Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!ucbvax!bloom-beacon!CALSTATE.BITNET!PAAAAAR From: PAAAAAR@CALSTATE.BITNET Newsgroups: comp.software-eng Subject: Re: Congress Finds Bugs in the Software Message-ID: <8912010709.AA10788@mwunix.mitre.org> Date: 1 Dec 89 07:02:24 GMT Sender: root@athena.mit.edu (Wizard A. Root) Organization: The MITRE Corp., Washington, D.C. Lines: 63 dbenson@cs2.cs.wsu.edu (David B. Benson) Quoted >[AAAS Science, 10 November 1989, vol. 246, p. 753] >[by M. Mitchell Waldrop] >Dig into any of the government's chronically over-budget and >behind-schedule development programs -- the Hubble Space Telescope of >the B1-B Bomber, for example -- and you'll find that a good fraction of >what gets labeled as "waste, fraud, and abuse" actually stems from >crummy software. Not only do the development agencies habitually spend >millions of dollars on operations software that is buggy, inadequate, >and late, they have to spend millions of dollars more to fix it. [...] >So, what can be done?[...] Back in England in the 70' the UK Government must have had a similar problem because it set up projects to select standard testing and modular programming Guidelines for the civil service. It selected a data directed design method (JSP) and added All-Path-Testing. Quality assurance techniques were formulated. Special courses were developed for managers at all levels upward from the just promoted programmer. It went thru many name changes. The last time I heard it, was called SDM. I learned it, taught it and still use it. 6 months of research were needed to work out the theory behind it. In the 1980's Britain's cost cutting Prime Minister (you know who) asked why software was still too expensive,... the Civil Service answered that the problem was that the program specifications were less than perfect. The service investigated all known Structured Analysis and Design and Project Management methods....and selected SSADM.... And started to watch for what is now called CASE products. SSADM is also data directed but involves data flows, data engineering, dynamic analysis and design, etc,... However the specifications were informal. So I guess that problems of verification, completeness and ambiguity remain. The Ministry of Defense now requires a formal specification for "critical" software. >So, what can be done? In the near term, very little, says the report. Design, quality control, specifications, and analysis can all be improved in the USA but much USA government software is done outside so the UK process can not apply. Possibly the first step would be real competition... why not two companies PRODUCING the software, and only the BEST getting paid for it?...only joking of course. ------------------------------------------------------------------------- Richard J. Botting, Department of computer science, California State University, San Bernardino, 5500 State University Parkway, San Bernardino CA 92407 PAAAAAR@CCS.CSUSCC.CALSTATE paaaaar@calstate.bitnet PAAAAAR%CALSTATE.BITNET@CUNYVM.CUNY.EDU --------------------------------------------------------------------------- Disclaimer: The above are my own opinions and may or may not reflect the past or present policies of Her Majesty's Government --------------------------------------------------------------------------- Brought to you by Super Global Mega Corp .com