Internet Engineering. Jan Nikodem, Ph.D. Software Engineering

Transcription

1 Internet Engineering Jan Nikodem, Ph.D. Software Engineering

2 Software crisis Software Engineering Lecture 2

3 The term "software crisis" was coined at the first NATO Software Engineering Conference in 1968 by: Peter Naur Nationality;Dutch, astronomer, Niels Bohr Institute, Technical University of Denmark, University of Copenhagen. Peter Naur /25

4 The term "software crisis" was coined at the first NATO Software Engineering Conference in 1968 by: Friedrich. L. Bauer Nationality;German, mathematician, theoretical physicist, Technical University of Munich Friedrich L. Bauer /25

5 The major cause of the software crisis is that the machines have become several orders of magnitude more powerful! The Humble Programmer (EWD340), Communications of the ACM(1972). Nationality; Dutch, theoretical physicist, mathematician Eindhoven University of Technology Edsger Wybe Dijkstra /25

6 To put it quite bluntly: as long as there were no machines, programming was no problem at all; when we had a few weak computers, programming became a mild problem, and now we have gigantic computers and programming has become an equally gigantic problem. Nationality; Dutch, Eindhoven University of Technology Edsger Wybe Dijkstra /25

7 Brooks's management adage (1) Adding manpower to a late software project makes it later. The Mythical Man-Month (1975). Nationality; USA, applied mathematician IBM, University of North Carolina at Chapel Hill Copyright owned by SD&M ( Frederic P. Brooks 1931 Copyright owned by SD&M ( 7/25

8 Brooks's management adage (2) The nine-women adage: "Nine women can not make a baby in one month". Nationality; USA, applied mathematician IBM, University of North Carolina at Chapel Hill Copyright owned by SD&M ( Frederic P. Brooks /25

9 Corollary of Brooks's adage (1) There is an incremental person who, when added to a project, makes it take more, not less time. The Mythical Man-Month (1975). Nationality; USA, applied mathematician IBM, University of North Carolina at Chapel Hill Copyright owned by SD&M ( Frederic P. Brooks 1931

10 Corollary of Brooks's adage (2) 1. It takes some time (ramp up) for the people added to a project to become productive. 2. The number of different communication channels increases with the square of the number of people Nationality; USA, applied mathematician IBM, University of North Carolina at Chapel Hill Copyright owned by SD&M ( Frederic P. Brooks /25

11 Brooks s conclusion There is no Silver Bullet ( the final result of looking for the software crisis solution) 11/25

12 Essence and Accidents of Software Engineering Brooks use Aristotelian language to separate two kinds of problems in software engineering. 1. Essence is the difficulty inherent in the structure of the problem. 2. Accident includes difficulties that in any particular setting go along with the production of software, or mistakes that happen but are not inherent to the Copyright owned by SD&M ( particular task. 12/25

13 Essence and Accidents of SE 1. Essential properties are those properties that a thing must have to be that thing: A car must have an engine, wheels, and a transmission in order to be a car. 2. The accidental properties arise by happenstance and do not affect the basic car-ness of the car. A studded snow tires or racing slicks, an automatic or a manual transmission. Steve McConnell, From the Editor IEEE Software, Vol. 16, No. 2, /25

14 No Silver Bullet Essence and Accidents of Software Engineering 1.Fashioning complex conceptual constructs is the essence. 2. Accidental tasks arise in representing the constructs in language. Past progress has so reduced the accidental tasks that future progress now depends upon addressing Copyright owned by SD&M ( the essence. 14/25

15 Standish Group CHAOS report (1) 31% of IT projects canceled before completion. Average IT project cost is 189% of original estimates. Average timeoverrun is 222% of original estimates. 15/25

16 Standish Group CHAOS report (2) 88% of all IT projects are over schedule, over budget, or both. On average, only 61% of originallyspecified featuresare implemented. 16/25

17 Standish Group CHAOS report (3) 5% of software projects are terminated before they produce anything. 66% are considered to have failed. Of those that do complete the average cost blowout is 43% 17/25

18 Standish Group reflection on SE (1) The fact of the matter is: developers commit to unrealistic estimates of effort, cost and schedule. 18/25

19 Standish Group supposition (1) Perhaps cost estimates supported a bid-towin strategy were unrelated to the actual amount of work to be done. Perhaps the initial schedule was driven by business requirements alone and did not consider the product size/functionality 19/25

20 Standish Group supposition (2) Perhaps there was no real awareness of the developer s process productivity. Perhaps there was no reality check performed upfront to see if what had been proposed and accepted was actually feasible. 20/25

21 Standish Group supposition (3) Perhaps, there was no motivation to develop a realistic cost and schedule baseline. Perhaps, but in the meantime. 21/25

22 Standish Group reality The lost dollar value for USA projects in 2002 is estimated at US $38bn with another US $17bn in cost overruns. 22/25

23 There are two issuesat the heart of the estimating the challenge : 1) the need to understand and express (as early as possible) the software problem domain and 2) the need to understand our capability to deliver the required software solution within a specified environment. Then, and only then, we can accurately predict the effort required to deliver the product. (Garmus, Principles of Sizing and Estimating Using (IFPUG) Function Points 23/25

24 Readings W. Wayt Gibbs, Software's Chronic Crisis,Scientific American, 1994 EdsgerW. Dijkstra, EWD340: the humble programmer. Communications of the ACM, 10, 24/25

25 Useful links E.W.Dijkstra Archive: Home page Fred Brooks -Wikipedia, the free encyclopedia Welcome to The Standish Group International, Inc. 25/25

26