Joffe Decks

Came across an interesting article today via HackerNews by Ben Joffe, who figured out how to stack a deck of cards so that regardless of how its cut, you are dealt winning cards. Seems like he figured it out by iterating over possible deck combinations until his algorithm found one that met the requirements. Pretty cool.

I do not condone cheating in poker, it is immoral, dangerous and often illegal, but out of curiosity I was wondering if it is possible to stack a deck so that no matter which way it is cut, the winning hand is always dealt to the dealer. This would be fun to use as a magic trick, to be able to consistently predict the winning hand.



The Universal Singularity

A good friend of mine, David Mangold, recently started an excellent blog called The Universal Singularity, which covers topics ranging from the implications of the discovery of water around carbon stars to ongoing efforts to build subaqueous human colonies.

Here’s a quote from his going series on his philosophy of existence:

“Cosmic evolution”, “biological evolution”, and “technological evolution” are not separate processes, but one in the same. This is central to the theory of Universal Singularity. While we necessarily create categories in the observed world to assist us in understanding it, they have little intrinsic meaning. Was there a fixed time at which cosmic evolution gave way to biological evolution? Or a time when suddenly biological evolution was replaced with technology? Of course not. Rather, the whole process is a continuum, a gradually changing system that tends to create pockets of ever greater organization and complexity. We humans, like the rest of the Universe, are the result of this process. Actually, “result” is the wrong word. We are this process. We are small bits of the Universe that through this process have gained a level of awareness that allows us to question our origins and our fate.

You can read more on his blog here.

AOL-Files Head Nod

I love running into folks that remember AOL-Files:

To be honest, my Dad gets the nod for buying me a VB book that included VB5 back in 5th grade (1997). But, I’ll have to say that you (with that AOL-Files.com site) and this guy named “Oogle” inspired me with the “hacker” curiosity by about 6th grade. In this case it was all black-hat though =)

From a HacherNews post about how folks got started in programming.

Good times.

Comparing Recurly, Spreedly, and Chargify

The following chart compares three recurring payment options I am considering for jMockups:

1) Paypal Website Payments Pro with Recurly or Spreedly

2) Authorize.net with a merchant account with Chargify

The calculations assume a recurring $8/mo payment from customers:

Click to view full size

You can download the Excel sheet here.

Recurly & Paypal Website Payments Pro Calculations:

  • Paypal is $30/month plus $0.30 + 2.9%/transaction for sales of $0 to $3,000; $0.30 + 2.5% for sales of $3,000 to $10,000; and $0.30 + $2.2% for sales of $10,000+ (source). For example, if I have 100 customers who are paying $8/mo, my monthly Paypal expenses are $30 + ($0.30 + 0.029 * $8) * 100 = $83.20.
  • Recurly is $29/month plus $0.20/transaction for up to 200 transactions/mo; $69 + $0.10/transaction for up to 500 transactions; and $199 + $0.09/transaction for up to 2,250 transactions/mo (source). (Note that if you’re not billing monthly, it becomes a bit more complicated because they also look at how many users you have in addition to the number of transactions — see their pricing page for more info.) So for 100 customers–assuming 1 transaction/mo/customer–Recurly expenses are $29 + $0.20 * 100 = $49.
  • Total monthly expenses for Paypal Pro + Recurly = $132.20. Total monthly revenue for 100 customers at $8/customer = $800. Total profit = $800 – $132.20 = $667.80. In this example, 16.5% of the revenue would go towards payment processing.

Spreedly & Paypal Website Payments Pro:

  • Speedly is a flat $19/mo + $0.20/transaction. For 100 customer, that works out to be $39/mo.
  • For Paypal Website Payments Pro and Spreedly the total is $83.20 + $39, or $122.20, or 15.2% of the revenue.

Chargify with Authorize.net and a Merchant Account Calculations:

These numbers will vary a bit based on the rates you get on your merchant account. These calculations assume a $0.25/transaction fee, a $9.95/mo monthly statement fee, a 2.19% Vista/Mastercard discount rate, and a $25 monthly minimum.

  • Authorize.net is $20 + $0.10/transaction (source). For 100 customers, the monthly cost would be $20 + $0.10 * 100 = $30.
  • For the merchant account, the fees would total $9.95 + $0.25 * 100 + 2.19% * $8 * 100 = $52.47.
  • Chargify is free for up to 50 customers; $49/mo for up to 500 customers, $249 for up to 5,000 customers and upward from there (source). For 100 customers, it would cost $49/mo.
  • Total monthly expenses for Chargify, Authorize.net, and this merchant account is $49 + $30 + $52.47 = $131.47, or 16.43% of the revenue. About the same as Paypal + Recurly or Paypal + Spreedly.

Final Thoughts

For an $8/mo subscription, you’re going to be paying 20% or more of your revenue to the payment processors until you reach 50-70 customers. That’s amazing to me.

For this comparison, Chargify wins hands down when you don’t have many customers because their service is free up to 50 customers. The more customers you have, the closer the options become. Also, remember that you don’t have to use Recurly or Spreedly with just Paypal Website Payments Pro — that’s just what I’m looking it; do you own calculations before making a decision.

It’s also important to note that the difference isn’t that much when you work it out. For 2,000 customers, the difference is less than $150/mo, but when you consider that you’re making $16K/mo at that point, things like customer service, how easy their API is to use, and what features they offer become a lot more important than the price difference.

Finally, if you notice any errors in these calculations, please let me know.