[CSUSB]>> [CNS]>> [CSE]>> [R J Botting]>> biba.php

Bibliographic Item (1.0)


  1. Roland Racko
  2. Hedging your Bets(Parts I and II)
  3. Software Development Magazine(Jul 1995)pp75-77 & (Aug 1995)pp73-76 & (Sep 1995)pp79-84

    Viri that infect the software process and may cause a problem when implementing OO methods. questionaire handed out to 40 reps at Software Development 1995(Feb 1995).

    You only know something is reusable when it has been reused 3 times or more.

    Bad object: no good map to user domain, difficult argument lists, instance data, fragmented or oversized objects.

  5. OO_process_virusses::=following,
    1. No way to measure cost of maintaining code. ->- will not be able to measure main benefit of OO. OO costs more when creating and saves when created.
    2. Focus on rapid delivery rather than other qualities. ->- forget reuse and bad inheritance structures, These take time to think out.
    3. low user rapport: specs on coffe can lids ->- loss of reuse because business domain objects and ruleslow knowledge of are not visible on a coffee can lid.
    4. Low knowledge of SD: coupling & cohesion ->- easy to make bad objects and lose reuse
    5. No tracking of source of bugs or system features. ->- end up blaming method
    6. No system for naming things ->- harder to find and reuse objects
    7. An addiction to writing tight code ->- it takes several versions to get a reusable version, then optimize that.
    8. Managers shorten all project estimates - manipulation, lack of trust, lack of respect, loss of two way communication->- reuse lost to meet time pressure.
    9. Bugs fixed in unaid and unrecorded overtime (25% of them).->- bugs go away and all else goes to the wall: reuse, maintenance, design,...inheritnce.
    10. No time to assimilate new skills -> bad objects.
    11. Changing two objects and another change as well.

    In part III claims the average shop does not have all problems... except in source code control. The most occurring problems being in the hasty scheduling and overload areas(2,8,9 above)

Search for bibliographic items containing a matching string.

(Search uses POSIX regular expressions and ignores case)

Search for a specific bibliographic item by name.

To see the complete bibliography (1Mb+) select:[Bibliography]