Re: [wildfly-dev] Search Index: Call for help
That's a good point Scott.
The search is limited by the quality of the data inout we use for indexing. Currently the
index is being created from the model descriptions and, the keywords. Although matching
keywords have a higher priority when queries are executed.
The keywords are the cheap win for us. But in the long run the underlying model
descriptions should be improved as well. These are often ambiguous or not very descriptive
at all.
/Heiko
On 13 Nov 2014, at 16:02, Scott Marlow <smarlow(a)redhat.com> wrote:
For the jpa-metrics token, I propose that we change the description
to reflect that these are metrics. Before updating the spreadsheet, is there any reason
why the description for jpa-metrics should be "The configuration of the JPA
subsystem.;"
We can probably change all metrics tokens as well if we agree. Perhaps to something like
"The JPA metrics.;
On 11/13/2014 08:36 AM, Heiko Braun wrote:
>
> We are looking for subject matter experts who help to improve the search
> in the console by providing reasonable keywords.
>
> _It takes you 2 minutes of your time_.
>
> Here's how you can help:
>
> 1) Open the spreadsheet[1].
> /(The "token" represent a place in the user interface. /
> /The "resources" are the management model resources referenced by those
> places.)/
>
> 2) Pick one or two rows that are appealing to you
> /(i.e. a subsystem you know, a component you contributed to)/
>
> 3) Provide 5 keywords that relate to the resources referenced by that token
> /(i.e. "token=modcluster" => keywords={"load-balancing",
> "reverse-proxy", "clustering"})/
>
>
> Please put the keywords right into the spreadsheet.
>
> Thanks for your help, Heiko
>
>
> [1]
>
https://docs.google.com/spreadsheets/d/12T-rty5AfrsxeT7FIUjcyHC1j4G_VHWBA...
>
>
> _______________________________________________
> wildfly-dev mailing list
> wildfly-dev(a)lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/wildfly-dev
>