tomcat-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Peter V. Gadjokov" <>
Subject RE: Case Sensitivity in URLs (was RE: BugRat Report #92 was close d(apparently by: Craig R.)
Date Tue, 12 Sep 2000 03:50:28 GMT
>Not just my opinion. Putting the onus on me to prove why it's a 
>hack doesn't chage the fact that it's a hack. 

You're the one calling it a hack, the onus is indeed on you. It is 
reasonable, in technical communication, to expect participants to 
justify their statements, whether these are purely their opnion or 
a combination of their and other's opinions. And I think it's also 
perfectly reasonable for me to say 'your view it's a hack is no 
reason not to implement it'. I happen to think Servlets and HTTP 
itself hacks. I still find lots of use for them and the packages 
that have undertaken the laborious task of implementing them. 
"It's a hack!", whether true or not, whether universally 
acknowledged or not, is almost entirely irrelevant in this case 
but I'm happy to hear your views _just in case_ they contain 
something that concludes the discussion of the feature outright. 
So telling me it's a hack once again is not moving us forward. 

>Fact is if you "get your way" with issue it's going to cause 
>porblems for me later when I have to port your ".war" to UNIX, 
>Netware or AS400, all of which run Tomcat, btw. 

That's not neccessarily the case and a different conversation. I'm 
not trying to 'get my way', I'm trying to discuss the inclusion of 
what I believe is a useful, neccessary feature. We're still debating 
whether such a feature _should_ be included. Let's suppose that we 
agree we should consider it and move on to discussing its possible 
implementation. Perhaps a requirement for any implementation will be 
'the feature does not cause problems for Nick Bauman when porting 
Peter Gadjokov's WARs to AS/400' and perhaps it turns out it's 
impossible to meet such a requirement and we drop the feature or 
change the requirement. But all of that is, as I said, is a separate 
conversation. Let's not mix up design and implementation and
philosophical reasons to dismiss the feature and practical, 
implementation-specific reasons not to implement it. 

P.S. At the risk of being an even more annoying pedant - MacOS X and 
Darwin running on HFS+ _are_ Unix. 



View raw message