[greater good pledge] request for feedback

[vsoch]

hey @pydicom/pydcom-fff @pydicom/pydicom-core !

I'm creating a new initiative, The Greater Good that aims to highlight projects with no underlying incentives or ulterior motives - those that are "for the greater good." The Pydicom family of tools (I believe) fits strongly within the criteria, and I'm hoping that the community would like to participate. Participation simply means adding the community name to the pledge, and then (optionally) adding a badge to any associated project repos.

Details are provided at the page, and any maintainers / users here are welcome to also join the organization The Good Labs if open source transparency is important to them. It's still early (I'm going to be creating media and other community interactions) and right now I'm getting in touch with projects that I've worked with.

[darcymason]

Hey @vsoch,

I think this has noble intentions, but I have mixed feelings about it, difficult to express, but I'll try...

This is a very quick reaction, so perhaps it is an interpretation of wording.

I understand where this is coming from -- coroporate interest has changed a lot of the earlier greatness of the internet. I just read an article today "Who's behind the smiling faces of some Airbnb hosts? Multimillion-dollar corporations". I think that is in line with the direction of this effort, to try to maintain transparency in open source.

But I personally wouldn't want to take this pledge for pydicom. Here are some reasons:

• it would seem to exclude "freemium"-type offerings. I personally think this is quite a valid model and should be encouraged, if it is done well. If the free offering does really provide good value, I think it is actually a good thing to "reward" the developers monetarily. I subscribed to Duolingo plus for a year, for example, to encourage development of what is for me a great product (not open source, but the point holds just the same). We have no premium pydicom (for more than 10 years), but I wouldn't want to exclude that -- a case might emerge where some special features might be of interest to a subset of people, and if complex enough to develop, or of interest only to a few, those features might not ever get done without remuneration.
• I don't see how it could be "enforced". It really relies on a declaration from someone. Even if they say so, that doesn't make it so, or they can twist the meaning a bit. Any violation would have to be exposed through an information campaign. But I think that can happen now, similarly to the airbnb article I linked.
• there is a danger of sorts from people abusing free service. If we had 50 requests a day to help with someone's specific problems, we wouldn't be able to handle that, even to triage them to decide which are valid open source questions. The answer for some requests might then be, "pay us to consult with your particular situation" and then "free" requests would naturally decline, I think, and migrate to ones truly related to the open source project. And as a paid consult, we could help people through much more complex problems than we could do on a volunteer basis. That's not a bad thing at all; both sides got something of value.

I know there could be an argument that we pledge to start with, and simply recind if/when a premium thing (or whatever) happened. But I don't want to really have to think about that or judge every action we take based on that. And it really just weakens the whole thing for anyone looking at a particular open source project - what does the pledge actually tell you, if it can be taken away at any time? I would pay much more attention to the history of the project, what people have actually done rather than what they have said.

Aside from pydicom, the pledge to not indirectly profit would leave hardly any company left to sign this pledge. Companies that release or contribute to open source indirectly profit. They are using that open source themselves for their product line or internal testing or some other reason, or to try to make 'their way' the defacto standard (to reduce their costs).

Bottom line, I think it is too absolute a declaration - there are so many situations where some kind of "profit" could actually be beneficial for an open source project, giving value to users willing to pay, while giving oweners/contributors incentive to keep working on it, rather than moving on to their next interest. And the only way to truly gauge a project is by what they do, not what they say.

[vsoch]

hey @darcymason ! You are correct that there are many things you could read into - but the easy question to ask now is "Does my project currently exist for the greater good?" This is mostly to say that your project is in spirit of being for the greater good - you are free to change your mind, receive financial help, and it's not any kind of contract or promise. I wrote up some FAQ this morning that might help to clarify - https://good-labs.github.io/greater-good-pledge/faq/ and please let me know if you want to see another question answered, or another example scenario. Now I'll address some of your questions.

• The pledge doesn't say anything about receiving funding to support a project. For example, support for developers, traveling for conferences, and talks are really important for community growth.
• For the Duolingo example, it's great that you supported them. The pledge doesn't make a value statement about entities that are so successful that they can offer a product. What I'd point out is that a smaller (similar) initiative, or generally, other open source projects on GitHub that choose to not go this route would not get the same attention as Duolingo, maybe simply because Duolingo can pay more developers, advertisers, etc. I agree that value should be given (by way of money) to all kinds of projects to send the message "I value you." This is sort of a different thing than the incentives behind a project. It could hold true for projects that do / do not sign the pledge.
• it's not enforced, and it's not a promise. It's a best effort to say that a project is in spirit of being for the greater good. If things change? That's okay. There definitely won't be any public shaming or removal, that sounds terrible.
• You don't need to look at the details of every action that you would take on behalf of pydicom. It's really a simple question of the incentives behind creating it. Did you create it because you wanted to start a company and sell software? Or did you create it to primarily help researchers with reading dicom in Python. I'm fairly sure the latter.
• You are right that pledge implies "forever promise" and that's not the case, for a human it's really saying "This project is in spirit of being for the greater good." Similar efforts I've encountered might be along the lines of pledging to wear a helmet when I bike, or to recycle. I can't promise to wear a helmet, every time for the rest of my life, or that I'll never be wasteful again, but by signing it's saying that I have awareness about these things, and will do my best. Is there another term / way to phrase it (instead of pledge) that you think would fit better?
• You are correct, the pledge could not be taken by most companies. Most companies have open source strategy teams that directly ask "How can we profit from open source models." This isn't bad, it's just a different beast than a project that grew organically and became popular out of community value / need,
• Getting some kind of financial support for a project because it has value is a great thing. Providing software, and being rewarded because it's good, is a different thing than providing the software because you want to charge your users. Again, it's about spirit and intention.

I am thinking that Pydicom fits in the spirit of the greater good, but I want to address your concerns. If it's too absolute a declaration, how can I make the criteria more clear? What language / wording can be changed, or questions answered? We can discuss here, or if you want to bring up points about the criteria, you can open an issue directly https://github.com/good-labs/greater-good-pledge/issues. Thanks for your thoughtful response, I really appreciate it.

[darcymason]

Hi @vsoch, thanks for the detailed replies. I'll have to review your arguments in more detail, but here are some thoughts:

Maybe it is because I am one of the 'good ones', if I can say so, that I fear having to periodically self-judge. For sure pydicom was originally released for the greater good (after I thought a little about a commercial angle, and figured I didn't have the time or energy to give the level of commercial service I would want to. so why not let others benefit from it). But let's say I started getting all kinds of offers to consult. So far, so good, my primary intention is still for the greater good... but then let's say it becomes full time, then more than full time so I hire two people and form a corporation. (BTW, I wouldn't ever actually see this as a possibility, but it is by way of example). Now I have a commitment to the health of my company, so I could still say my objective is for the greater good (and really believe that), but clearly there are also other demands in my mind as well.

In that scenario, what I would say is much more important to others looking at using pydicom, would be the potential conflict of interest. Each of the owners/maintainers (at least those) could simply say whether they have a potential conflict of interest. It is a relatively clear and measurable thing - 'I make money supporting this software as a consultant', or 'I own/work for a company that makes money supporting this software', 'I get travel costs to speak about this project'. Then people can make their own judgments about how much/little that seems to be influencing the direction of the project. BTW that would have worked in the AirBnB story linked in my former comment - if people had simply been informed that the 'host' was actually working for a company, there would have been no news article to write.

I think conflict of interest is a well-used concept in the scientific world and should serve as a model. People don't start a talk saying "I confirm that my main intent in this work is to promote general human knowledge", they say "Disclosure: I consult for company X which has a financial interest in this work". The latter to me is much more transparent, and provides much clearer information for others to base their decisions on.

[vsoch]

That's a great idea! So if the pledge had criteria about disclosing conflicts of interest, that would be more friendly to different kinds of projects? I assume that we would need to edit the statements about profits to be in line with that. For example, here is throwing out an idea:

The Greater Good Pledge

As a representative of an open source project or group, I affirm that
the community to which I belong practices transparency, and makes a best effort to exist
for the greater good. Specifically, this means that:

• Any direct or indirect profits from the software as a result of the project, community, or supporting company providing a product or services are disclosed in a FUNDING.(md|yml).
• Conflicts of interest for one or more maintainers are listed in a CONFLICTS.(md|yml)
• The majority of contributors represent dispersed community (and not required participants for a company)
• The community strives for open communication for discussion around development.
• Software is provided under an open source license, with no private distributions or versions with special privileges.

By taking this pledge, I am committing to a continued best effort to maintain this
transparency of communication, and incentives. I understand the spirit of the initiative, and am free to change my mind. I affirm that the open source project or community that I represent exists for the
greater good.

Including @pdurbin on this for feedback, I like the direction of lessening the severity about profits, and instead focusing on disclosure.

[vsoch]

If it helps, here is a draft that you can edit (suggestions mode so I can see!) directly, or comment on points.

[bnmnetp]

@vsoch thanks for inviting me to the discussion.

I like the disclosure model as well, and my path to where I am today has taken lots of twists and turns that may make for good discussion or help clarify.

For Runestone Interactive my path has been:

1. Start the project during a sabbatical year, as a side project and in my desire to create a better model for writing books for CS education.
2. Learn that most profs and especially high school teachers do not have the desire or ability to spin up a server to run the back end of Runestone. So that led to what is https://runestone.academy
3. Watch in amazement as the number of schools and students using Runestone continues to grow. Need money to pay for hosting so start a small campaign to accept donations through Paypal, or more recently, Patreon.
4. I got so focused on Runestone that when I had a chance to take an "early retirement" incentive from my college I did so (I'm not even 55), with the idea of finding a business model that can make Runestone sustainable even after I'm done with it. I'm not looking for profits, I'm just looking to pay myself a decent salary, cover hosting costs, and in an ideal world offer a chance for students to get internships to help with development of Runestone.
5. The first step down this path was to create an LLC -- In our litigious society this is pretty important. I don't want to end up bankrupt if someone hacks the system, and releases Johnny's grades, and then Johnny's parents sue me because he doesn't get a job with Apple. I should have done this step as soon as I started running a server and allowing people anywhere to create classes on it.
6. In my first year of non-retirement, Google approaches about using the platform in an educational initiative -- bringing CS to small colleges that have never offered CS before!! They want to pay me to consult. And are not charging the schools. Cool! But probably not a long term business model, and probably not a long term relationship. Support for Runestone is part of several outstanding grant proposals with the NSF and others.

So, at the end of the day, I firmly believe Runestone exists for the greater good, (http://reputablejournal.com/LifeLongLuther.html#.XMnqmJNKjUI) and I have the numbers to back that up. But existing, and continuing to exist does require some type of business model, and I'm more than happy to disclose how that model works and where the money goes. Building a community takes a lot of work, and I'm happy that I can do that, and so much the better if I can do it full time because I have a way to get funding.

With respect to transparent communications, We are open source, and most of our conversations take place on github. One initiative that I'm trying is that we are having some bi-weekly video hangouts. I'm happy to have anyone join in the hangouts but we don't publish minutes or anything like that, so one could argue that is less than transparent.

[suever]

Just wanted to chime in here with some thoughts.

• Typically these sorts of initiatives are taken on and accepted by a project due to them resonating with the team somewhat organically. This pitch feels a little bit like using pydicom to further the exposure of the Greater Good Pledge (not that that's a bad thing). Is there not a better arena to do so? I appreciate the information being provided here, and it's certainly something to consider, but it may be best to give Darcy (and other interested parties) some time to chew on it and decide for themselves whether this truly resonates with them and is something that we want to adopt for the project.

• Personally, I don't think people select a particular software package because the people working on it are great people with disclosed conflicts of interest (although great people behind a useful project is a huge plus), they use software because it solves a problem for them and carries a license that is compatible with their use-case. Do we really care if Siemens or Phillips are paying people to actively develop pydicom to be more than it is? The licensing of pydicom is what ultimately protects it and keeps it "for the greater good", is it not?

• Sometimes "conflict of interest" is a little grey as I may contribute to projects in my spare time that my employer could possibly benefit/profit from, but they aren't telling or paying me to do it. Would we really have to declare that? It seems like a hurdle to contributions which all projects need to survive and thrive.

I would suggest we wait to see wider adoption on projects outside of pydicom and your own projects to see how it is received by the community at-large (I only see three projects so far). Additionally, I would suggest working with various distribution channels to get the word out about the Greater Good Pledge so those with whom this resonates can sign the pledge and demonstrate this broader support. I definitely think it's a great idea, we just need to make sure that we're letting projects choose it for themselves.

[darcymason]

@bnmnetp's story is quite instructive - it does show how quickly things can change and take on a money angle of various kinds.

I appreciate the comments @suever just made, and maybe it is better for pydicom to sit back somewhat. I'll think on that a while. But I will say for me I'd be much more likely to sign on with a focus on disclosure more than statement of intention.

[vsoch]

@bnmnetp after hearing your story I'm thinking that the disclosure model is a better fit. It's also a good example of acting as a business, for very good reasons, but being for the greater good. Your project is exactly the kind that I'm wanting The Good Pledge to support!

@suever thanks for your feedback! I didn't mean to force anything on anyone - this particular kind of project "for the greater good" is important to me, and instead of watching passively I want to do something about supporting said projects. I'm new and green to the "How to create an initiative" and it seemed like a logical first step to reach out to projects with the pledge and ask for feedback (which has been fantastic, by the way). I'm hoping to get enough feedback to draft some early version 1.0.0, and then share the Good Labs more publicly (Twitter, etc.) to allow it to grow as you mention - the project owners finding it and not the other way around. You're also correct that people find and use software because they value it. But there are cases when software that isn't super great gets attention and developers artificially because they, for example, go on Twitter and see that it's the trendy thing to be working on. Is this a bad thing? Maybe not, if it brings people to improve something to a point of it being much better, it's great! I don't want the Greater Good Pledge to make value statements about who does / does not sign the pledge. It's just turning on a small signal of awareness of having transparent communication and incentives. It's probably a much simpler thing than you are imagining.

Do we really care if Siemens or Phillips are paying people to actively develop pydicom to be more than it is? The licensing of pydicom is what ultimately protects it and keeps it "for the greater good", is it not?

Nope! We totally don't. And yes, the license and separation from those companies do (somewhat) protect it. Now, it would feel very different if Philips released an open source Python library with sole intention to make profit, hired 10 full time developers to work on it, hired an advertising company to create branding around it, and then advertised it as a good-feeley, non-selfish open source initiative. Is that the truth? Not really - it's an example of hijacking open source culture for a non-disclosed, selfish goal. This is the kind of thing I would want to promote with this initiative. If Philips did that? They would need to disclose the funding, conflicts of interest, and really engage the community to get contributors outside of the company.

Sometimes "conflict of interest" is a little grey as I may contribute to projects in my spare time that my employer could possibly benefit/profit from, but they aren't telling or paying me to do it. Would we really have to declare that? It seems like a hurdle to contributions which all projects need to survive and thrive.

The conflicts of interest are with respect to yourself and your relationship the project, not with respect to your company. I am not super familiar with the academic "conflict of interest" definition (is there one?) but I think If you contribute to projects on your spare time, for personal interest or other, that's your thing.
@darcymason could you give an example of what you would consider a conflict of interest?

Additionally, I would suggest working with various distribution channels to get the word out about the Greater Good Pledge so those with whom this resonates can sign the pledge and demonstrate this broader support.

This is in the plan! This first week, I'm just reaching out to communities that I've interacted with. It's definitely under 10.

I definitely think it's a great idea, we just need to make sure that we're letting projects choose it for themselves.

Huge +1. Thanks for the feedback!

For now, don't worry about making some decision to include Pydicom, this is more a discussion about the criteria, and how this original draft should go out. I (still) think Pydicom would qualify, so I'm genuinely interested in any/all feedback. You've already given a lot. :)

[vsoch]

hey @darcymason @bnmnetp @suever we've done a lot of updates to (what is now called) the Greater Good Affirmation, and I'm hoping I can again ask for some feedback! Specifically:

• it's no longer a pledge - there is nothing to sign or promise.
• it's been hugely simplified to focus more on transparency and incentive
• It's an affirmation that a project or organization can take in good faith, and all this means is affirmation that a/your project is for the greater good by adding a badge to your repository.
• I've updated the About / FAQ to reflect this

What do you think?

https://good-labs.github.io/greater-good-affirmation/

[darcymason]

Hi @vsoch, I've only had time for a quick glance, It certainly looks better, ... but I will have to get back to you with more detailed feedback.

[vsoch]

sounds good :) Happy Friday, @darcymason!

[mrbean-bremen]

(moved to discussion, in case this is not shown)