SPDX License List v1.12 - OSI Approved column added

[Jilayne Lovejoy]

(I've cc'ed Karl at OSI and the OSI license group email, so they are aware
of this conversation)

Re: Juergen's second point below about having more options than "yes" or
blank for the OSI Approved column on the SPDX License List - I don't think
that's a good idea.  For example, "not OSI compliant" would require someone
making that call and is open to interpretation as to meaning, e.g. Does that
mean the license is simply not listed as being on the approved list or does
that mean the license was submitted for approval to OSI and rejected?

My understanding of the OSI Approved license list is that it's pretty
straight forward: there are some licenses that have been approved and they
are listed with no information as to why many licenses are not listed.  I
think we simply want to mirror that information.

How we mirror it, I'll leave up to the tech team, but for the time being as
the keeper of the SPDX License List master spreadsheet, I'm fine with
maintaining the column manually (and welcome other eye balls
cross-checking!).  I don't think the OSI approves new licenses that often
and thus, this field won't change that often, which may mean the effort for
some kind of automatic syncing is not worth it in so far as a cost/benefit
perspective.  

That being said, it would be great if OSI could notify SPDX when a new
license is approved, so we can add it to the SPDX License List. Which then
leads me to think that our adding-licenses process may want to include an
automatic addition for any new OSI Approved license?

Stopping now...

Jilayne

Jilayne Lovejoy |  Corporate Counsel
jlovejoy at openlogic.com

720 240 4545  |  phone
720 240 4556  |  fax
1 888 OpenLogic  |  toll free
www.openlogic.com

OpenLogic, Inc.
10910 W 120th Ave, Suite 450
Broomfield, Colorado 80021


On 6/9/11 11:27 AM, "Gary O'Neall" <gary at sourceauditor.com> wrote:

> Hi Juergen,
> 
> Very good points.
> 
> I like the idea of keeping it in sync automatically, but I worry about the
> reliability of any implementation.  I took a look at the HTML source on the
> OSI website and the license list is simple HTML text.  This can be parsed,
> but any changes to the website structure would cause the parsing to break.
> 
> If OSI could provide an XML page or an RDFa page of OSI approved licenses
> which reference the URI's of SPDX licenses, we could accurately keep these
> in sync.  If there is someone at OSI I can connect with on this topic we
> could explore this approach.
> 
> On you second point, I'll leave that up to the folks who are maintaining the
> license spreadsheet.  I was thinking of a simple checkbox in the UI for the
> OSI approved, but it is actually easier to just copy the text from the
> spreadsheet column.
> 
> Gary
> 
> -----Original Message-----
> From: Juergen Weigert [mailto:jw at suse.de]
> Sent: Thursday, June 09, 2011 2:15 AM
> To: Gary O'Neall
> Cc: 'Peter Williams'; spdx-legal at fossbazaar.org; spdx-tech at fossbazaar.org
> Subject: Re: SPDX License List v1.12 uploaded
> 
> On Jun 08, 11 21:51:04 -0700, Gary O'Neall wrote:
>> I'm not sure if a formal decision was made, but I am in favor of adding
> this
>> property.  If no one objects.  If no one objects, I'll update the website
> to
>> display a column for OSI approved and implement the machine readable
>> proposal in the tools once it is approved.
> 
> I am in favor of adding that column too. This makes me think of two
> additional thing:
>  - As OSI also adopts our names, it should be possible to automate the sync
>    from OSI to SPDX for that column. This is an important item, as an
>    outdated column could do more harm than no column at all.
>  - OSI keeps their approved list short, although there are many licenses
>    out there, that appear OSI compliant, or even close duplicates.
>    I wonder if it made sense to design the column in such a way that we
>    could say: "OSI approved", "Not OSI compliant", "undefined"
>    Where "Not OSI compliant" is chosen when a reason is known, why it
>    cannot be approved.
> 
> Am I making things too complicated?
> 
>         cheers,
>                 JW-