Time-delay GPL
Paul D. Fernhout
PlantStudio software to general manufacturing processes)
http://www.kurtz-fernhout.com/PlantStudio/
under a "time-delay GPL" of three years. That is, to use the software, you
need to pay per licensed machine to stay current, or wait till the product
becomes GPL three years after release.
id Software has done this informally with its games (five years for them,
and it is not formal, and you don't get the source with the product).
http://en.wikipedia.org/wiki/Id_Software
Here is a proposed readme file. How hypocritical would it sound coming from
me? :-)
====
"""
This software is what I call "time-delay GPL".
Source code is available for the software when it ships, but the source
code is under a proprietary license for the first three years from the date
it is first released. After the three years is up, the license becomes GPL
version 3 or any later version.
This means, to use any version of the software from the last three years,
you have to pay for the software per machine you want to use it on,
and any changes you or others make for that version make can only be used
by people who have paid for that version, until that version of the software
becomes GPL.
I am sorry to do this, creating a temporary artificial scarcity of the
software. Ideally, this work would be supported by grants or tax dollars or
by a basic income. But, that is the society we live in right now, and I can
either chase grants or write software. This is a compromise to get the
software written and out there.
The model for this is what id Software has been doing, with releasing
versions of game software they make under the GPL after about five years.
This also reflects the notion that in the internet age, with quicker
distribution, that copyright should only last three to five years
(something Richard Stallman has suggested, even though he might not approve
of this?).
The good news for people who support this software by buying a copy is that
I have every incentive to keep working on improved versions with the clock
ticking. :-) And, you have the source code, so down the road, you know the
software can be maintained by a community someday as well as made part of
free software distributions. So, it can become a definitive free standard
to preserve your time investment in learning to use it.
And if you don't want to pay for the software, you can just be patient for
three years. :-)
"""
====
Any thoughts on this possible compromise? It's sort of like a variation on
shareware, with a free software twist that makes it totally different. :-)
Obviously, growth may be limited, but I'm just trying to be a small software
developer who can pay the bills, not the next Microsoft or even WinZip.
Realistically, I think the "service" model of software support is a failure
in many cases (not all), encouraging bad coding practices if you build a
product. After all, in a service model, there is a lot of incentive for the
software to have rough edges. I know I can build good software products, and
respond to customer feedback, but it takes many months (even years
eventually) of full-time work to make a good product. A basic income would
support that kind of work, but that's not what we have yet.
I'm not sure if I will do a time release GPL, but I'm thinking about it. I'm
also not sure what practical problems it may have as well with third-party
plugins, an issue to think about.
I'm also not certain about what product I would make. Part of me would
rather work towards a social semantic desktop, but NEPOMUK already spent all
the grant money for that. :-) And I don't see a social semantic desktop
focused on open manufacturing doing as well under a time delay GPL as a
stand alone tool like a 3D editor. Without being free from the start, such a
platform just won't have much momentum, and the point of social semantic
desktop is to be social.
In my thinking, for a time-delay GPL product, you'd get the source code with
the product -- you just can't use it under the GPL until three years from
release. This would create *temporary* artificial scarcity though. Which
obviously I am no big fan of. But three years seems to be reasonable?
Frankly, most open source projects I've seen only have a few key
contributors anyway, as it takes a long time to get inside the codebase and
understand it to where you can make significant contributions. Yet, having
the code is great peace of mind, and is a way to move things forward
eventually if the main developer moves on to other things.
As I said, I can either write code or chase grants. And if I chase grants,
chances are I won't get the grant and the software won't get written either.
From:
http://p2pfoundation.net/Towards_a_Free_Matter_Economy
"Both grants and prizes introduce serious hazards for poisoning the
self-selection and cooperative advantages of the bazaar development model."
There are lots of other things I can do, but most seem to entail moving to
expensive places to live, requiring two-incomes to pay for a house and not
being able to homeschool (the "Two-income trap"). NIST actually seems like
the best match as a job to open manufacturing, but is in an expensive area.
I'd definitely recommend anyone here looking for work doing open
manufacturing who does not have a family (or is willing not to homeschool)
and is mobile look hard at NIST.
"Sustainable and Lifecycle Information-based Manufacturing"
http://www.nist.gov/mel/msid/dpg/slim.cfm
Anyway, all my choices are risky or entail changes at this point. This
Time-delay GPL seems like one of the lesser evils?
--Paul Fernhout
http://www.pdfernhout.net/
Trent Waddington
Fernhout<pdfer...@kurtz-fernhout.com> wrote:
> id Software has done this informally with its games (five years for them,
> and it is not formal, and you don't get the source with the product).
Umm.. you're intending to give the source code in the first 3 years?
What will happen is that someone will buy your license (or just warez
a copy if they are floating around) and write their own version of the
software and make that new version immediately GPL. They'll claim
they did it completely clean-room and you won't be able to prove
otherwise.
Trent
Paul D. Fernhout
Trent-
Thanks for the feedback.
I've spent many years trying to move beyond a mindset of creating artificial
scarcity. And the issues you outline, playing a cat and mouse game of who's
got the source, and setting up barriers, and pitting experienced programmers
against young programmers in and adversarial way fighting over money and
ideals, and so on, it's all rather distasteful. And wasteful too.
Still, the only way I had time to move beyond that was by my wife being
enmeshed in that world, so it is not like I had a completely sustainable
solution in that sense either for our current economy. Still, I did try to
make some small free contributions towards transforming our economy towards
a post-scarcity one, but the effects of any of that remain to be seen.
Anyway, I'm still trying to figure out a way to follow Emlyn's suggested
theme to both make things free and also sell them. :-) And I'm considering
Ben's comment that the game market is saturated. (Both comments were made in
the for-profit vs. non-profit thread a while back.) The iPhone is up to
55,000 applications last I saw, so I'm losing interest in that platform. :-)
It's interesting to me that the first comment here was not about whether
this approach is ethical, or whether it is it hypocritical for me
specifically to consider it, but is it profitable? Thanks for your concern
for my welfare. :-)
You're right, doing something with the source code in violation of the
license may be an issue, but how likely it is, and how much it would effect
the bottom line for such a product is a different issue. Here are some
thoughts on that.
Software of any size can be non-trivial to deal with. For example, over the
last few years, I've spent in total several person-months full-time
converting StoryHarp (and two other programs) from Delphi to Java. Even with
that, StoryHarp (the smallest of the three programs) is still not completely
finished (and was put on the back burner as I started going through the
mental anguish of shifting gears after eight years of becoming more of a
free nurturer than a market competitor). I think it will take another month
of work to polish StoryHarp and write new help and more examples and a new
tutorial and so on.
Granted, translating code and using a different GUI library is different
than just massaging it a little, but the less you do, the more obvious the
similarities would be. And what matters there is what the community thinks,
as I'd expect most GNU/Linux distributions are not going to redistribute
code that looks tainted, whatever the claimed license. Even if there was no
way to prove anything legally, the similarities would create all sorts of
issues in distributing something.
One reason for that is that there is a lot that goes into a serious
application besides the source. There are icons, help files, examples, other
support files, mailing lists, web pages, tech support, new versions, and so
on. These are all hurdles even as the original source itself may take a lot
of energy to wrestle with (even if you wrote it yourself once upon a time
:-). That's one reason software is often written by teams who do different
aspects.
One person violating a license might be easily imaginable, but imagining a
whole team of people working together to violate a license seems much harder
(again, possible, but unlikely). There are all sorts of more interesting
things a team of people could do, including just writing their own
GPL-from-the-start application in their own way. And from an indy developers
point of view, the problem is not getting copied so much as it is getting
noticed.
So, I guess the bigger the application is, the less I see trying to pass of
a slightly rewritten version as original being a likely problem.
Some kid can choose to spend months rewriting the source of something that
will be GPL in a couple years anyway, or they could instead make a funny
video for youtube or do something else more interesting socially or
technically than changing variable names and control loops in code someone
else wrote to make it look different. Which would most kids chose to do?
Even if a kid does decide to prove a point by making a questionable
contribution to a community, most people in the community might view very
that "contribution" very negatively (even if they don't like proprietary
code of any sort).
The fact is "maintenance" in that sense is generally some of the most
boring computer work that few people willingly want to do. That's one reason
a lot of free software is not so good -- it's fun to write the first or
second version, after that it can become hard work that you do only because
it is important to you idealistically, socially, or financially. Most
projects just aren't important enough to most people who start them,
especially if other things (relationships, children, jobs, family,
vacations, recreation) get in the way.
So, I can wonder if it would really be worth it to anybody to make some kind
of a point by doing what you outline as a possibility? A person would have
to combine in one personality wanting to promote some ideal (free software
or being 'leet), while also being willing to violate a proprietary license
(breaking the law, even if indefinite copyright is clearly unjust at this
point), and then they would also have to be willing to do a lot of work.
Even all that is not enough, as that rare person would then have to decide
to pick on this one specific project (given all the other code out there to
do stuff to in various ways). Sure, that is possible, but is it likely?
Of course, it the source was only put under the GPL in twenty years, or if
the product was completely proprietary, then I could see there would be more
people willing to do something about that. Their would be a stronger
consideration of an unjust copyright length for the internet age and the
notion of civil disobedience. For a three year copyright term today, it is
harder to make the case that civil disobedience is justified.
So, the choice of the number of years of the time-delay seems important. I
was waffling between five years like id Software does informally and three
years. I think three years just feels better, as something that is within
the scope of younger kid's teenage years between, say, thirteen and sixteen,
some of the prime antisocial hacking years for some. :-) A thirteen year old
can either make it difficult for someone to develop that product or they
could do something else that is more fun and have that product's code
legitimately under the GPL when they turn sixteen. Sure, some thirteen years
olds may do it just to be obnoxious (or for some absolutist stand, which may
not even be wrong :-), but hopefully, they will find other things to do?
Again, it's possible, but is it likely? It would also clearly be in their
own best interest for this model of self-limited copyright length of three
years to replace other proprietary software distribution models, so maybe it
would get some sympathy points there? I don't know. It is hard for me to
guess what today's teenage anti-social (or even pro-social) programmer would
be thinking. I'd certainly be interested in anyone's comments on that who
had a more current insight into that world.
Also, the risk of this happening could be traded off against maybe getting
more sales up-front by being open, and people wanting to support the idea
somehow. Even retail stores trade off the certainty of shoplifting against
the rewards of being in business. From a marketing standpoint, I'd be a lot
more likely to buy software with such a license, including occasional
updates, assuming the software did something I wanted. If others felt the
same way, that might translate into better initial sales and more publicity.
So, it is risk versus reward.
This approach does trade off the community aspect of software development,
of course. The code could not go up on SourceForge for example. But, while I
do write open source stuff, and have projects on SourceForge, aside from
some small contributions to a few other projects, the things I really do are
essentially big applications written by myself (or, in the past, myself and
my wife). So, community as far as coding has never been a big issue for me
in practice in my own independent work, even as community as far as customer
feedback has been a big issue shaping what we work on or how it changes
(that rule of 80% of innovation comes from customer suggestions). I've coded
collaboratively on projects in corporate settings, but that came with a lot
of other corporate baggage, so it is hard to compare with some of the better
collaborations one can see on the web.
Now, that independence may be a failing on my part, of course. It also
reflects just the fact that I've been writing software for thirty years, so
my habits and mindset were formed in a different and more isolated-developer
age. Strengths are often also weaknesses, depending on the context. So, in
my particular case, with this particular licensing strategy, it may make
more sense for me than for more social people. I'm not that directly
extrovertively collaborative, even though I am indirectly introvertively
"stigmergically" collaborative by posting free stuff or free thoughts for
people to build on or transform, as I in turn also build on what others do
(complex replies, bringing in links, using software libraries, writing
libraries, etc). So, in that setting, not having a community of developers
for a specific project does not mean lack of collaboration, it's just a
different type of collaboration.
I feel what I do as a developer also reflects the reality that certain types
of large desktop applications (especially in new areas) take a big
investment of time to get into and understand, and few people are able to
make that investment due to lack of the time even if they are willing. Other
classes of programs may have different dynamics. That's one reason I've
gotten so enamored of the basic income idea lately, so people will have the
time to do bigger projects. One reason there are so many abandoned free
software projects is in part just that the need to earn an unrelated income
has intruded on people's concentration. I've seen endless posts by free
software developers that if they only did not need to have a day job they
would be doing more for some important (to them and others) software
project. SourceForge would be a very different place if we had a basic income.
Modularity affects this too. The more modular a program is, the easier it is
for people to make a good contribution in a limited time. But the core of
many applications is rarely modular -- it may have modules, but they become
so interdependent, there often is a big learning curve to get the gestalt.
Better software tools help with this, of course. But there still often seems
to be a learning curve for all this for a big package.
What I feel is more likely is your other point that someone will post the
code or warez the product. That takes next to zero effort. Or in the case of
cracking a registration scheme, it takes just enough effort to be
challenging to some young programmers:
http://www.google.com/search?hl=en&q=storyharp+crack
So, what do do about that is an issue.
It's not clear to me if it would be best from the start to have the code on
a website anyway, or if you only get it when you purchase the product. Also,
it is not clear if the product should have a demo, have a shareware
registration scheme, or just be downloadable on purchase of a license (with
source). More stuff to think about if I pursue that idea of self-limited
copyright.
Still, one alternative to a time-delay GPL is just to ask for donations
(like tip jars). But, I've never seen a one person programming project that
was sustainable on that basis (that does not mean they are not out there).
Advertising is a possibility too, but as I turn it off myself, I'm not to
enamored of building a business around that.
Lego Digital Designer, by the way, is an excellent example of a
free-as-in-beer product sponsored by a company to sell more product. Concord
Consortium is a great example of a non-profit funded by grants to make
free-as-in-LGPL software (although Concord is so school-focused and I am so
Gatto/Holt/Unschooling-oriented that it is unlikely that would be a good
match). So, clearly foundation or commercial sponsorship is possible it
theory, but I am just not in those channels, and so I weigh the risk of
trying to get in those channels (time, restrictions) versus doing something
else.
As a thought experiment, let's say Bryan used this method to release some
new fancy set of DNA he was developing for DIYBio, but which he needed to
fund the development of with investors to pay for equipment and years of his
time, like some new lines of custom genetic material for synthetic meat.
Let's ignore any ethical issues of DNA hacking, as well as whether or not it
was a matter of life-and-death for people (or animals) to have synthetic
meat (some might argue both points, and that changes the issue). In that
simple case, if Bryan said, "you have to pay for any cell cultures I sell
you, as well as restrict their spread in various ways, but only for three
years", well, would that sound reasonable to a company today who wanted to
produce synthetic beef and get some of those culture lines? Would it sound
hypocritical to hear Bryan saying it today? Would we think less of him, for
all his other work? If I were inclined to pay for that sort of thing in
*today's* economy, would I? Would I accept Bryan's compromise as reasonable,
even while bemoaning it? I think I would. And frankly, I think people would
ever pay *more* for those lines than completely proprietary lines of cells,
which might help with the business plan. And sure, anyone would prefer the
cell lines were free from the start -- except, under our current economy
organized under dismal-science economic ideology, the free-at-the-start cell
lines are often just not an option (unless they were developed as a hobby or
grant-funded, both perfectly fine things to do, of course, if Bryan can
manage them, and preferable if feasible). Sure, in three years, everyone
will have these cell lines, but if Bryan has confidence in himself, a three
year head start over competition with himself seems like something he can
win. :-) At my age, that may be more questionable. :-)
At least this approach limits the artificial scarcity to a short time frame,
one more reasonable for the internet age and fast modern transportation,
considering how copyright was only twenty years long in the age of the Pony
Express. Would people cheat on these cell lines, and give them to friends,
maybe. But would the discerning consumer want to buy the products of things
that were not somehow certified by the producer? It may also depend on how
high the original cost is, too.
If coders (gene or software) wanting to make an identical product from the
source put enough effort into making the code look different (formatting,
comments, variable names, structure), they might just as well just write it
from scratch. Most effort in software goes into design and redesign anyway,
not the coding (even if the code represents the design). For any existing
product, it is much easier to write a reverse-engineered clone than the
original, because you don't have to go through a maze full of lots of wrong
turns. That's always been true:
"Software Is Hard"
http://gamearchitect.net/Articles/SoftwareIsHard.html
"Scott Rosenberg coins this as Rosenberg's Law: Software is easy to make,
except when you want it to do something new. The corollary is, The only
software that's worth making is software that does something new."
Another way of looking at this is, if General Motors said their next car was
proprietary for three years, but it came with a DVD of all the engineering
plans and the designs for all the process machinery to make the car too, and
then after three years anyone could use the design, would we be more likely
to buy the car or less, from a marketing standpoint? Would we here be
cursing General Motors ("not good enough") or praising them ("wow, they are
going open")? What would Richard Stallman say? And if GM did that, would it
be a good thing for the world, and put pressure on other companies to
follow? Or would it just be seen as harming efforts to make cars that are
free from the start?
Of course, I'm not GM. One might rightfully say a three year license would
be a step backwards for me, even if it was a step forwards for GM. It
certainly feels like a step backward emotionally in some ways. On the other
hand, being paid to write software that was going to be under the GPL in
three years would also be a big step forward for me financially, and so a
big step forward emotionally in other ways. :-)
Anyway, copyright was supposed to be a bargain, where authors and publishers
get exclusive rights to print something an author wrote, but for only a
limited time (twenty years or so). The bargain was broken by a collusion of
legislators and publishers distributors of copyrighted works, who lengthened
their terms as theft from the public. It is obvious now that copyright terms
should have been shortened rather than lengthened because information goes
around the world more quickly now. So, as RMS suggests, copyright is no
longer a good bargain for society and the artist if it is indefinite, and
prevents making derived works or free distribution.
But, we are still stuck where we are right now, including most post-scarcity
public assets (general tax dollars, charitable dollars, education dollars,
defense dollars) all siphoned off into creating more proprietary artificial
scarcity (like with the Bayh-Dole act). So, what can we do? Well, hobby work
is one. It is a reasonable suggestion to think it should be enough. And it
probably is. But for people who need income anyway, is their anything they
can do to be not as bad, assuming they can't get hooked up with the
post-scarcity dollars?
Services, sure. But writing a complex product or making a complex mechanical
design is still not the same as services. MakerBot is sold out of the open
CupCake CNC system,
http://store.makerbot.com/cupcake-cnc/cupcake-cnc-deluxe-kit.html
but really, they are offering a service of systems integration and a buyers
club for parts more than people are paying for the design (which is free).
So Makerbot industries has a niche by spanning the digital and physical
worlds, and are essentially using a free design to drive sales of services
related to physical items. A desktop software application does not have as
much of that need for paying for integration into the physical world. And in
any case, doing integration and teaching classes and so on is not the same
thing as actually building and testing and iterating a complex digital
design. This has been a perennial problem of trying to work on post-scarcity
digital products while surviving in an economy oriented around managing (and
even creating) scarcity.
It seems like, in the case of complex things that might have update streams
(songs might be different), that voluntarily scaling strong copyright back
to three years (and then still requiring attribution or copyleft or maybe
even not) might be a step forward, as well as attract interest and
emulation. Still, I'm open to suggestions it is categorically a bad idea
(and you raised some reasonably practical objections).
This idea is also in a sense laziness, in that if I banged on enough doors
of enough foundations, maybe I could find support for writing a program that
was GPL from the start? Of course, if that process of finding support took
three years, :-) would the world rather have the software now, even if it
cost something? Given in both cases the software would be GPL in three
years? :-) The thing is, I don't know how many doors I would need to bang on
to find funding for a social semantic desktop or for a tool to breed 3D
designs. So, if it took a month full-time, then that would be a better
choice. It is just a big unknown, although everything I know about it
suggests dealing with the government or foundations takes a long time even
when it works, and essentially, getting grants is really a specialized
skill, same as writing code. It takes time and practice to get good at it.
But until you succeed, you may have little to show for it.
There is another way that this is laziness. If I write something myself, I
don't have to account for each hour, and I can change the direction of the
software as I want. If the software is grant funded in a usual grant-ish
way, I'll need to track my hours for the project as well as go through all
sorts of negotiations with the government agency or foundation if the
project changes direction. So, there is an advantage to just making a
marketed product as an indy developer in those senses. The market does have
some advantages in that sense.
Still, if I was willing to travel (and could afford it), I think going
around to a lot of foundations and telling them why they should be funding
free software and open manufacturing would be a great idea. :-) Even if it
was not to get a grant, but just to raise their collective consciousnesses,
Or to give talks at conferences foundations go to, and so on. Building on
this from something sent in 2001 to the Markle Foundation:
http://www.pdfernhout.net/open-letter-to-grantmakers-and-donors-on-copyright-policy.html
(By the way, it's not clear to me foundation supported works should be in
under the GPL instead of in the public domain or under a simple attribution
required license or at most under the LGPL like Concord uses, but that is
another issue.)
But no doubt foundation staffers by now are reading Clay Shirky and so on,
and use Wikipedia, so it's not clear it makes sense for me to do that
foundation consciousness raising myself. Someone without kids, who loves to
travel, and who is extroverted (or who is introverted and likes to perform,
like Sting) might be a better choice to do that. :-) There's a lot of people
like that in the world. :-) It would be a rough life for someone like me,
who would rather be back in his cave making digital stuff, whether code or
emails. :-)
Another thing along those lines is to try to become a pundit about
post-scarcity. But I don't think that would suit me, especially with all the
expensive travel required, and the continual public speaking, though I've
thought about it. I can already think of several people positioning
themselves for careers there (or already having one), and I'm not inclined
to compete with them.
I'd rather just be helping them conceptually on the back-end. Alfred hardly
ever left Bruce Wayne's mansion, right? :-) From:
http://en.wikipedia.org/wiki/Alfred_Pennyworth
"The Pre-Crisis comics (i.e., comics published by DC Comics between 1938 and
1986) established Alfred as a retired actor and intelligence agent who
followed the deathbed wish of his dying father, Jarvis, to carry on the
tradition of serving the Wayne family. To that end, Alfred introduced
himself to Bruce Wayne and Dick Grayson at Wayne Manor and insisted on
becoming their butler. Although the pair did not want one, especially since
they did not want to jeopardize their secret identities with a servant in
the house, they did not have the heart to reject Alfred. ... Current issues
of the various Batman comics seem to indicate that Alfred is a pioneer in
and has also mastered several fields of rose breeding (even creating his
own, the "Pennyworth Blue"), computer programming, computer engineering,
electrical engineering, chemical engineering, mechanical engineering,
nanotechnology, and biotechnology as he singlehandedly builds, programs, and
maintains much of Batman's next-generational technology such as the
Batcomputer."
Anyway, I'm still not sure I will pursue this approach, but it is
interesting to develop the idea (pros and cons).
Thanks again for the feedback.
--Paul Fernhout
http://www.pdfernhout.net/
Trent Waddington
Fernhout<pdfer...@kurtz-fernhout.com> wrote:
> Thanks for the feedback.
FYI, I read about 1/3 of your email. Long and rambling seems to be
the norm on this list, and there's only so many hours in the day.
Folks, can we try to get the signal to noise ratio a little higher?
Thanks,
Trent
Paul D. Fernhout
And I thought your comment about making email posts the size of a text
message (160 characters) was a supportive satirical joke. :-(
You must be new around here; it's all noise, at least my part of it. :-)
http://en.wikipedia.org/wiki/Noise
"Calling some signal or sound noise is often a subjective distinction. One
person's maximum-volume music listening pleasure might be another's
unbearable noise. An annoying background hiss interfering with short-wave
radio broadcasts was found to be due to extra-terrestrial, indeed cosmic,
processes; listening to this "noise" to the exclusion of all other signals
with ever more sensitive antennas and receivers is now the science of radio
astronomy. Radio astronomers are still plagued by noise in their signals —
but now it is thermal noise generated in their equipment interfering with
wanted signals from the cosmos."
What matters most sometimes is how we creatively shape the noise in our
lives into personal meaning. :-)
Still, this idea might help; maybe you could help build it into the Linux
kernel? :-)
http://www.google.com/search?hl=en&q="weak+signal+detection"
Then weak signal detection could automagically run on all text data streams
like emails on idle cores to automatically build a summary or flag text you
wanted to pay extra attention to? :-)
Seriously, better software tools might help somehow, though not soon, and
probably not in kernel space. :-)
We all make our choices in life on how we want to spend our time.
Anyway, thanks for reading some of what I wrote.
--Paul Fernhout
http://www.pdfernhout.net/
Emlyn
My two cents: Don't do it.
1 - You are violating your own ethics. You clearly don't want to close
your code in this way, even for a short time. If you really wanted to
close it, you'd keep it hidden.
2 - I assume you want this non-gpl period because you want a lock of
some kind; a way to give yourself a monopoly for a period of time. I
don't think it'll really help much though. If people want to pirate
it, they will. The small sliver of law abiding people who will pay you
if you are the only supplier, but might not if there were other
options, how many of those people are there really?
3 - Meanwhile, in the free software arena, you'll lose credibility for
this odd approach.
Do you have a copy of "Here Comes Everybody", by Clay Shirky? He
describes the mass amateurization of some professions in the modern
connected world (journalists and photographers are the two examples he
gives). I think packaged client side software is slowly being eaten by
mass amateurization (hooray! what an amazing development for the
world!), and is a bad place to be looking for commercial income at the
moment. One of the best indicators of this is that you'll find
yourself having to do all kinds of nasty protectionist things in order
to make any money, and I suspect they only work for the big incumbents
in any case, and even there probably not for all that much longer.
Remember when there used to be Shareware? Ah, the 90s. I don't really
miss it though :-)
All the software I see these days that has a hope of making money has
a server based component. Server side gives you control, which is good
for you. It also means extra options for the users, which is good for
them.
Some software lends itself to this better than others. Facebook, yes.
Image editing software, not so much. Your project, maybe it closer to
a client side thing than a server side thing? I'm not sure.
A problem with server based solutions is they don't feel very open,
but this needn't be the case. You can make entirely open solutions
here, keep your integrity intact and also service your wallet. Here's
my plan for how:
I'm assuming a client-side type of application. Let's say yours is
largely about designing 3D objects, which end up saved to files, and
is in this class.
1 - Make the client side app. Make it do everything you currently want
it to do. Make it entirely free, release the code under the GPL. Give
it away.
2 - Look at how you can add value by providing a server side. The most
obvious way is to provide sharing capabilities for whatever "files"
your program produces. Think social network. Lots of social tools.
Make this side of things entirely optional, but damned useful. You can
charge for this.
3 - Release your server side code for free, GPL it. Make it possible
for people to set up their own private (or public!) server if they
want to.
So you end up releasing an excellent, 21st century product, entirely
free, and value add by providing a hosted server side service.
Maintain your reputation and I think customers will follow.
Really work hard on making sure people can use your entire solution
for free on their own infrastructure if they want to, never hobble it.
You don't need any artificial lock in here. Just the fact that you are
clearly trustworthy and competent and committed to openness means that
when people see and like the server side stuff, they'll be most likely
to use your service, even if there are other free alternatives. Oh and
I imagine integration with your servers would probably be very
slightly easier than with any others people set up, even if it's just
because the menu options go "Use Official Server | Configure a private
server".
btw, you get excellent trust by making the server side free, because
you address the "what if this commercial service folds?" problem.
People know that at worst they can always host it themselves, so they
feel safe using your commercial service (oh remember to provide import
and export functions).
--
Emlyn
http://emlyntech.wordpress.com - coding related
http://point7.wordpress.com - ranting
http://emlynoregan.com - main site
Paul D. Fernhout
> My two cents: Don't do it.
Thanks for the well thought out reply with some server-based alternatives.
> 1 - You are violating your own ethics. You clearly don't want to close
> your code in this way, even for a short time. If you really wanted to
> close it, you'd keep it hidden.
You're certainly right I have no interest in closing a product. And I do see
an ethical conflict with creating artificial scarcity. I guess I'm weighing
that against the temporariness against anything getting written at all or
working for someone else on their proprietary stuff as a software developer.
But, obviously I can switch professions, too. :-) Or try for grants. Or
something out.
Essentially, it is true, my ethics about artificial scarcity are in direct
conflict with how much of the global software industry works, even the
in-house development stuff.
But ethics, while they may often seem black and white, in practice quickly
become shades of gray (or even a rainbow). For example, do I pay for CPUs
that are closed? Yes. Even Richard Stallman does that. I'm not saying he
likes it.
I have other ethical obligations in my life though, and I am stuck in this
crazy artificial scarcity system. So, there are tradeoffs I am not happy
about involving conflicting responsibilities. It would be better to find a
path that did not involve tradeoffs, of course. So, I really appreciate your
comments to that end.
> 2 - I assume you want this non-gpl period because you want a lock of
> some kind; a way to give yourself a monopoly for a period of time. I
> don't think it'll really help much though. If people want to pirate
> it, they will. The small sliver of law abiding people who will pay you
> if you are the only supplier, but might not if there were other
> options, how many of those people are there really?
I know when we find one of the few programs for kids worth paying for, we
are happy to pay for them (and they are very cheap).
http://grubbygames.com/professor_fizzwizzle.php
http://amanita-design.net/samorost-1/
And even of the many free-as-is-beer ones we have found, many funded by
advertising, few are free-as-in-speech.
At least from when we sold stuff as shareware (ten years ago), there were
people who were genuinely happy to pay for good value. Example:
http://www.kurtz-fernhout.com/PlantStudio/userssay.htm
"Just a word of praise for such an excellent program as Plant Studio. I'm a
registered user, utilizing your excellent program in most of my work, which
is shown in many galleries and about to be seen commerically. I pass praise
of this software to any and all, hoping each who have downloaded your wares
will take the time to register and pay the price, which is far cheaper than
the true value of the program to be sure. In the 3D world of art, it's a
dream-come-true to find inexpensive programs for developing realistic plants
and trees, so I consider Plant Studio totally indespensible. I plan on
getting one of the tree generators, but the cost is high and I need to
compare finished work in order to make my choice as to which one. However,
Plant Studio is excellent as is, able to generate an unlimited array of
living plants, real and imaginary, and for a cost that is unbelievable."
But, times may have changed. Certainly the 3D world has moved on. And also,
it is not 100% clear to me I can bring this kind of thing to arbitrary
manufacturing, so it is risky technically in that sense. And while I know
more than I did then, I'm not so sharp as I was then, either. :-)
What I am thinking about is just a license that says no redistribution for
three years, even as it comes with source, but after three years from
release, then what you have is under the GPL. I don't especially see any
other restrictions, but maybe there could be a registered and non-registered
version might be possible as trial shareware, with the source available
after registration?
From my knowledge of the shareware market from a long time ago, it seems
like almost all the "piracy" preventing things are self-defeating. There are
a certain percent of honest people out there who want to support their
suppliers. Well, there are free riders out there too. But focusing on the
free riders (or free-rider wannabees) often then loses the honest folk who
believe in commerce and exchange.
Of course, I obviously believe in commerce and exchange when I plunk down a
credit card in the grocery store. Or pay my utility bill. :-) Not that I
have that much choice. It would take a lot of resources to cut down trees
around our house and rebuild the soil, too. Gardening can be pretty
expensive these days. Still cheaper than the store sometimes, but it's still
not free of cost.
> 3 - Meanwhile, in the free software arena, you'll lose credibility for
> this odd approach.
Well, maybe. Maybe not. That's something I'm exploring. There are a lot of
free software developers that also write proprietary code (often for a day
job). I'm not sure it is so black and white there. And there are no doubt
people stuck in the same situation, looking for win/win alternatives.
> Do you have a copy of "Here Comes Everybody", by Clay Shirky? He
> describes the mass amateurization of some professions in the modern
> connected world (journalists and photographers are the two examples he
> gives). I think packaged client side software is slowly being eaten by
> mass amateurization (hooray! what an amazing development for the
> world!), and is a bad place to be looking for commercial income at the
> moment.
Great book, and I'll generally agree with what you said. And what the
professional amateurs are not eating up, Google and others are. :-)
Still, while it is easy in the free software community to see all the free
software, it does seem to me there are still a lot of people who pay for
specialized software, especially if it has a direct business use. If it
saves time, and there is no obvious alternative that is as cost-effective,
it is a good business investment. It would just really be foolish
economically not to have it. That's the economics of that sort of thing.
Example: :-)
http://www.shopbottools.com/software.htm
I'm thinking, maybe something good enough to complement those somehow? :-)
On the other hand, commercial software incurs overhead in marketing,
protection, distribution, management, return on capital wagered, and so on,
all of which drives up the cost and can diminish the quality (in terms of
responding quickly to user feedback once there are several levels of
management or sales people involved). So, "free" stuff by one or two
developers has a lot going for it as a production method.
Pros and cons on both sides, from a business point of view. Most good
business people want their suppliers to stay in business in the long term,
because switching suppliers can be painful and costly.
> One of the best indicators of this is that you'll find
> yourself having to do all kinds of nasty protectionist things in order
> to make any money, and I suspect they only work for the big incumbents
> in any case, and even there probably not for all that much longer.
Nah, I'm not too worried about any of that, beyond what the license says.
Still, there is the hidden aspect of enforcing a license, and creating
felons. One reason we stopped doing shareware is I did not like making
criminals out of people who did not pay for the software past the trial
period. So, there is a dark side even just to a proprietary license.
I don't have any serious worries about "piracy". It's getting attention that
matters as an indy developer. I'm not even worried too much about the code
being out there. Any experienced free software developer learns not to look
at code until they've figured out the license. And even without code, for an
experienced developer, most of many applications can be reverse engineered
on inspection just by looking at the menus and dialogs and windows for
people who are so inclined to copy things. There may be still big hurdles,
but enough is learned from inspection to make it a lot easier to build a
clone than the original in the sense of knowing what basic functionality is
essential, deciding on which is most of the trouble of writing an
application. Having the code may sometimes actually slow somebody down if it
is not in their preferred language and changes are planned. :-)
> Remember when there used to be Shareware? Ah, the 90s. I don't really
> miss it though :-)
That's probably generally true, but some shareware is still thriving on the
Mac, it seems.
Growing more common now for games seems to be a web-based version with a few
abilities (flash or Java), and a pay-per-download version.
> All the software I see these days that has a hope of making money has
> a server based component. Server side gives you control, which is good
> for you. It also means extra options for the users, which is good for
> them.
>
> Some software lends itself to this better than others. Facebook, yes.
> Image editing software, not so much. Your project, maybe it closer to
> a client side thing than a server side thing? I'm not sure.
Oh, it isn't defined well. It could be anything, even as I have some ideas
in mind I wanted to do back when the PlantStudio software was under
development ten years ago.
Still, running a server is a different thing than writing software. And it
is a big commitment and a bunch of stress to keep anything running 24X7
(having done that. :-) It's a big commitment to put up a service. You can't
in good conscience walk from it a year later if something else comes along.
So, it is in a way a different sort of ethical quagmire. :-)
Basically, what you seem to me to be saying is run a service of running a
server, one I am an expert in running. Well, maybe that is not a bad
business to be in, even if it can be stressful for one person. So, I'll have
to think about what you wrote some more.
I can see that, if I was going to go that way to a server, yours is great
advice on integrity and honest business.
Actually, if I were to try to follow your advice, doing a social semantic
desktop might actually be a better choice than a 3D application. While a
social semantic desktop is by definition intended as a desktop application,
coordination services go through the network, and may well work most easily
with coordinating servers or other network services.
So, maybe you've rescued the social semantic desktop option for me? :-)
But actually charging can still get a little fuzzy. Like getting people to
pay a monthly fee for some access to something or other?
> btw, you get excellent trust by making the server side free, because
> you address the "what if this commercial service folds?" problem.
> People know that at worst they can always host it themselves, so they
> feel safe using your commercial service (oh remember to provide import
> and export functions).
Yes, I agree. My wife is trying this with her Rakontu project, with hopes to
provide related consulting on it.
Thanks, Emlyn, for all the challenging ideas. I'll need to think about them
more. I appreciate the gift of your time in trying to come up with
innovative win/win solutions in this area.
--Paul Fernhout
http://www.pdfernhout.net/
Emlyn
>
> Thanks, Emlyn, for all the challenging ideas. I'll need to think about them
> more. I appreciate the gift of your time in trying to come up with
> innovative win/win solutions in this area.
No problems. Oh and there's something I forgot to mention, unrelated
to the server based model.
If you are the copyright holder, you can release under GPL and also
provide commercial closed source licenses for those who want to pay.
So if you think there is a commercial market for the software, release
it this way, let individuals and open sourcers use it under GPL as
they see fit, and provide a way to pay decent money for corporates who
feel the need for closed source and support.
If, for instance, it is extensible with plugins (say representing
particular types of equipment, or processes, or whatever), and to
create them you have to consume libraries, then the GPL will extend to
those libraries I think. This means that corporates who want to create
their own closed proprietary plugins are going to have to shell out
for closed licenses.
Samantha Atkins
- samantha
Eugen Leitl
>
> Paul, I largely admire you but you are way over-wordy. Do you hear
> it now or to you wish to write more pages (plus references of course)
> in response?
and not top-posting? Thank you.
> - samantha
>
> On Tue, Jul 14, 2009 at 4:32 PM, Paul D. Fernhout
> <[1]pdfer...@kurtz-fernhout.com> wrote:
>
> Trent Waddington wrote:
> > On Wed, Jul 15, 2009 at 1:07 AM, Paul D.
> > Fernhout<[2]pdfer...@kurtz-fernhout.com> wrote:
> >> Thanks for the feedback.
> >
> > FYI, I read about 1/3 of your email. Long and rambling seems to be
> > the norm on this list, and there's only so many hours in the day.
> >
> > Folks, can we try to get the signal to noise ratio a little higher?
>
> And I thought your comment about making email posts the size of a
> text
> message (160 characters) was a supportive satirical joke. :-(
> You must be new around here; it's all noise, at least my part of
> it. :-)
> distinction. One
> person's maximum-volume music listening pleasure might be another's
> unbearable noise. An annoying background hiss interfering with
> short-wave
> radio broadcasts was found to be due to extra-terrestrial, indeed
> cosmic,
> processes; listening to this "noise" to the exclusion of all other
> signals
> with ever more sensitive antennas and receivers is now the science
> of radio
> astronomy. Radio astronomers are still plagued by noise in their
> signals
> but now it is thermal noise generated in their equipment
> interfering with
> wanted signals from the cosmos."
> What matters most sometimes is how we creatively shape the noise in
> our
> lives into personal meaning. :-)
> Still, this idea might help; maybe you could help build it into the
> Linux
> kernel? :-)
> streams
> like emails on idle cores to automatically build a summary or flag
> text you
> wanted to pay extra attention to? :-)
> Seriously, better software tools might help somehow, though not
> soon, and
> probably not in kernel space. :-)
> We all make our choices in life on how we want to spend our time.
> Anyway, thanks for reading some of what I wrote.
>
> --Paul Fernhout
>
> >
> References
>
> 1. mailto:pdfer...@kurtz-fernhout.com
> 2. mailto:pdfer...@kurtz-fernhout.com
> 3. http://en.wikipedia.org/wiki/Noise
> 4. http://www.google.com/search?hl=en&q=
> 5. http://www.pdfernhout.net/
--
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A 7779 75B0 2443 8B29 F6BE