I think it's easy to assume MS should have run a test for leap years, perhaps they only checked to see if dates functioned properly. My contention isn't that MS shouldn't have done more, but that most companies don't do more. David's example is evident that Apple doesn't check everything. Checking just the sheer number of security updates Apple applies to OS X over time it's clear they don't check everything, and it's my belief they can't. Or the product would never be released.
The iphone 2.1 update was mainly pushed out as solely a bug fix for the problems in 2.0. What is the difference? Obviously Apple doesn't check it's OWN code even under Tom's standards. On Mon, Jan 5, 2009 at 8:55 AM, David K Watson <[email protected]>wrote: > This is just plain bizarre. > > It is plainly obvious that no one is going to "check every line of > code in every third party component". Such a task would be > almost as difficult as originating the code in the first place. That > question is purely rhetorical and as such needs no answer. > > There is, however, such a thing as acting with due diligence to > make sure that those third-party components actually work for > you in the way you envision. Not checking every line of code, > but certainly running a test suite to uncover any potential problems. > In the case of a essential calendaring component, it would be > reasonable to assume that this would include checking various > critical dates, like Dec 31, Jan 1 for every year and Feb 29, Mar 1 > for leap years. Heck, it would be a fairly easy test to automate, > checking every single day well beyond the expected life of the > product would not be unreasonable. Plainly, no one at Microsoft > made sure that this test was properly done--not terribly surprising > because we already had indications that Zune 1.0 was a rush job. > > Now, there is room to argue as to whether or not Apple is this > sloppy as often as Microsoft seems to be. I tend to think not, but > you can argue otherwise with examples like the most recent software > update, which in a few cases did not download fully but tried to > execute anyway. > > On Jan 5, 2009, at 12:00 AM, COMPUTERGUYS-L automatic digest system wrote: > > Subject: Re: Zunepocalypse >> >> Statements that include terms like "every," "all," "none," "always," or >>> "never" usually have the answer built into the question. Your >>> ridiculous, stacked-deck question is unworthy. >>> >> >> As Mike notes, you have finally answered the question, rather >> circuitously, >> with "No, Apple does not check every line of code in every third-party >> component." If you admit (finally) that no one checks -every- line of >> code >> in -every- component, then you cannot classify MS as incompetent because >> it >> failed to check -this- one line of code in -this- one component. >> >> There simply is no logic in your position that no one completely validates >> every component, yet MS is clueless because it failed to do exactly that. >> >> > > ************************************************************************* > > ** List info, subscription management, list rules, archives, privacy ** > ** policy, calmness, a member map, and more at http://www.cguys.org/ ** > ************************************************************************* > -- Make sure you support your local CarbonONset programs! ************************************************************************* ** List info, subscription management, list rules, archives, privacy ** ** policy, calmness, a member map, and more at http://www.cguys.org/ ** *************************************************************************
