>> On May 26, 1:03?am, John Larkin
>>
>>
>>
>>
>>
>> wrote:
>> > >> > >> >>
>> > >>> John Lark>> > >>>> >>
>> > >>>>> jmfbahciv writes:
>>
>> > >>>>>> Patrick Scheible wrote:
>> > >>>>>>> The monitor and the interface are the same for late versions of
>> > >>>>>>> Windows. ?They are released together, sold together, and used
together.
> > >>>>>> So what? ?It's still an app. ?Doesn't anybody realize that keeping
>> > >>>>>> this distinction is important in design?
>> > >>>>> Lots of people realize the distinction is important. ?Unfortunately,
>> > >>>>> none of them were making design decisions for late versions of
>> > >>>>> Windows.
>>
>> > >>>>> Peaked roofs are an important design feature for buildings. ?Buildings
>> > >>>>> with flat roofs always leak sooner. ?But I'm not going to look at a
>> > >>>>> flat-roofed building and say it has a peaked roof.
>>
>> > >>>>>>> Oh, Windows has so *many* bugs, I'd hate to ascribe all of them to
>> > >>>>>>> just *one* of its design flaws...
>> > >>>>>> I can safely state that it's the _source_ of most of the bugs. ?Think
>> > >>>>>> about the mindset of the developers. ?It is more important to keep
>> > >>>>>> the monitor running. ?If the app kills the exec code, you have
>> > >>>>>> security risks up the ying-yang. ?If the app is separated from
>> > >>>>>> the exec code via UUOs, it cannot destroy the core image of the
>> > >>>>>> exec.
>>
>> > >>>>>> Then you can keep the app bugs separated from the monitor bugs.
>> > >>>>>> The app may die but the integrity of the rest of the system
>> > >>>>>> (think file structure and networks) stays steady.
>> > >>>>> Windows NT and later does try to keep application crashes from
>> > >>>>> bringing down the whole OS. ?There is a barrier between user code and
>> > >>>>> the monitor. ?It's just that the GUI is on the monitor side of the
>> > >>>>> barrier.
>>
>> > >>>> It wasn't originally designed that way. Cutler was adamant to make it
>> > >>>> a "small kernel" OS with all the dangerous stuff in protected space.
>> > >>>> The first version of NT was like that. His enemies inside Microsoft
>> > >>>> then moved in and broke it.
>>
>> > >>>>formatting link
>>
>> > >>> It had nothing to do with "enemies". ?It had to do with a mindset and
>> > >>> not having a proper development infrastructure in place before mass
>> > >>> shipments.
>>
>> > >>> /BAH
>>
>> > >> At one stage in the NT development there were "pipelines of piss
>> > >> flowing" between competing groups.
>>
>> > >> Good book: read it.
>>
>> > >Did the book mention the work that JMF and his group were doing
>> > >w.r.t. NT developement? ?Does it mention the work that MS
>> > >contracted DEC to do w.r.t. NT development?
>>
>> > >/BAH
>>
>> > Don't recall... I haven't read the book in a while. I'm intoTrollope
>> > lately, and some spectroscopy stuff.
>>
>> > John- Hide quoted text -
>>
>> > - Show quoted text -
>>
>> Are you are that there is a Trollope group on yahoogroups?
>>
>> Bruce B. Reynolds, Trailing Edge Technologies, Glenside PA- Hide quoted text -
>>
>> - Show quoted text -
>
>That prior posting should read :"Are you aware that..."
>
>Bruce B. Reynolds, Trailing Edge Technologies, Glenside PA
YFHPFH (You Fixed His Post For Him)