[sldev] Re: [VWR] Web login without llmozlib

Lawson English lenglish5 at cox.net
Wed Dec 26 09:32:21 PST 2007


Argent Stonecutter wrote:
> [...]
> The message after the status code is routinely delivered to end-users, 
> by all kinds of software, when it's the best available response (for 
> example, a fatal error with no body). The only software that I have 
> run into that goes to great lengths to hide this information is 
> Internet Explorer... and any time IE does something different than the 
> rest of the world the odds are pretty overwhelming that it's not the 
> rest of the world that's messed up (this is not to say that anything 
> by Microsoft is suspect, just that IE seems to have become Microsoft's 
> resident foulup fairy).
>

Actually, anything Microsoft does IS suspect. They may have improved in 
recent years, but it really WAS the unofficial motto of the DOS division 
of MS that "DOS ain't done 'til Lotus won't run."

The OS division at Apple found that the main reason they couldn't 
upgrade the Mac OS past a certain point was because it would break 
certain "important word processing software," no matter how they did it 
(It wasn't purchasing NeXT that allowed Apple to bring out a new OS, it 
was falling RAM prices that allowed them to run two OS's simultaneously 
so they could support legacy MS software without interfering with the 
features of the new OS).

Even MS's own software wasn't immune to MS's interesting policies 
concerning how software was written. According to insiders who revealed 
MS's so-called "best software practices" after their non-disclosure 
agreement had lapsed, MS would allow any application programmer access 
to the source code of hte OS and, if knowledge of the value of a 
variable was required, they had two  choices: they could request that 
the variable be exposed via a function call OR they could do what was 
done in at least one case: examine the call frames of the internals of 
the text editing routines, implement a callback function, and walk back 
the required number bytes through the call stack to grab that variable 
"the hard way." This latter approach, of course, required an upgrade of 
the application software with each new OS release or patch or 
potentially for every new processor....


...hmmm guess MS's practices haven't changed, eh?



Lawson






More information about the SLDev mailing list