search

admercs / gapfraction-website

Website for the gapfraction package for R: https://github.com/admercs/gapfraction
Apache License 2.0
1 stars 0 forks source link

lidR does not rely on LAStools #1

Open Jean-Romain opened 7 years ago

Jean-Romain commented 7 years ago

Hi,

In your page "pre-processing" you wrote:

By far, the simplest option for pre-processing data is to follow the lidR package tutorial, which utilizes LAStools under the hood.

This is wrong. lidR relies only on its own code (and other R package) and does not have any link with LAStools. Please correct this error in this page and everywhere it could occurs because I don't want people to believe there is a link between me and LAStools. lidR is not a wrapper over LAStools. It's confusing for people and I have two problems with that:

  1. If there are bugs or misconception or anything which goes wrong with my package I don't want people to complain to Martin Isenburg (it already happened once).
  2. lidR is not LAStools. Different strengths, different weaknesses, different purposes. If people believe they can use lidR like LAStools they will be disappointed (it already happened once).
  3. I spend so much time writing an open source framework, don't tell it relies on a closed-source software it's not kind :cry:

Regards

admercs commented 7 years ago

Hi @Jean-Romain,

Am I incorrect that lidR relies on both LASlib and LASzip, both found within the LAStools project on GitHub? These are the open-source components of LAStools. Is that not why Martin Isenburg is listed as the second author of rlas? I believe it would be misleading to suggest that lidR doesn’t rely on LAStools.

The functionality of lidR also follows that of LAStools to a large degree, which I think is a good thing. These are common operations in LiDAR remote sensing of forests. It is not meant as a criticism, but was a good design choice in my opinion.

Jean-Romain commented 7 years ago

Hi

rlas relies on LASlib and LASzip to read and write las and laz file that's right. LASlib/zip are open source parts of of LAStools that's right too. However lidR is not related to LAStools. lidR uses its own algorithms and does not make any call to LAStools close source binaries. lidR is not an R wrapper of LAStools. But rlas is an R wrapper of LASlib. It embeds the Martin's source code for technical reasons. This is why Martin is a copyright holder of the package.

What I mean here is that your description says that lidR uses LAStools algorithms under the hood. This is not correct this is why I asked you to be more accurate in your description because it adds confusion. Users should be very aware of what they use (whatever their software choice). An algorithm is not a button to click. It's a choice to make. And so far we are far from this objective. This is why adding confusion should be avoided in my opinion.