elements; this can be edited by the user. In designing a query form, the user can specify whether an input field is to appear blank, pre-initialized to a constant, or shown as a query-list. The query-list feature eliminates user guesswork and spelling mistakes on the query form, by pre-fetching a list of allowed valued from the database. This is particularly useful for scientific databases, given that taxonomic and other scientific names tend to be long and easily misspelled. Although keyword searching is not supported directly by SQL, our software permits the scientist to search for keywords, even when it is not known in advance which fields should be searched. When submitting a form, QueryDesigner automatically composes an SQL query, establishes a connection to the database, receives the results, and displays them. By default, QueryDesigner automatically suppresses the printing of empty (or NULL) fields to prevent screen clutter. A pull-down is used to limit the number of results returned at a time. A slider at the bottom of the screen permits the user to scroll through results one record at a time. Clicking the SAVE-AS-SPREADSHEET button permits the user to save results in a form accepted by most spreadsheet software. Finally, QueryDesigner offers advantages in terms of its architecture. Accessible from a Web browser, QueryDesigner is platform independent as well as location-independent of the underlying database. Perhaps the most important advantage is that QueryDesigner is independent of any particular DBMS software. Through JDBC, QueryDesigner supports most relational databases that are addressable over the Internet.

Hypertext Transfer Protocol (HTTP)

Abstract: In this report (written in Greek) are presented the basics of the hypertext transfer protocol.

Method and system for injecting code to conditionally incorporate a user interface component in an html document

Abstract: A method and system for injecting code into an existing HTML document that enables the conditional generation and incorporation of an injectable component is provided. The injection mechanism provided enables a user to display an add-on user interface component within a web page using a technique that is transparent to the normal processing within a browser application. The injection mechanism installs interceptor code as a proxy server between a client browser and a server. The interceptor code monitors all HTTP messages sent from the client browser and injects code into each HTML document that is sent from a server to the client browser. The injected code when executed by the client browser conditionally inserts HTML code into the received HTML document. The inserted HTML code is then used by the client browser to generate a new component, for example a user interface component. The code injected by the interceptor code is dependent upon the type of the browser and whether certain tags are present in the received HTML document. In addition, the code may be written in a script language, such as Javascript or VBscript.

Presenting web pages with discrete, browser-controlled complexity levels

Abstract: A method of browsing the Worldwide Web of the Internet and presenting elements of a web page to a user. The web browser requests a desired HTML and then requests other elements of the page based upon user defined priority levels. Elements are presented to the user in priority sequence as soon as received eliminating unnecessary waiting for lower priority or unrequested elements, as well as partial display of desired elements.

Method and system for generating materials for presentation on a non-frame capable web browser

Abstract: A method and system are disclosed for managing Internet presentation materials in a single file format for ease of administration while presenting to an Internet requestor only those portions of the file requested, for maximum performance. Also disclosed is a system and method for presenting Internet materials using borderless presentation areas, where the background specification is decoupled from the presentation area specification. The invention also relates to a system and method for using a dynamic web page builder to generate and manage multiple instances of information to be simultaneously displayed in multiple presentation areas, in which one of the presentation areas contains table of contents information listing various selectable web pages stored in a single file. The table of contents information is continuously displayed on-screen when any of the items listed in the table of contents is selected for ease of navigation through a web site.

Symbiota - A virtual platform for creating voucher-based biodiversity information communities.

Abstract: We review the Symbiota software platform for creating voucher-based biodiversity information portals and communities. Symbiota was originally conceived to promote small- to medium-sized, regionally and/or taxonomically themed collaborations of natural history collections. Over the past eight years the taxonomically diverse portals have grown into an important resource in North America and beyond for mobilizing, integrating, and using specimen- and observation-based occurrence records and derivative biodiversity information products. Designed to mirror the conceptual structure of traditional floras and faunas, Symbiota is exclusively web-based and employs a novel data model, information linking, and algorithms to provide highly dynamic customization. The themed portals enable meaningful access to biodiversity data for anyone from specialist to high school student. Symbiota emulates functionality of modern Content Management Systems, providing highly sophisticated yet intuitive user interfaces for data entry, batch processes, and editing. Each kind of content provision may be selectively accessed by authenticated information providers. Occupying a fairly specific niche in the biodiversity informatics arena, Symbiota provides extensive data exchange facilities and collaborates with other development projects to incorporate and not duplicate functionality as appropriate.