Radar 43

[chadwhitacre]

What are you working on this week and why?

last week

[chadwhitacre]

Roadmap (what?)

Short Term

Embarrassments:

• can't edit teams (Team Basics)
• not using immutable ids (https://github.com/gratipay/gratipay.com/issues/835)
• widgets are broken (https://github.com/gratipay/inside.gratipay.com/issues/389#issuecomment-153884921)
• escrow imprecision (https://github.com/gratipay/inside.gratipay.com/issues/389#issuecomment-153929881)
• no payroll! https://github.com/gratipay/gratipay.com/issues/3433
• Team and ~user pages are the same https://github.com/gratipay/gratipay.com/issues/3852

Long Term

Color code:

• red—external forces that we're under
• yellow—administration
• green—product development
• orange—marketing
• blue—capitalization

[chadwhitacre]

[rohitpaulk]

I'm considering working on https://github.com/gratipay/gratipay.com/issues/3870. Do you think it's 'ready to start' @whit537?

[chadwhitacre]

@rohitpaulk Yikes! Okay ... does that help you in particular? Looks like it mostly helps Europeans, ya?

[rohitpaulk]

does that help you in particular?

Nope, it doesn't. We're relying 100% on PayPal and that seems dangerous to me - good to maintain other options, right?

Looks like it mostly helps Europeans, ya?

They're only US and Europe as of now, yes.

[chadwhitacre]

We're relying 100% on PayPal and that seems dangerous to me - good to maintain other options, right?

@rohitpaulk Yeah, definitely. If that's the motivation, another good option to pursue could be Dwolla (https://github.com/gratipay/gratipay.com/issues/726): they're US-only, but we've already cleared their compliance department.

[rohitpaulk]

Aha, well yh might make sense to do that first :)

[chadwhitacre]

@rohitpaulk Whichever, I'm happy to get the ball rolling on https://github.com/gratipay/gratipay.com/issues/3870.

[techtonik]

I mostly work from Windows nowadays, so my roadblock for everything is #480. Maybe I will get some hours to do something about it.

[Changaco]

Dropping by to wish a happy Republic Day to @rohitpaulk :-) (I know about it because the French president is attending this year.)

[chadwhitacre]

lol

@Changaco Why be so coy? Just ask @rohitpaulk to work on Liberapay with you, already! Sheesh! :stuck_out_tongue_winking_eye:

[chadwhitacre]

And a felicitous Lotus 1-2-3 Day to everyone! :-)

[Changaco]

I have no intention to do that, I'll stop posting here since I'm annoying you.

[chadwhitacre]

I'll stop posting here since I'm annoying you.

I'm pretty far from annoyed. Gratipay is in a good spot, we've got a lot of interesting stuff to work on, and I'm thoroughly enjoying it again for the first time in years. If @rohitpaulk or anyone else wants to work on Liberapay with you, that's no skin off my back. The worst-case scenario is that Liberapay manages to significantly overshadow Gratipay, in which case you may find me joining you as well. ;-)

Best of luck with the launch! :-)

!m @Changaco

[chadwhitacre]

Inbox 3(!), GitHub 1, L2 Support 0, Vendors, etc. 0.

[Changaco]

My bad for jumping to conclusions, I should have said "if I'm annoying you".

[chadwhitacre]

:-)

[chadwhitacre]

Nice-looking js MVC library: http://mithril.js.org/. Seems the best fit I've seen yet with the sort of scope for a library that we've gotten comfortable with in aspen and postgres.py.

[chadwhitacre]

Okay! With https://github.com/gratipay/gratipay.com/issues/3826 completed, we've just landed the Balanced Shutdown milestone!

!m * :dancer:

Quite a ride since this all started ... I'd say we were pretty spot on, actually:

[W]e will have to change Gratipay in some yet-to-be-fully-understood fashion in order to survive the transition away from Balanced, and it will be hard.

:flushed:

Seems like we could combine this news with an announcement about our new accounting system once we get through our meeting with the CPA on Thursday (#350).

[chadwhitacre]

Okay!

[chadwhitacre]

Inbox 2(!), GitHub 1, L2 Support 0, Vendors, etc. 0.

[rohitpaulk]

Dropping by to wish a happy Republic Day to @rohitpaulk

Thanks, @Changaco :)

[chadwhitacre]

Interesting development, Uphold (https://github.com/gratipay/gratipay.com/issues/3870).

[chadwhitacre]

We're relying 100% on PayPal and that seems dangerous to me - good to maintain other options, right?

@rohitpaulk I think that in the long run, the best way to reduce our risk of depending too heavily on other processors is to build our own AML program (#119), so that we can participate more directly in the global financial system. I guess my view has been that PayPal is "good enough" until we can get AML done and reapproach Transpay (#417) and/or banks (e.g., Citizens). Do you think the risks with PayPal are great enough that we need an alternative sooner than we'll be able to get AML done?

Remember, too, that most of what we need as part of AML overlaps with what we need anyway in order to bring back payroll (identity verification is the biggest piece).

[chadwhitacre]

In which case, Uphold gives us a really good example of what we need to build. :-)

https://uphold.com/en/security-protection/uphold-protecting-you https://uphold.com/en/security-protection/you-protecting-you https://uphold.com/en/transparency https://uphold.com/en/compliance https://uphold.com/en/help/for-law-enforcers https://uphold.com/en/legal/membership-agreement

[rohitpaulk]

Do you think the risks with PayPal are great enough that we need an alternative sooner than we'll be able to get AML done?

Yes, I do. PayPal could kill Gratipay, @whit537. Adding Uphold as a payout method is an order of magnitude simpler in our current scenario than getting AML done is, don't you think?

[chadwhitacre]

Adding Uphold as a payout method is an order of magnitude simpler in our current scenario than getting AML done is

Probably three or four, actually. :-P

PayPal could kill Gratipay

We have a ticket for pursuing underwriting with PayPal: https://github.com/gratipay/inside.gratipay.com/issues/305. I guess what you're saying is that we should have an alternative in place before we push on that. Yes?

[rohitpaulk]

I guess what you're saying is that we should have an alternative in place before we push on that.

I'd want us to have an alternative after too. We're bus factor unity on PayPal - and that isn't good for business. We should have an alternative in place so that we don't end up running around chasing one if and when PayPal decides to split on us. I'd make the same case for PayPal (or any other provider who seems worthy) if Uphold was our only payout method :)

[chadwhitacre]

The limitation with both PayPal and Uphold is that users are non-transferrable. If we add Uphold today and PayPal goes sour tomorrow, we can't transparently move our PayPal users over to Uphold, we have to have everyone sign up all over again. If we hold the PII, on the other hand, we can switch from one bank to another under the hood at will.

[chadwhitacre]

Folding the https://github.com/gratipay/gratipay.com/issues/3870#issuecomment-175729354 thread back into here:

in the glorious future we want to be able to have users give us their identity info and bank details, so that we can deposit to them ourselves (via Transpay or ACH) rather than having another brand in between. Right?

That is the goal, yes :) My aim here is to get us some stability while we pursue that goal.

But does adding Uphold really get us significant-enough stability, given that we can't transfer users from Uphold to PayPal?

[chadwhitacre]

In other words, PayPal going sour would be disastrous for us even if we have Uphold in place. We'll take a huge hit in processing volume—whether it's 50% or 100%, and Uphold is not going to be 50% of our volume anytime soon, given PayPal's much deeper traction—and we'll have to scramble to get people signed up again with Uphold. Adding Uphold on the spot when PayPal fails us adds an incremental amount of additional chaos to what will already be a chaotic situation.

[rohitpaulk]

Let's say we don't add a second provider right now, and Paypal leaves - What would we end up doing?

Would we -

1) Find another provider, work on the integration, and ask users to create an account with them?

or

2) Wait till we have AML in place and only process payments then?

If the answer is 2, then it probably doesn't make sense to work on integrating another provider. If it is 1 though - it does.

[chadwhitacre]

If the answer is 2, then it probably doesn't make sense to work on integrating another provider. If it is 1 though - it does.

Fair enough. Seems like that depends on how close we are to AML when "Gratipocalypse 2: The Revenge" hits. ;-)

[chadwhitacre]

PayPocalypse?

[rohitpaulk]

Adding Uphold on the spot when PayPal fails us adds an incremental amount of additional chaos to what will already be a chaotic situation.

The difference I see here is 'able to process payments' vs 'not able to process payments', and to me that's huge. We're a payments company, I really think that we should build redundancy into the foundations of our business.

When balanced shut down - we had a lead time of what, 2 months? And we still struggled to hit the deadline. If that happens again, we could be out of business for MONTHS

[rohitpaulk]

PayPocalypse?

Call it what you want, the fact remains :stuck_out_tongue:

[chadwhitacre]

I really think that we should build redundancy into the foundations of our business.

Agreed. The question is how to invest our precious development time: how much in building "weak redundancy" (Uphold), and how much in building "strong redundancy" (AML)?

We're a payments company

Yes, and we're also a payroll company—but we don't offer payroll! We're already out of business in that sense, and "weak redundancy" doesn't get us any closer to payroll.

[chadwhitacre]

@rohitpaulk I think we're basically on the same page about the bigger picture. What I'm hearing from you is that you're really motivated to work on Uphold to hedge against our eggs-in-one-PayPal risk. That's a real risk and you're right that Uphold is way more achievable than AML. I say go for it! :-)

[chadwhitacre]

@rohitpaulk What do you think about Uphold vs. Dwolla?

[chadwhitacre]

I mean, we can/should add them both, I suppose. Does it matter which first?

[rohitpaulk]

Another factor here is that I'm not sure about how much I can contribute towards AML, but with Uphold - pretty clear.

[rohitpaulk]

Dwolla is US only, AFAICT. Uphold is US + Europe, and they're expanding to China and India pretty soon

[rohitpaulk]

http://www.businesswire.com/news/home/20151014005723/en/Bitreserve-Uphold-Integrates-Global-Financial-Ecosystem

Phase 1: Live today: Members in 33 European countries can fund Uphold accounts via bank transfer; members in 30 European countries can fund accounts with credit or debit cards, in addition to bitcoin. Phase 2: November 2015: Members in the U.S. and China (limited beta) can fund Uphold accounts via bank transfer and credit or debit cards in addition to bitcoin. Phase 3: December 2015: Uphold will introduce both physical and virtual payment cards with Visa, MasterCard and Discover card (limited beta), allowing Members to pay merchants online or in-store, directly from their Uphold accounts. Phase 4: January 2016: Members in India (limited beta) can fund Uphold accounts via bank transfer and credit or debit cards in addition to bitcoin.

[mattbk]

If Uphold gets us closer to Payroll, I vote to add it as an option in addition to PayPal. If people really want to use Payroll, they'll get an account. If they don't want to get an Uphold account, we can work on additional routes in the future, but some Payroll seems better than no Payroll at this point.

[chadwhitacre]

If Uphold gets us closer to Payroll

I don't think it does, because for payroll we need to directly hold PII, for two reasons:

• In order to produce tax documents for The Enterprise.
• In order for The Enterprise to know whom to sue if ~picard or ~crusher go rogue.

[chadwhitacre]

I'm not sure about how much I can contribute towards AML

So let's talk about that! Maybe we are closer than we think? :-)

[chadwhitacre]

/me refers back up to https://github.com/gratipay/inside.gratipay.com/issues/479#issuecomment-174516999 ...

[chadwhitacre]

Basically, we need to start storing PII and checking it against Trulioo (#256). How can we break that down into pieces that you could start working on?

[rohitpaulk]

Well, baby steps :smiley:

[chadwhitacre]

!m @rohitpaulk :dancer: