Writing an academic paper and getting it published is a one way to help advance your career and promote your company. Most people learn how to write academic papers while pursuing a Ph.D. at a research university. Writing an academic paper is difficult and time-consuming, but anyone can write such papers. Here are a few suggestions and opinions about writing a research paper and submitting it to a formal academic conference such as the International Conference on Information Technology New Generations (see http://www.vteOnline.com/ITNG2010/).

• Your topic must represent new information rather than merely summarize existing information.
• For Microsoft authors, good choices are descriptions of novel techniques, frameworks, and algorithms.
• Microsoft authors, do not directly write about Microsoft products.
• Do not directly write about products under development.
• Your paper must briefly explain why your work is new and relevant.
• Your paper must have at least one paragraph which reviews existing related research.
• The tone of your paper should be more formal than the tone used for magazines or books.
• It is OK to use the past tense if you are describing some experimental type activity you performed.
• If there are joint authors you can use "we" as in, "We executed our program against six benchmark datasets."
• Even if you are the sole author, you should not use the word "I". Use passive tense instead.
• Your paper absolutely should have at least one data table with some sort of results.
• If your paper describes a system of some sort, you should probably have a figure which represents the system.
• Formatting your paper is extremely important. Each conference has a very specific template you must follow exactly.
• If your paper has an algorithm, an O(n) order of magnitude analysis is usually appropriate.
• Limit your presentation of code. Your audience is extremely smart and wants algorithms, not code.
• An "areas of future research" paragraph is usually a good idea.

Here are the titles of accepted papers from the 2009 ITNG software testing track:

• Using the Multi-Attribute Global Inference of Quality (MAGIQ) Technique for Software Testing
• The Design of an Automated Unit Test Code Generation System
• Model Based Software Estimation
• Test Object Model
• Automated Testing of Environment-Dependent Programs: A Case Study of Modeling the File System for Pex
• A Dependence Graph-based Test Coverage Analysis Technique for Object-Oriented Programs
• Declarative Testing: A Paradigm for Testing Software Applications
• The Architecture, Design, and Operation of a Virtual Network Hardware Emulation (VNHE) System
• The Design of a System for Testing Database-Centric Software Applications using Database Surrogates
• Test Automation Framework for Implementing Continuous Integration
• Testing SQL Server Integration Services Runtime using Models and Mock Objects
• TAO Project: An Intuitive Application UI Test Toolset
• An Approach for SQL Injection Vulnerability Detection