lucene-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mark Miller (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (SOLR-3619) Rename 'example' dir to 'server'
Date Sat, 14 Jul 2012 23:10:34 GMT

    [ https://issues.apache.org/jira/browse/SOLR-3619?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13414534#comment-13414534
] 

Mark Miller commented on SOLR-3619:
-----------------------------------

bq. So in calling it example (or template or skeleton or whaterver) gives people a hint that
it's not anything that they should expect to be sufficient for their need without some more
tuning (Solr is not GSA..)

I think that is exactly what you want to avoid - the feeling that you cannot get anywhere
without tuning. You should have a relatively good experience without tuning when you first
start - that's how people like software to work - really easy to get into, and then really
easy to tune as you need. That comes down to documentation.

I really don't think it helps to try and club someone up front and say, no, you must figure
it out all up front. That is when someone looks for something easier. Everything should work
the best it can without tuning. Then good documentation and crazy customizability work in
your favor. Lots of people don't clean up all the stuff in schema.xml and make their own.
They add a few fields. The remove a few things. Some pare it down to just what they need.
If initial building and testing you do works with the example schema and few or no tweaks,
great! You can pare it down later when you have learned more over time.

I think that's how most people try and come at things anyhow - try and get the existing stuff
up and improve and tweak as they learn more. We should make that easier and more intuitive.
There is no reason Solr should not be as good as GSA out of the box (though Solr not having
crawlers makes that an odd comparison). Easy out of the box and super configurable/tunable
is way better than hard out of the box and super configurable/tunable.

I also think you can do way more with Solr out of the box than you might think! Either using
dynamic field defs or adding just a field or three. Have you ever seen Hossmans out of the
box talks? They are fantastic. I think enough is configured right out of the box that there
are a variety of paths you can take to really quick cool setups.

For people that can get away with just a few teaks to what we ship, should we really be trying
to send the message that you really need to treat all these huge xml files as an example,
and actually, you need to figure out how to create new ones for your use case?

I think that rather than trying to put the burden on the user right away (don't think this
is gonna be easy buster!), we should be making the initial experience liquid smooth and as
hassle free as possible and maintain all of the options and configurability so that as they
learn Solr they are able to take it in whatever direction they need.
                
> Rename 'example' dir to 'server'
> --------------------------------
>
>                 Key: SOLR-3619
>                 URL: https://issues.apache.org/jira/browse/SOLR-3619
>             Project: Solr
>          Issue Type: Improvement
>            Reporter: Mark Miller
>            Assignee: Mark Miller
>             Fix For: 4.0, 5.0
>
>         Attachments: SOLR-3619.patch, server-name-layout.png
>
>


--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org


Mime
View raw message