Licensing question

[Chris Travers]

On Sun, Feb 28, 2010 at 8:09 PM, Mark James <mrj at advancedcontrols.com.au> wrote:
> On 02/27/10 10:51, Chris Travers wrote:
>>>
>>> Chris, Shareware isn't usually source-available, freely
>>> re-distributable, and non-crippled in trial form.
>>
>> I have seen a lot of shareware that was any combination of these.
>> Whether or not it is typical shareware, I think that is still the
>> correct genre to consider the license under.
>
> Chris, unless pan-handling is used as the means of payment, shareware is
> incompatible with FOSS licences. And the combination of a standard
> commercial
> licence with standard distribution and payment infrastructure doesn't make
> FOSS-like open development feasible. That's why a different system
> and licence was needed.


Shareware and FOSS are different concepts and incompatible.  Sorry if
I was unclear.

>
> If you've paid for the software, the developer is obligated to fix bugs
> for free. There's no problem selling support services on top of this
> for greater timeliness, or in the form of feature bounties.

I found a bug in Windows Server 2003 (not found in Server 2000)
environment variable handling that Microsoft plainly said they
wouldn't fix.  (The issue had to do with Unicode handling, empty, null
terminated strings, and set but empty environment variables.)  It
wasn't a problem everyone would run into but it kept some versions of
Websphere from installing without setting a bunch of environment
variables first.

Also, there are issues regarding prioritizing fixes.  I wouldn't
expect Microsoft or anyone else to fix every bug reported by anybody.

> A $100 fix would be impractical for poorer users. And in large companies,
> where someone is already being paid to manage the software, they can become
> more familiar with it by fixing it internally. Not to mention wanting to
> avoid
> the administrative overhead of the external contract (though this is offset
> with the optional burden of up-streaming a patch).

I am just saying where my time would be better spent.  YMMV.
Personally I would be happier if my customers sent me bug fixes
instead of paying me to do it simply because that allows me to focus
more on paid development work focused on new features.  But whatever
the customer wants....

> When an external package is both small and part of the same system I'm
> already familiar with, such as a framework plug-in,  I find that it's
> much quicker and easier to just fix things myself. The Rails Wheels licence
> is particularly apt for such smaller packages, rather than huge systems
> with many prime developers and contributors.

I have contributed patches to CPAN modules like DBD::Pg,
Excel::Template::Plus, and others.

I have submitted bugs to various FOSS projects as well.  It saves me a
bunch of time if the problem is not a showstopper to me to just file a
bug report.

> I don't think Vista is buggy, it just had some poor design choices.
> I think it's certainly more polished than any Linux distribution.

I didn't say buggy.  I said it had many good ideas which were poorly
implemented.

>
> How do the PostgreSQL developers support their work?
>
>  - EnterpriseDB employees, funded by closed-source extensions, closed
>   and restricted documentation, and paid support contracts?

Same with Green Plum.  However, not the same for CommandPrompt (these
days) or Red Hat (the combination of whom probably submit at least as
much code as EnterpriseDB and Green Plum put together).

Also, this dual licensing of BSD and EULA works very well for them but
only because they are a multi-vendor project with a rapid pace of
development.  This is something worth emulating for many projects, but
certainly not where you are going with your project, right?

>  - Corporate users whose work is supported by other closed software or
>   products?

Not sure what you mean here.
>
>  - People looking for work with one of the above two?
>
>  - Paid bugfixes and enhancements that their customers are willing to
>   have committed up-stream? (Private fixes and deployment services
>   don't directly help the distribution.)


It takes very little sales effort to have customers aware of the
benefits of submitting the enhancements upsteam.
>
>
> The Rails Wheels licence is designed to make it easier to be an IOSV
> (Independent Open Software Vendor). Everything open, without any corporate
> life support system or pot of gold.

Sure.  But it's not really comparable to FOSS.

>
> Email lists are a crap-draw. No-one's obligated to help. Many questions
> don't get a satisfactory resolution, particularly the more complex ones.
> That turns off many users.

Email lists depend on the willingness of the developer to help out and
encourage a community of folks to do so.

> The prospect of being given a free licence or a revenue share is a good
> incentive for all types of users to contribute, without the corporate
> appropriation that occurs when uncompensated copyright assignment is
> demanded by dual-licenced projects.


I think that if any of us had not considered the pro's and cons of
license fees, we wouldn't be on this list.

Call it what you want, but it is well outside the expectations of what
users expect from "Open Source" these days.

Folks on this list probably are aware of my harsh feelings for folks
who think that everything MUST be open source.  I think PostgreSQL is
a great counter-example, and shows what can be accomplished with the
BSD license if you have a vibrant community.  However, you aren't
going to generally entice FOSS developers over to the world of license
fees unless there is a real open source basis on which this can rest.


>
> Again, I'd be interested to know the employment circumstances of these
> developers, and to what extent some developers are prevented from
> offering such advice so as to not dilute the paid support services their
> employers offer.

Tom Lane is one of the most helpful people on the lists and he is
employed by Red Hat.  Many others seem to run primarily support
businesses (CommandPrompt, for example).

I have also gotten help from employees of Affilias, Fujitsu, and others.
>
> Other than the Command Prompt offering, all the other PostgreSQL add-ons
> are closed in some way, which loses all the open advantages we agree on.

Right, but they are closed because the community doesn't want them.
Nobody wants Pg to treat empty strings as NULLs like Oracle, for
example, except those migrating from Oracle.  What this means is that
EnterpriseDB offers everything outside this narrow area back to the
PostgreSQL community.

>
> If some of the PostgreSQL add-ons adopted a Rails Wheels-type licence, they
> could, like Command Prompt, open their work while remaining profitable,
> without having to focus on support over development.
>
> The restrictive nature of OSS is not suitable for some pieces of software,
> and this has driven developers away from the open model, throwing the
> baby out with the bathwater.

I wouldn't necessarily call it restrictive.  I don't see the BSD
license as restrictive.  However, some software shouldn't be OSS and
on that I would agree.  Consumer software, for example, lacks any
business model where one can really monetize the software outside of
panhandling.

> I just want to restore openness by providing a
> licence that relaxes the OSS model at appropriate points.

No complaints there.  Just not sure it is comparable to OSS.

>>> 3. Making it hard for the software writer to directly charge for their
>>>   software denies them the power of replication (mass production), which
>>>   supports the income of just about every big company and wealthy
>>> individual.
>>
>> Maybe.  I am not convinced.  I am working on replicating my efforts.
>> Plus there are other ways of replication too  such as selling turn-key
>> solutions and the like.
>
> Will you open-source the turn-key build?

Absolutely.
>
>>>   Customization is hard work, and to fairly reward one's talents one
>>> should
>>>   be able to get away from doing that alone.
>>>
>> Like it or not, this is the best way to perfect the software.  It gets
>> you a first-hand look at what customers actually need in the next
>> version.  Think of it as charging in advance for development instead
>> of in arrears for licenses.
>
> Why not do both?

Business decisions.
>
>>> Do you know what percentage of your users pay for support?
>>
>> I don't know what you mean by "my users."  If they are "my users" it
>> is because they are paying me for support.
>
> OK, I've now looked at your website. I was looking from the perspective
> of the support offered by a prime developer of a smaller package, rather
> than a more independent support service for a larger package.

Understand I am not independent of some of those projects.  I have
committed roughly half the changes in LedgerSMB since the fork, for
example.  I am independent of most of the other contributors though.

>
> BTW, how many of your customizations get incorporated into the standard
> package?

80-90% get distributed either in the standard package or into addons
in the same svn tree.

We are moving towards more addons and a smaller standard package for
LedgerSMB though.

>
> When the customers can't afford cost-plus for the required work, you
> must choose between rejection and cross-subsidization. Too much
> cross-subsidization always collapses. Charging everyone for using
> the software at least limits it.

I offer discounts to address some of this.

>
>
>> Honestly, I get more money out of a donation of code than a small
>> project of cash.
>
> Could you explain this more please.

Either it fixes a problem for me with less overhead on my part, or it
provides new functionality that lets me sell to a larger market.
Either way it does so much faster than if I am paid to do it myself.

> If your offering is compelling, unique, and not easily (clean-room) cloned,
> most users will pay modest licence fees, which can be automatically shared
> between developers according to an agreed level of contribution.

Consider LedgerSMB. (http://www.ledgersmb.org)

Distribution would be smaller, sourceforge would not be available to
host the downloads (meaning hosting expenses wold go up), and I would
lose out on the compelling perspectives of other contributors.  It is
great to jointly develop software with folks like Josh Drake, Chris
Murtagh, and Seneca Cunningham who can not only contribute code but
provide much needed perspectives in building quality software.

However, why I make money is because I know the code very well.
Nobody can out-produce me in this area.  So my time is worth more than
many other developers' time.

I am also the go-to person in the community so when people need work
done I am one of the first folks they come to.

Again, these are business decisions.

The key issue between OSS and your license IMO is that your license,
not being really open source or free software by expected standards is
that you have a tradeoff.  On one hand, you cannot leverage the open
source community as well, but on the other, you can demand licensing
fees.  That is a tradeoff that your business has to decide on.

Best Wishes,
Chris Travers