Add the initial version of the Polish stemmer #29
Conversation
| @@ -0,0 +1,21 @@ | |||
| MIT License | |||
There was a problem hiding this comment.
Nothing seems to actually reference this file, and the PR against the snowball repo doesn't mention this licence.
The Snowball compiler and other algorithms are all released under the 3-clause BSD licence, which is also a permissive licence and really pretty similar in practical terms (it does have an additional "no endorsements" clause, but that shouldn't make a difference to any reasonable use).
Are you happy to just license under 3-clause BSD? I'm not keen to have licence proliferation within the Snowball distribution.
There was a problem hiding this comment.
@tomek-ai @tomeklysiak Please respond to this part at least - can we drop this file (i.e. algorithms/polish/LICENSE)?
The PR to add the stemmer to the snowball repo didn't include it, so the stemmer code seems to be implicitly being contributed under the licence of the snowball repo, which is 3-clause BSD.
So it seems this is just meant to apply to the algorithm description in this PR. Whatever is meant, I really don't want to have a proliferation of licences in the project (and it would be unhelpful to have to think about licensing if, for example, we later wanted to reuse parts of the algorithm description as code comments).
It would be useful to respond to my other question too, but that one isn't really a blocker to merging.
There was a problem hiding this comment.
@ojwb @mitya57 The PR was created more than three years ago, but from what I remember, the Polish stemmer logic originates from an MIT-licensed Python project. The implementation in Snowball was created independently, using the same principles but rewritten from scratch and adapted to Snowball’s structure. The MIT license was included as an acknowledgment of the original inspiration.
There was a problem hiding this comment.
Thanks very much for clarifying.
That's all fine, I'll adjust the documentation to credit the creator and link to the original repo. We should be able to get this merged.
|
|
||
| [% algorithm_vocab([20, 'babicka', 'maaru']) %] | ||
|
|
||
| <h2>The stemming algorithm</h2> |
There was a problem hiding this comment.
What's the origin of the algorithm?
If there's a paper or a description of it somewhere, it's helpful to reference it.
If it's just something you devised, that's OK but it's useful to say that (and really you should get credit for creating it).
No description provided.