SpoC 007 - Security throughout the SDLC

From OWASP
Revision as of 07:34, 14 July 2007 by Pauloc (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Back to SpoC 007 Selection page


AoC Candidate: Caseydk

Project coordinator: Dinis Cruz

Project Progress: 0% Complete, Progress Page

Contents

Caseydk - Security throughout the SDLC

Educational and professional background

I have a BS in Electrical Engineer which although I rarely use it, I have been told that I "write code like an engineer". I value simplicity and functionality and work to refactor regularly to make my code reflect these principles. I am a core contributor to dotProject - LAMP-based Project Management System - and prior to starting my own company worked at the Department of Justice with the SDLC on a daily basis. During my time there, I provided Independent Verification and Validation, Risk Management, and Technical Review for various projects related to internal systems and homeland security. Most of these projects were inter-Departmental and required the coordination and involvement of numerous stakeholders from numerous organizations.

Participation and leadership in open communities

In terms of Open Source, I have been a core contributor to dotProject since 2004 and have recently joined the Zend Framework project. Within the dotProject community, I have been a big proponent of ease of use while simultaneously working to close various issues related to security, permissions, and general suitability of the system. While this is simple when you have complete control over your environment, the vast majority of our community is using shared hosting and therefore we have to secure the system multiple ways simultaneously to cover every possibility. My specific contributions were the implementation of additional permissions types - previously View and Edit were often confused - and the implementation of .htaccess to lock down all sub-directories, many extraneous files, and still allow for a variety of access methods. While I don't consider it perfect, I believe we have made huge strides in the past year or two.

The opportunity, challenges, issues or need your proposal addresses

Currently, security across the SDLC is often viewed as an item to checkoff. While this may have worked in the past where access to a system required physical presence, it is short-sighted and dangerous in our currently connected world. The concept of "the seven touchpoints" is a step forward and offer a model of active security processes and procedures, but once again, these are often treated as items to checkoff the list. I propose a research effort to develop and support a compelling call to action so that organizations see that they must implement the practices and procedures out of their own best interest. Recent examples such as the Department of Veterans' Affairs, TJ Maxx, and many others show that organizations with bad practices put us all at risk.

Objectives or ways in which you will meet the goal(s)

I see this effort working out into two major pieces. While there is a fundamental need to understand the security perspective, there's the secondary aspect of getting in the mind of an Evil Guy(tm). First, there is the research and documentation phase. I propose to gather a series of real-world security policies, operations, and practices. While this in itself may not be interesting to most, I further propose to supplement this information with real-world examples of these practices in action with a focus on the local Federal Agencies and their supporting organizations. I believe this perspective is something that should provide for interesting discussion and comparison. Out of these various policies and actions, I hope to combine this to identify how "real practices" compare with "best practices" and discuss areas where these differences do and do not make sense. This portion of the project will be entirely done by me.

The second and more interesting part of this project is an Abuse Case Repository (ACR). Many system architects and developers have a single view of their system: How it's supposed to be used. By having a repository of Abuse Cases, I hope to provide a one-stop-shop for nefarious thoughts and practices to get minds rolling and ideas flowing. Hopefully by seeding this with a number of real-world examples and small explanatory code snippets, this will encourage other people to make contributions and suggestions... not only on new and interesting security problems, but also potential solutions. Yes, this sort of repository has the ability to be used for evil purposes. In order to reasonably protect contributors, code snippets would have to be scrubbed to remove system-identifying variables and information. This effort would be started by me but would include anyone with information and ideas to share.

Specific deliverables and a rough project schedule so we can track progress

Milestones:

  • April 9th - Notification / Kickoff
  • April 23rd - Draft Outline of Document
  • April 28th - Abuse Case Repository (ACR) Design/Draft/RFC
  • April 30th - Gathering information on practices, policies, etc
  • June 1st - ACR Launch
  • June 4th - Evaulate real-world information gathered, determine value, usefulness, and targets for completion
  • June 15th - Draft of final report, status report on ACR
  • July 9th - Project "complete"
  • Ongoing - Convincing people to participate in the ACR effort.

Long-term vision for the project

I believe the most valuable aspect to come out of this project will be ACR and I see it outliving this effort as time goes on and more individuals become aware of the risks and challenges they face. Since it will not solve their problems and simply identify potential holes, I believe it can only grow as new attack vectors, technologies, and problems arise. Developing a normalized notation for these risks will be key in the effort.

Any other reasons why you and your project should be selected

I believe that my involvement in the local Washington, DC development community - NoVaLUG, DCLUG, DCPHP, NoVaJUG, NoVaRUG - puts me in a special position to gather this information. Yes, dome of it can be found online in various documents, but there are volumes of information in the hands of the developers and system administrators. This information will not be classified, sensitive, etc, but may be considered embarrassing and therefore would will never see the light of day. In addition, I am active with a number of groups and have connections to various security researchers who may be able to help seed and potentially promote the effort.


Back to SpoC 007 Selection page