IMPORTANT, MUST BE READ... : Boehm's Top ten Software Defect Reduction list
Title : Boehm's Top ten Software Defect Reduction list
Boehm's Top ten Software Defect Reduction list
IMPORTANT, MUST BE READ... A spell back, Barry Boehm & Vic Basili wrote a prissy summary of best ways to larn improve character software. Their advice nevertheless applies to embedded systems today. Below is their listing (in bold) amongst my commentary (the parts non inward bold).1. Finding too fixing a software job later delivery is oftentimes 100 times to a greater extent than expensive than finding too fixing it during the requirements too pattern phase
Bug ready terms gets worse every bit your software gets closer to deployment, because y'all convey to non exclusively pass a lot of fourth dimension tracking downward the source of the bug, but every bit good retest the organisation later the fix. It is mutual to discovery situations inward which a põrnikas "escape" into plain units costs dramatically to a greater extent than than 100x. Consider having to scream upward a fleet of cars to install a põrnikas fix, or do maintenance visits to thousands of sites to manually install a fix. (Over-the-air fixes convey their ain problems, but that's a topic for about other time.)
2. Current software projects pass most xl to 50 percentage of their endeavour on avoidable rework.
As the joke goes, the start 90% of the projection is spent on design, too the minute 90% on debugging. The cheapest way to debug is to avoid putting the bugs into the software inward the start place. The side yesteryear side best way is to discovery them at the betoken of introduction (e.g., peer review of pattern earlier code is written) rather than at organisation test.
3. About eighty percentage of avoidable rework comes from twenty percentage of the defects
If y'all convey a "bug farm" that's oftentimes non because the code is bad, but rather because the underlying requirements too pattern are bad. If ane module has a lot of bugs y'all should rewrite the module rather than perish on patching it. However, if during the rewrite y'all mightiness good discovery that the job isn't actually the code, but rather a bad pattern or unclear requirements. Writing novel code for a bad pattern ultimately won't solve the problem.
4. About eighty percentage of the defects come upward from twenty percentage of the modules, too most one-half the modules are defect free.
In add-on to comments for #3 above, modules amongst high cyclomatic complexity tend to last hard to attempt too tend to last to a greater extent than bug-prone. Keeping a confine on complexity tin help amongst this problem.
5. About ninety percentage of the downtime comes from, at most, 10 percentage of the defects.
It makes sense to weight testing on the areas that are the highest risk. For desktop software this oftentimes corresponds to the mutual occupation cases. For embedded systems too other mission-critical systems this every bit good way testing failure detection too recovery for high-cost failures.
6. Peer reviews grab sixty percentage of the defects.
Yes, really. Peer reviews grab the bulk of defects. Why aren't y'all doing them? (If y'all are doing them, are they catching at to the lowest degree one-half the defects?)
7. Perspective-based reviews grab 35 percentage to a greater extent than defects than nondirected reviews.
When y'all do reviews, give each reviewer a checklist amongst a unlike laid of areas to think most or unlike piece of work to play. For example, command flow, information flow, exception handling, testability, coding style.
8. Disciplined personal practices tin trim defect introduction rates yesteryear upward to 75 percent.
As much fun every bit it is to last a coding cowboy, on average fifty-fifty the best programmer volition innovate fewer bugs yesteryear next a methodical technology scientific discipline exercise rather than slinging code. As mentioned above, the cheapest bugs to ready are the ones that never made it into the code. Beyond this, at that spot are practices such every bit PSP too TSP that are shown to dramatically improve character without actually changing productivity.
9. All other things beingness equal, it costs 50 percentage to a greater extent than per source pedagogy to prepare high-dependability software products than to prepare low-dependability software products. However, the investment is to a greater extent than than worth it if the projection involves pregnant operations too maintenance costs.
In other words, if a production scream upward puts your companionship out of business, it's worth investing inward practiced software character upward front. In my sense if y'all are already transportation a mission-critical production you're already spending that 50 percentage to a greater extent than (and too then some), but nevertheless transportation defects. This isn't maxim pass fifty-fifty more. Rather, doing peer reviews too about other basic software character practices y'all tin improve character at the same terms you're already spending. Testing software into submission is only non the way to go.
10. About xl to 50 percentage of user programs comprise nontrivial defects.
If y'all convey programmable features, your customers volition convey bugs inward what they do. And don't forget that your fiscal management spreadsheets are user programs (i.e., can, too oftentimes do convey bugs).
Items #1 - #10 from:
Boehm & Basili, Software Defect Reduction Top 10 List, IEEE Computer, Jan. 2001, pp. 135-137.
You tin read the master three-page article here:
https://www.cs.umd.edu/projects/SoftEng/ESEG/papers/82.78.pdf
IMPORTANT, MUST BE READ...
Thank for your attention Boehm's Top ten Software Defect Reduction list
my blog Boehm's Top ten Software Defect Reduction list, Have a nice day.
Now you read article Boehm's Top ten Software Defect Reduction list this permalink article is https://fairemirima.blogspot.com/2017/11/boehms-top-ten-software-defect.html Thank you and Best regards. You Can read nice Tips below. It was always better to choose topics that interest you or in wich you at least have some knowledge about . When creating targeted internet copywriting , you have to stick with your strong points , or everyone will know it . Make a list of all of the things and or topics that you are interested in . . . How much do you know ? Can you tell it as a story ? That is The essence of writing for the web . You Have to know your subject well , or nobody will believe you it is always better to impress someone then upset them . When Writing Targeted Internet Copywriting , you have to choose your appropriate target group of customers . without a target group of customers , you could ramble on incessantly about random subjects for days on end with no essence of a final goal . You always have to keep in mind who your customers are and what they are looking for . . . . . . . . . IMPORTANT, MUST BE READ...
0 Response to "Boehm's Top ten Software Defect Reduction list"
Post a Comment