Home > opinion > “Lidar News” publishes uninformed (L)GPL rant

“Lidar News” publishes uninformed (L)GPL rant

In its “Random Points” column, the June 2015 issue (Vol.5 #4) of Lidar News, recently renamed LidarMag, contains an opinion piece called “Open Source Mania” (PDF) by Lewis Graham, a director of the board with ASPRS, the organization that defines the LAS file format.

The article contains grains of interesting and potentially relevant comments on the LGPL, but without properly spelling things out: The LGPL – if not amended with a static linking exception as in the LASzip license – has “copyleft” implications when the library code is statically linked, which is somewhat similar to but not as strict as the “strong copyleft” nature of the GPL. I recommend reading the LGPL section of the “Copyleft Guide” or a good article on the Open Source “risks” and considerations during corporate acquisitions and mergers.

Having said that, Lewis Graham’s piece contains many inaccuracies and unfair judgements:

1) The author underhandedly attacks Martin Isenburg’s broadly supported attempts to have the LGPL-licensed de-facto standard LASzip accepted as an Open Standard and then goes into a rant about the GPL, while lumping both licenses together as “viral”. The LGPL – not the GPL – explicitly allows the use as dynamically linked library without any licensing impact on the main program. Combining the attack on the LASzip community with a rant against GPL while ignoring the main difference between LGPL and GPL is unfair to say the least.

Lewis also fails to mention that the LASzip license is actually LGPL with an additional clause that explicitly allows even static linking without licensing impact on the main program.

2) He falsely claims that “Open Source” was never defined but fails to mention the Open Source Initiative (OSI) who coined the term and provided exactly that Open Source definition.

3) The article misleadingly mentions the Free Software Foundation as the “anchor organization” of Open Source, repeats the old misunderstanding that “Free Software” should not cost money and fails to mention the “Free Software” definition by the FSF.

4) The author spreads long debunked FUD (fear, uncertainty, doubt) about the GPL, claiming that all “code [that] touches GPL code in any manner [..] is now GPL”. In particular he makes false claims that the following uses of GPL code would be “viral”:

  • Executing a GPL licensed program from a script via process execution (“Unix fork”) explicitly does not impose any licensing restrictions on that script, as per GPL FAQ
  • Derived works that incorporate GPL licensed code, do not automatically become GPL licensed. Only when the derived work is distributed the following applies (quoting GPLv2, section 6): “Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions.”

5) The author repeatedly refers to an “Open Software Foundation”. He seems to confuse or conflate Open Source Initiative (OSI) and Free Software Foundation (FSF). The actual “Open Software Foundation” – later merged into “The Open Group” – has not much to do with either OSI, FSF or software licenses.

6) The author keeps calling the GPL “toxic”, while in fact the GPL is a widely known and court-proven license in the software industry that many companies use successfully as part of their business models, for example Redhat (Linux), Oracle (MySQL), etc. Especially Dual Licensing based business models actually benefit from the relatively restrictive nature of the GPL.

7) The author praises the MIT license as “reasonable” because of its permissive nature (i.e. not imposing any significant licensing restrictions on derived works). As he makes that judgement he takes only the perspective of companies that want to use Open Source libraries in their proprietary closed-source products. He ignores the perspective of Open Source developers, communities and companies who want to protect their work from embrace and extend and other hostile take-over strategies and deliberately use copyleft licenses like the GPL to protect their software.

8) Overall, the author fails to accept that it is the copyright holders freedom to chose a license that suits their needs and intentions.

Advertisements
  1. September 21, 2015 at 08:00

    The part about LGPL being a strong copyleft is not accurate. The requirement is to allow the user to upgrade and modify the LGPL part. One way to do that is to dynamically link. Another is to ship object code (of your closed source application) and allow the user to relink. Neither of those requires your source to be made public, nor do they impact on the licensing of your source.

    • September 21, 2015 at 08:37

      Hi Brad. Thanks a lot for the feedback. Greatly appreciated! :)

      Do you mean what Lewis Graham or what I wrote about the LGPL’s copyleft nature is not accurate?

      This is my statement: “The LGPL – if not amended with a static linking exception as in the LASzip license – can have “strong copyleft” implications when the library code is statically linked”.

      What I was trying to acknowledge was Graham’s sentiment that LGPL acts very similarly to GPL when you only consider static linking. I also wanted to say that what he wrote about LGPL and copyleft was still unfair given that LASzip license has the added static linking exception clause.

      Let’s see if and why my statement might be ambiguously worded or even inaccurate. Maybe you can suggest a more precise rewording of this statement or the surrounding paragraph. I will be happily to improve my article.

      • September 21, 2015 at 17:56

        I meant that your statement is not accurate. You have to provide ability to update the LGPL part. How you do that is up to you, but static linking alone does not have “strong copyleft” implications.

    • September 22, 2015 at 12:38

      Ok, thanks Brad. I reworded that sentence to better distinguish between LGPL’s weak copyleft and GPL’s strong copyleft.

  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: