[chemfp] new licensing option in chemfp 3.4

[Andrew Dalke]

Hi all,

The upcoming release of chemfp 3.4 has a new licensing option. I will distribute pre-compiled wheels for "manylinux" 2010 and 2014 (see https://github.com/pypa/manylinux ). These should work on most Linux-based OSes.

If you are interested in trying it out, do:

  python -m pip install chemfp -i https://chemfp.com/packages/

Let mw know if you want a more in-depth trial and I'll send you a no-cost three-month license key for your own use.

The pre-compiled version comes with some restrictions. People will need a license key in order to:

  - generate FPB files
  - create or search in-memory fingerprint arenas with more
     than 50,000 fingerprints
  - perform Tversky searches
  - perform Tanimotio searches of FPS files with
     more than 20 queries at a time.

See http://dalkescientific.com/LICENSE for the full license.

I also plan to allow no-cost academic use of the pre-compiled binary under my academic license agreement at http://dalkescientific.com/Dalke_Academic_License.pdf , with a one-year license key. Academics can also purchase the source code under the same license terms, which has no time restriction.

Why the change? Up until now I've maintained two versions of chemfp:

   chemfp 1.x - a no-cost open-source version, which only support Python 2
   chemfp 3.x - a paid-for version, which supports Python 2 and 3
     (various licenses available: open source, proprietary/corporate, proprietary/academic,
      at different price points.)

I will continue to maintain the chemfp 1.x line so it can be used as a FOSS baseline for other similarity search tools.

However, all of the underlying chemistry toolkits dropped support for Python 2, making it rather less useful for that goal.

The new licensing model lets people use chemfp 3.x to generate fingerprint data sets which can then be used by chemfp 1.x or any other fingerprint program which understands the FPS format.

I hope it also encourages people to get their toes wet with chemfp, find out they like it, and decide to pay for expanded abilities.

Best regards,

				Andrew
				dalke at dalkescientific.com