Mailing-List: contact dev-help@ctakes.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@ctakes.apache.org
MIME-Version: 1.0
In-Reply-To: <CE708D32.4D78%robert.spurrier@explorys.com>
References: <CE708D32.4D78%robert.spurrier@explorys.com>
Date: Tue, 1 Oct 2013 14:53:58 -0400
Message-ID: 
 <CAPqz87qMyPVBPp7GTRbeboPBJ4WksLo+gfwQmz3wGMZnfzhnDA@mail.gmail.com>
Subject: Re: Converting cTAKES to UIMAfit
From: Pei Chen <chenpei@apache.org>
To: dev@ctakes.apache.org
Content-Type: multipart/alternative; boundary=047d7bd756e2a9231c04e7b2793b

--047d7bd756e2a9231c04e7b2793b
Content-Type: text/plain; charset=ISO-8859-1

Agreed,
There have been few good places where uimaFIT may be useful:
Feel free to check out any existing items and add/create items:
https://issues.apache.org/jira/browse/CTAKES-135?jql=project%20%3D%20CTAKES%20AND%20text%20~%20uimafit

On the top of my head:
1) Type System detection
2) JCasUtil (selectCovered, etc.)
3) Input Parameter handling
4) Annotator, Pipeline instantiation/creation

Note: There may be a valid use case to have a few default XML
configurations (mainly in production use where you may want to swap configs
without having to recompile code.)


On Tue, Oct 1, 2013 at 2:31 PM, Robert Spurrier <
robert.spurrier@explorys.com> wrote:

> Hello,
>
> Working with UIMAfit makes annotator/pipeline development a sheer joy. I
> thouroughly dislike keeping track of annotator descriptors  during
> development.
>
> Is it worth converting cTAKES to use UIMAfit as exclusively as possible? I
> would not mind taking on that responsibility if It meant blowing away as
> many descriptors as possible =]
>
> Thanks,
> Rob
>

--047d7bd756e2a9231c04e7b2793b--