: 1. It does not make any user-facing changes in 1.3 -- no configuration is
: touched at all. This was one of the major support issues.

But it does affect users ... it opens them up to start seeing plugin 
developers using new/different syntaxes they aren't familiar with -- 
syntaxes that will definitely confuse people and increase the number of 
questions about how to configure things.

: configuration post 1.3 but we should not force plugin developers to use an
: ugly format for the next 1 year (till spring or whatever comes along) if
: something can be done about it so easily.

As I said in the email that spawned that Jira issue -- Plugin writers are 
free to load their own custom config files containing more complicated 
information -- those files can be any syntax they want, and by being 
seperate files it will not suprise people that they are a different 
syntax, and it will be clear what they should ask questions about: "I'm 
looking at the FooRequestHandler config file, if i don't understnad it i 
should ask about the FooRequestHandler config file syntax"  when you start 
allowing drasticly differnet syntaxes for different instances 
(FooRequestHandler vs 
BarRequestHandler) of the type (RequestHandler) directly in solrconfig.xml 
then users don't really know what to ask about ... the expectation is that 
within the solrconfig.xml the syntax for configuring RequestHandler X is 
the same as for RequestHandler Y ... the names/values of the options may 
be differnet, but the syntax should be the same

: I wasn't sure if a -0 is the same as a -1 but I'll roll this back if I made
: a mistake in handling this in the correct fashion.

http://www.apache.org/foundation/voting.html

You essentially had two people saying "yes"; two people saying "I won't 
get in the way, but I'd rather we didn't do this"; and one guy (me) who 
didn't specificly "vote" but did express a lot of concern that this will 
introduce confusion ... that doesn't really sound like a lot of support.

If I'd been in your shoes, I wouldn't have commited.  Anytime you have 
multiple committers who are concerned about the impacts of something, it's 
probably wise to discuss some more.



-Hoss

Reply via email to