Source code documentation

Archiving of software and documentation What is archived, for how long (software, revisions, source code, documentation) Where is the source code archived Can the source code be made accessible to regulatory agencies Is a periodic check of data integrity ensured ... 

Standardized directory structures for source code, documentation hies, hnal executables, conhguration hies, and model inputs/outputs allow tracking the software development process and also help in easily integrating multiple, independently developed modules. Standardized directory structures allow easy detection of con-hicts in the names of functions, scripts, or conhguration hies. 

GOTS refers to government created software, usually from another project. The software was not created by the current developers. Source code, documentation, analysis, and test and results are generally included with GOTS software. 

The scientific area is often characterized by a highly fluctuating and international workforce. This makes it necessary to document the source code and to represent e. g. the class structure in order to introduce new staff to the system easily. Conversely, non-existent documentation may lead to the necessity of reimplementation. 

Because the programs are compiled (. exe format), they can be run on any PC under Windows, but by the same token, they can no longer be tinkered with. For most users, this will not constitute a disadvantage, as even the authors of very well designed and documented source code occasionally run into problems when attempting to change some minor detail. Those who habitually tweak every screw they run into are aware that the old GW- or Q-Basic world was a much simpler place than today s universe of objects, mainly because the barebones syntax was easily memorized. 

The validation umbrella not only covers in-house systems, but also covers vendor systems. Much of industry today is dependent to some extent on vendor-supplied electronic systems and, consequently, the vendor s validation of those systems. Because of this, there are validation issues to assess at the vendor including how the vendor addresses change control, testing and documentation, source code, integration, and implementation of the system during development. Implementation support also must be assessed. 

Cmcial documentation needed in the development phase includes end-user manuals, a unit test summary report, a user acceptance test plan, results of the database design, results and methods employed in the source code evaluations, and a trace-ability analysis. A source code traceability analysis verifies that all code is linked to established specifications and established test procedures. 

The design This includes all the documents and source code associated with the work of satisfying the specification it may be withheld from customers. 

Component software demands a complete separation of interface specifications from implementations. The interface specs, rather than the source code, define what a component will provide and expect when used. In fact, if you are ever forced to use an existing component that does not have an explicitly documented interface, it is usually worth writing a specification of it as perceived and used this document greatly simplifies testing and reduces the time to evaluate the suitability of new and alternative versions of the component. 

One of the problems with inheritance frameworks is that they are white-box in nature The extender of a framework must study and understand the source code in order to make extensions and to understand which methods to override, which other (template) methods that override will affect, and how the set of overridden methods must behave for the framework to function properly. The real reason for this legacy is that the vertical interfaces of class frameworks have rarely been documented explicitly, abstracting away from the code. 

Other deliverable artifacts Documents, source code, tests, executables... 

All of the areas of proof described above provide evidence concerning the current operations of the computer systems in place, but can those same controls be expected to continue to function over time Certainly a trend of control provides some presumption, and annual SOP review procedures provide a degree of assurance, but the most significant evidence of system reliability lies internal to the software and is documented only through a review of that source code itself. 

The requirements for hardware validation are identical to those of any other equipment in use, comprising the OQ/IQ/PQ cycle, except that in the PQ, it is the test of software used. The software validation comprises functional testing, in which defined inputs produce outputs that meet expectations or specifications a thorough examination of source codes, database designs, programming standards, control methods, and support documentation or a quality-assurance program that includes alternate plans, contingency practices, record retrieval, and security practices... 

The objective of this problem is to explore the performance of stiff and nonstiff user-oriented initial-value-problem software. Acquire the Fortran source code and the documentation for Vode from www.netlib.org. The VODE package enables the user to select either stiff or nonstiff methods. 

After acquiring the FORTRAN source code and documentation for Dassl (or other differential-algebraic solver) from www.netlib.org, write a simulation program to solve this problem. 

Inspections are testing practices in which source code and documentation are examined in either a formal or informal manner in order to discover errors, violations of standards, inconsistencies, and other problems. The objectives of these reviews are to verify the following ... 

Code audits are formal and independent reviews of source code by a person, a team, or tool, to verify its compliance with software design documentation and programming standards. In addition, the software code is evaluated against its supporting documentation to ensure that it correctly describes the code. 

Another, independent, person should verily calculations, algorithms, SLC documentation, source code, test script execution, test results, validation plans, and qualification protocols, and so on, as applicable. 

In any case, the software supplier or the developer must follow a documented change management procedure, so that changes to source code are specified and documented. 

Inspection A manual testing technique in which program documents [specifications (requirements, design), source code or user s manuals] are examined in a very formal and disciplined manner to discover any errors, violations of standards or other... 

The second section reviews documentation in relation to CSV and the procedural controls required for regulated operations. Change management and control, training, and security (introducing key technology-driven services such as user/data authentication and access control), are followed by guidance on source code issues, and suppliers qualifications. 

All experiment sequences and Fortran programs described above are available from the authors on 8 inch floppy disks. In addition, source codes and documentation for each program is available. 

The ASPEN system does facilitate the inclusion of a user s own model either in FORTRAN source code or compiled into an object code. The proprietary model may rely on the entire physical property system to calculate the required properties, however, the user routines must have the correct interface calls. These are to be documented in the ASPEN user manual. 

GAMMA is a good example of many of the benefits and compromises listed previously. The bulk of the library was the doctoral work of Dr. Scott Smith at the Eidgenoessische Technische Hochschule (ETH) in Zurich, Switzerland. Subsequently, Dr. Smith took a job at the National High Field Magnetic Laboratory in Florida. Both sites shared in hosting a web site to provide open access to the source code, examples, documentation and compiled binaries for many platforms and compilers. As a code library, many of the obstacles to cross-platform support go away, especially the need to support an application across different graphical user interfaces. 

Because of the nature of InChIKey (and every hash in general), collisions are possible. This fact comes directly from the limited number of possibilities a 25-char-acters-long string can contain. Even though collisions of InChIKeys are inevitable in the future, it is not possible to say when the first collision will occur. The official InChl documentation (documentation published with the InChl source code, version 1.02-Beta http //www.iupac.org/inchi/download/index.html) states that the probability of a collision in a set of 1 billion InChIKeys is 2.0 x 10 20%. However because the second part of the InChIKey is based on InChl layers that do not exist (are empty) for many structures (such as isotopic layer, stereochemistry layer, etc.), a more realistic estimate must be based on collisions in the first part of the InChIKey alone. In this case the same source states that the probability of a collision in a set of 1 billion structures increases to 2.7 x 10-9 %. However, even this means that unless we are extremely unlucky, InChIKey should remain unique for quite a long time. It... 

A user may incorporate these models into custom written software by including the source code in the main program and programming the proper calls to the Handbook routines. The subroutines are extensively commented, and detailed documentation for the input arguments, output arguments, and detailed calling procedures are provided in the beginning of each subroutine. The routines are compatible with versions of the 1977 ANSI FORTRAN standard. 

---