incubator-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Kevan Miller <>
Subject Re: [VOTE] publish Tika 0.1-incubating
Date Mon, 07 Jan 2008 14:52:53 GMT

On Jan 3, 2008, at 11:09 PM, Jukka Zitting wrote:

> Hi,
> On Jan 4, 2008 5:31 AM, Niclas Hedhman <> wrote:
>> +1 for this release, and note to developers to have a DISCLAIMER file
>> separately in both source release and the META-INF dir of the binary.
> Will do.
> On a related note, when and how did the DISCLAIMER file come to exist?
> I guess it's a legacy of Wicket, since that was the first time I
> recall seeing the file. To me it seems debatable whether the
> disclaimer is more prominent within the README file or in a separate
> DISCLAIMER file, but I'm OK to go with the latter if people prefer
> that.
>> I am also curious to know how Andy's and Ryan's contributions "came  
>> to be". Do
>> we have any paper trail of this?
> The notice for Andy Clark was added for the NekoHTML [1] dependency
> (ALv2). The Tika codebase doesn't actually contain any of that code,
> just uses it as a dependency, so AFAIK the attribution in NOTICE
> wouldn't actually be needed. It got added [2] a while ago when we were
> considering a release bundle that contains all the dependencies. The
> notice was briefly questioned [3] and discussed [4] on tika-dev@ but
> so far no action was taken on it.
> The notice for Ryan Ackley is more substantial, as the Tika codebase
> contains a few classes (part of [5]) from his now-dormant
> [6] codebase. Those classes found their way into Tika
> via the Nutch and Lius Lite codebases and their status was considered
> on tika-dev@ with Ryan on Cc [7].

IMO, the attribution for Ryan Ackley in the NOTICE file should include  
the copyright for those source files (i.e. "Copyright 2004 Ryan  
Ackley"). Could also ask for permission to remove the copyright from  
those source files or move them to the NOTICE file as described in

Everything else looked good...


  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message