explore relationship with USV

[chadwhitacre]

I've had a couple conversations with @andyweissman from Union Square Ventures (USV), and today we started directly discussing the possibility of a partnership between USV and Gittip. Let's move the conversation here from Twitter for the sake of deepening it.

The two questions on the table are:

1. Does Gittip fit USV's investment thesis?
2. Where's the profit?

Does Gittip fit USV?

I suspect that "yes" is a possible answer, but of course it's ultimately for @andyweissman and his team to investigate and decide.

One specific question so far is, "What are the network effects?" And I need some help understanding how to answer it. Is this a marketing question, like, how do we take this thing viral? Are we looking for market estimations and growth projections here? Or what?

Where's the Profit?

I think the second question is going to be trickier to answer, because I've staked out Gittip as an open company, with these principles:

1. Maximize transparency.
2. Don't compensate employees.
3. Price to cost.

Pricing to cost precludes profit in the traditional sense, the thinking being that through Gittip we're creating a "push" marketplace for gratitude, such that the part of the price traditionally assigned to wage and profit in the exchange economy can now be reverse-traded elsewhere.

So we're going to have to think creatively.

Here's how I think it could work, in two stages:

• Stage 1: USV gives money away on Gittip. This would be up to a certain point, defined by dollar amount, time period, size of user base, etc. Having money available in the system incentivizes people to join and use Gittip, thus fueling Gittip's growth to critical mass. I believe we want to weight this money to new people. Like, we match up to $N per user. We want this money to help get people in the door, and then we want to wean them off of it fairly quickly if possible. As with one-off payments (allow for one-off tips #5) and pre-payments (Add funds at any time #113), this would need to be clearly separated in receiver's profiles and history, so they were well-prepared for the extra money to dry up eventually.
• Stage 2: Gittip pays USV back, with interest. This would be up to a certain point, defined by ROI, time period, size of user base, etc., and would take the form of a tax levied against activity on the site, parallel to how payment processing fees are handled now.

The key here is that the investment is in the Gittip community, and the profit is returned from the community. It's not an investment in Zeta Design & Development, LLC per se, and the profit is explicit in the pricing. Gittip is buying USV's capital just as we're buying Balanced's payment processing services. It's potentially an open, transparent win-win.

In fact, we could set it up strictly on an individual basis, as a loan to each person who joins. They'd be given the $N fairly quickly (probably not immediately), and then repay it with interest an order of magnitude or two more slowly. To the extent that we individualize it, we will probably need to make it optional. "Would you like to accept $N in matching donations from USV? Repayment terms are: _____."

What are the risks for USV?

If we match only recurring gifts, then the basis of repayment--the recurring gift stream--ought to be fairly resistant to erosion. That's why we're basing the site on recurring gifts in the first place, because we want an erosion-resistant foundation with which receivers can pay their mortgage. We ought to be able to structure any loans such that it would be difficult for a good-faith actor to take the loan money and then not be able to repay it.

A bad-faith or difficult actor could conceivably delete their Gittip account after receiving some or all of the loan money, thereby not repaying it. However, they are strongly disincentivized from doing so because, given a properly-structured loan system, they will have to have built up a significant funding base in order to receive significant loan money, and they won't be able to take that gift stream with them if they delete their account. The repayment terms should be gentle enough such that sticking around and continuing to collect the gift stream is worth it, even if it means repaying the loan with interest. We could also continue to maintain the former users payment stream in the event of a premature cancellation, funneling it now to repayment of the loan. Since all of the money would be available (the user having cancelled) the loan would be repaid much more quickly. We risk ill will from donors, however, who will already be upset to have been suckered in by the bad-faith actor in the first place. Being then asked to pay off the bad actor's debt would be difficult for donors.

What is the return for USV?

I don't know. As spec'd above we're basically talking about consumer loans. I believe we're in a different risk/reward ballpark than traditional venture capital investments. We knew this was going to be weird, didn't we? :-)

What about other investors?

Now, as I propose this I'm put in mind that @bradmontgomery has already expressed an interest in matching gifts on Gittip through his company, WorkForPie (see #208). In the spirit of openness and an inclusive table, we may need to devise this as a system that multiple investors can participate in, with fair terms all around. Eventually, it seems like people with a balance held in Gittip might wish to pay in to such a system as a way to invest their money without it leaving Gittip.

That said, @bradmontgomery is presumably looking to build goodwill for his company through his matching donations, and may not wish to be seen to participate in the extraction side of the equation. Or he may! I thought it only fair to loop him into this conversation, at least.

What Does the Gittip Community Think?

The most important question is, what does the Gittip community think of this idea? Do we want to introduce loans into Gittip? Are there other possible relationships with investors besides this one that could make sense? What considerations am I missing?

[chadwhitacre]

Personally, I'm conflicted. I don't want to water down the essence of Gittip, but I also don't want to be an impractical purist. However, the matter of Gittip's essence could actually be seen to be practical and not puritan, in that a muddy message is bad marketing.

The point of Gittip for me is gratitude. I want to live out of gratitude rather than necessity, if I can. Gittip for me is about metaphorically exciting us to a new electron state. I love #165.

I personally am much more turned on by something like the community chest (#84), where existing users would set aside money to be re-gifted automatically to new users. It would be a "Welcome!" gift from the Gittip community. Or maybe that's done as a special case of one-time gifts (#5). Whatever. The point is that rather than repaying a loan (gray, Dickensian, usurious), the new user is then culturally incentivized to pay it forward in their own time (Dinotopian color bursts!). Is it different if the loan is felt to originate from within the Gittip community? If it's attached to a person? Dunno, maybe.

I really appreciate @andyweissman for his willingness to engage Gittip and the concept of open companies. Certainly USV companies like Kickstarter and Etsy and Tumblr and Twitter are world-class teams doing amazing work. And surely there's a good bit of capital under management over there. I want Gittip to grow into something amazing, and venture firms like USV specialize in making that happen. If there's a way I can tap into that without selling Gittip's soul, I want to, as a matter of resourcefulness and doing right by Gittip by doing everything I can to help it grow. I definitely feel the lack of a full-time team. Here's App.net with a crowd-sourced $803,000. There goes Meteor.js with a top-shelf $11,200,000. I'm not funded myself on Gittip yet, and I'm juggling Gittip with other concerns. On the other hand, if you squint you can see a team of open source contributors starting to emerge amongst an engaged community, and of course we can look back at Linux and Python and Postgres and all of the other fantastic projects that were started on fumes. Maybe the best way to be lean is to be lean. Slow and steady. But really it's fast and steady that wins the race.

Gittip has already led to the development of open companies and open partnerships. Is there an open venture somewhere in here, too?

[canweriotnow]

Github really needs a +1 system for comments so I can stop leaving trivial comments.

That being said, +1 this^^^^

[bradmontgomery]

Thanks for the inclusion, Chad!

I haven't talked with @cliftonmckinney about any of this, yet, but I think it's safe to say that Work for Pie is interested in tipping out of gratitude (about 90% of our company is built on open source software), and we do hope that generates good will toward our company. Ultimately, we're building a company that has at least two goals:

1. Contribute code to the open source ecosystem
2. Promote people and companies that also support open source

With those two things in mind, it makes sense for our company to tip!

However, as an individual, I'm not at all interested in anything resembling a "loan". Incidentally, I'd prefer to re-gift any tips that I receive, so I may not be your target audience anyway. I've been much more a beneficiary of open source software than a contributor, so tipping gives me a way to support the people that have unknowingly given me so much.

[cliftonmckinney]

Work for Pie's perspective is that we've always seen our (forthcoming) contributions as something we're giving away not to get back. On the other hand, I'm also fairly sure our contribution will be significantly smaller than USV's. We're thinking something on the magnitude of $100 per month to start.

Our reasons for doing this are clearly twofold: we want to give back to the community that helps us exist, and we don't doubt for a second that it's pretty good PR. In our opinion spending money on Gittip sure beats LOTS of other places that money could go. I think it's a viable option for plenty of companies that benefit from open source, and I said as much in our blog post about it. :)

The question we're really trying to answer is how best to donate. I'm almost inclined to say that for donations like this there is a "blind" process from the donating company's perspective. Ideally, I think we'd prefer to donate a certain amount a month, and then some algorithm determines where that money goes. I definitely see the value in more money going to new community members to "get them started" as it were. In the beginning our thinking was that we might limit our donations to people who are already part of the Work for Pie community, but now that seems a bit like "purchase necessary to win," so maybe we'll avoid that altogether and just donate completely blind.

I welcome additional thoughts!

[hansent]

Not sure what I think about the "loan" idea. In terms of repaying...are you talking about repaying with money you would be receiving from tips in the future, or something I would be liable for personally in case gitip's stop coming in? In that case, I'd much rather just go to a bank or local credit union and get a loan I think...

where is the profit?

I think the the question where is the profit? is really the crux of the matter. My thinking on this issue sort of falls along two lines right now.

1. using a looser definition of profit:

   putting funds into gitip via investments/tips can be profitable for the giving entity by improving the software / ecosystem of software they are using. The thesis being: "by funding open source work this way, I will spend less or equal amounts of money, while ending up with something that is much better/useful than if I would go spend this money on a traditional closed product/company."

   I think traditional thinking or the main argument against it would be along the lines of: "but all this open source stuff is being done for free anyway..why should I 'donate' money if I get it for free." However, this is exactly where I think gitip may have the potential to be truly disruptive, because it challenges this assumption by essentially testing this hypothesis: If these people are producing all this stuff because they want to, in their free time, imagine what they could be producing if they were able to do it full time without worrying about making money doing something else on the side. (Given my personal experience of myself and other like minded people, I think it is a very valid question to ask. Coding is a creative process more than anything else, and "artistic" freedom can have a big impact on the outcome)

2. it's ok for an open company to make profit

   Chad's definition of open company includes 3 main points:

   1. Maximize transparency.
   2. Don’t compensate employees.
   3. Price to cost.

   I think the most important, and maybe only necessary point is: Maximize transparency. I guess for me, given maximum transparency, I don't care that much about the other two. I'm fine with people being paid for work they are doing, I'm fine with paying for a service I consume. If the open company is being completely transparent about who is getting paid for what, how it spends its money, how much of it is profit, and where that profit goes, I can make a personal decision on whether I want to support it or not. Such an open company could even be "forked," if all it's products, processes, and finances are "open source". If someone wants to make an open company that doesn't compensate employees and prices to cost, that's great. However I think these last two requirements make the concept unpractical for many situations, where a transparent company could otherwise excel. Who makes decisions, and how those decisions are made probably needs to be transparent as well.

   this would obviously solve the whole where is the profit? problem. i really think this could be a better way for companies to function. An open source company may a better way to run a business, in the same way open source is a better way to produce software. i think all the benefits from trade secrets or secreticve negotiations used in traditional business could be far outweighed by the confidence/trust it gives consumer/customer/employee/partner and the other benefits of community/network effects that come with complete transparency and openness.

[chadwhitacre]

@hansent Well put, thank you. I agree with your delineation of the two lines.

I think the most important, and maybe only necessary point is: Maximize transparency. I guess for me, given maximum transparency, I don't care that much about the other two.

I definitely feel a kinship with other companies that practice (1) even if they don't practice (2) and (3). It's been building in my field of vision even this week:

• Balanced has now opened up their API design process
• cf. App.net's API design intitiative
• Kalzumeus aka @patio11 aka BingoCardCreator publishes his financials--this is the closest I've seen to what I want to do with Gittip/Zeta
• Mixergy shows his financials
• "Transparency Life Sciences is the world’s first drug development company based on open innovation."

I learned about B Corporations when I floated the "Open Company" post. Without (2) and (3) it seems like we're more solidly in B corp territory, though there's more bureaucracy over there (for better or for worse--larger corps need the formalities).

If someone wants to make an open company that doesn't compensate employees and prices to cost, that's great. However I think these last two requirements make the concept unpractical for many situations, where a transparent company could otherwise excel.

Right. (2) and (3) are important for Gittip to keep Gittip's incentives aligned with its users'. If Gittip is truly going to become a place to find a livelihood then I have to find my livelihood there. But I think you're probably right that it'd be helpful to factor that out and assign it to Gittip uniquely, and let the idea of maximizing transparency bloom on its own. That is, I would be excited about and would welcome an #opencompany movement that dialed in on transparency and dispensed with the points about compensation and price structure. I would and will definitely participate in that.

Still not sure where the "hard" profit is in Gittip, though. Plenty of "soft" profit, but that doesn't help @andyweissman. :-/

[hansent]

Still not sure where the "hard" profit is in Gittip, though. Plenty of "soft" profit, but that doesn't help @andyweissman.

well, maybe (plenty "soft" profit) > (a little "hard" profit) ?!

edit: fix this post, and never reply by email using inline replies again [done --ed. ;-) ]

[andyweissman]

@whit537 this is a great overview of how you think about things, so first thanks for that. Now a couple of specific thoughts:

1. As to network effects, the question is, does the value of Gittip to a user depend - or increase - with how many other users there are utilizing the service. To me that's central USV "investment thesis" question. I think the answer is yes, but the discussion would be interesting.
2. As to the the open companies principles, they are admirable, big time. An open question, which started above, is, can an open company also make a profit. I think there are two ways to answer this, the first is, what do the principals of the open company want to do (that's a personal question). And the other, a more macro question, the answer to which is yes. In other words, as @hansent commented above, I believe you can be an open company and dedicated to transparency, fairness, righteousness etc - and make money. It is hard, but there is way for certain companies - those whose value is comprised of the contributions of the users, to forge a different path where the company fulfills a duty to its constituents - its shareholders/owners, employees, and users - in a different fashion. If you extend this out a bit, Gittip (like Etsy, or Kickstarter for example), could create a new ecosystem and business. That business could make money. That business could be valuable. It is there - at that point - that an investment firm like USV - would bemost interested. btw - lot's of open questions here, natch. But could gittip become the trusted intermediary for XXX (insert whatever XXX is, I am not sure yet). I guess one thing I am thinking is can you take open source product development principles and apply them to a company form or structure. And if so, what does that company do? Is Gittip the right place for this experimentation?
3. Finally, as to investments, USV as a firm can be thought of quite simply. We have investors, and our job is to make a return for those investors, and due to the risk we take that needs to be an outsized return. The primary way to do that is to participate in helping companies that transform entire industries, primarily by creating new markets and economies.
4. @hansent agree with this - "(plenty "soft" profit) > (a little "hard" profit) ." In fact, I think the role of new internt intermediary companies should be to take as little as possible - leaving most of it to the users. I also think that in doing so, with network effects, these companies can grow to massive influence and value.

[chadwhitacre]

Closing this as I'm pruning old issues.

I'm grateful for this relationship, but I don't see it flowering into a full partnership.

[chadwhitacre]

Reopened and reached out on Twitter. I heard from Jason Calacanis yesterday in private email, and while it's looking like that won't go anywhere (Jason's not ready for an open call; fair enough), hearing from Jason did get me thinking: we have Teams now! Teams could be the way forward for an "open venture." 💃

Scenario:

• We make it so any team member can add funds to a team account on Gittip.
• We add an investor to the Gittip team.
• The investor adds funds to the Gittip team on Gittip.
• The investor sets their take the same way anybody else on the team does.

What do you think, @andyweissman? Interested in an open call to unpack this? :-)

[chadwhitacre]

IRC

[hansent]

Teams is awesome, and I like your scenario. However, I'm not sure the way the teams feature works right now in terms of everyone setting their own take would produce the incentive for venture capital to use it as a funding model.

I like the idea of there being no difference really between the investor and other team members...everyone is a member contributing some sort of resource.

But for there to be an acceptable incentive for ventures to be funded like this, whoever is contributing the money (and --at least when dealing with larger sums of money-- maybe this goes for other team members in general as well) needs to have some legal protection with respect to promises about who will take how much money in the future.

Unlike e.g. a developer that contributes value to the project over an extended period of time getting resources paid out every week; the whole idea of venture capital is to put a lot of resources in first, and then get a return further down the road...so the risk is all upfront.

[hansent]

needs to have some legal protection with respect to promises about who will take how much money in the future.

I guess this is already the case with teams only allowing you to 2x your take every week. I feel like 2^n grows pretty fast though in terms of being able to cause some financial damage...but actually its at worst like taking out payment for the next one ahead of time, after which if trust was broken, the team could react accordingly...So maybe someone taking too much is not really an issue.

Being able not to take (enough) money out though may be a problem for the investor...what keeps the rest of the team from taking money initially, and then once the project starts making a return just kicking the investor of the team?!

[chadwhitacre]

@hansent All good points! Trust is certainly the basis of Gittip Teams. I see legal structures as a backstop to trust, a sort of baseline or lower limit. Seems like any legal structuring to a Gittip investment would occur outside of the scope of Teams proper? I think we could hammer out the details with a sufficiently motivated and creative investor. :-)

[chrismdp]

Ultimately I'm not sure how the investor relationship works with an investor who cannot guarantee a return, except through what the community chooses to give back. That's a lot of trust the investor has to place in a system. Are there any investors ready for this challenge? To get this off the ground some legal basis for the investment would probably have to be put in place. Perhaps Gittip might crowdsource a standard set of investment terms to allow investors to easily sign up with a great leve of confidence?

How about finding a way of running an experiment with a small (but not insignificant) amount of money, to see how the Team construct copes with the idea of an investor in a business, and how a community reacts?

[chadwhitacre]

except through what the community chooses to give back

The key with the Teams system is that everyone sets their own take(!). My proposal is that an investor could do the same, so they'd at least be on equal footing as the rest of team. Investors are part of the team! :-)

I'm not sure how the investor relationship works with an investor who cannot guarantee a return

Returns aren't guaranteed anyway, are they? Ventures are risky.

The main thing that's negotiated in an investment is the percentage of ownership shares. That's the basis of a financing contract. One option would be to add an investor as a team member at a fixed percentage. Do we need a second class of team member for investors?