Currently on a team that handles ~225M requests per day on a single endpoint and we have dozens of endpoints with similar volume (+/- 50%). We haven't landed on a solution yet but are currently exploring the following:
> All data is offered for non-commercial use. With speedy APIs, all data can be served by just a couple of virtual machines for less than a coffee a day.
Very impressive! What does your backend stack look like? Are you using any caching or does every API call hit the binary file?
> The project went live 2 weeks ago and is slowly being used.
The API backend is written in Swift with some code in C (conversion Float16, interpolations, some solar radiation libraries). Data are read directly via mmap from local NVMe SSD disks. To scale, I can add new VMs and copy my binary files.
There is no frontend cache and I want to keep it this way. Next steps include point-of-presence API virtual machines in different counties. Via GeoDNS this is even faster than a CDN.
Peak API usage was around 50.000 API calls per hour, but on average it is quite low.
I do not actively prevent commercial use. I stated a non-commercial license as precaution to prevent all kinds of legal complications.
Mainly I also intent to reach open-source developers to make access to weather dat easier without the need to instruct users to sign-up for an API key somewhere.
It was clear from the start that SARS-CoV-2 would become endemic. Very interesting write up + data here.
> Cure-Hub's data shows less antibody diversity in vaccinated individuals than naturally infected individuals. Fewer unique antibodies provide the virus with an avenue for immune escape through S protein mutations.
The data compares natural immunity vs vaccinated immunity. I'm curious to see this data compared against individuals who were infected and were then vaccinated vs. individuals who were vaccinated then infected.
I have no primary sources to offer regarding this question.
But here's some pure speculation if you're interested - assuming immunological memory operates in a 'most recently used' fashion, then it would follow that individuals will develop an immune response that trends toward being most effective at neutralizing the viral profile that was last seen. Consequently, the studies and data that evaluate natural immunity vs vaccinated immunity could be fairly applicable. In practice I'm sure you'd see a blending of the two immune responses, but timing of exposure could play a big factor in weighting that balance.
*Incredible* product. I signed up for the monthly membership and, if all goes well, I will likely stick around or buy the lifetime membership.
I currently use Undebt It[0] to manage my debts. It supports all kinds of debt payment plans like snowball, avalanche, cash flow index[1], etc. One feature I love is the "extra debt payment" that basically throws an addition X dollars at the highest interest debt (or whatever debt is prioritized by the selected repayment plan). When that debt is paid off, the amount rolls over to the next debt on the list. This helps a person to pay off debts faster and reduce total interest paid across the lifetime of the debt(s). For a family with a few credit cards, a car payment or two, and a mortgage this can save them tens of thousands of dollars.
Three suggestions:
1. Create an "extra debt payment" feature that goes towards my prioritized debt(s). This amount is in excess of the total minimum payment amounts. It could be towards the highest interest rate, lowest balance, split across all accounts, etc. When a debt is zeroed out the payment automatically adjusts to target the remaining debt(s). Example: two debts, each of $500/mo and an additional $100 "extra" per month for a total outflow of $1100/mo. When the first $500/mo is paid off then new outflow is $600/mo.
2. For each debt "account" add an option to "roll over" the monthly payment when the debt is paid off. This amount would then target the next prioritized debt amount. Example: two debts, each of $500/mo for a total outflow of $1000/mo. When the first $500/mo is paid off the "extra" $500/mo targets the second debt, for an unchanged outflow of $1000/mo.
Combining #1 and #2 example: two debts, each of $500/mo and an "extra" payment of $100/mo for a total outflow of $1100/mo. When the first $500/mo is paid off the "paid off" $500/mo targets the second debt (as well as the "extra" $100/mo), for an unchanged outflow of $1100/mo.
3. Stack the icons on the chart. The vertical scaling of the chart is wonky when adding a 10+ items all in the same year - this makes the chart seem squashed and difficult to parse.
Again - GREAT work. Definitely sent this to a few friends. If you don't respond I will shoot you an email in a few days. I'm sure you're very busy with the new release. Good luck!
Thanks for the thoughtful response! And sure, feel free to continue the conversation over email if you like.
Currently, when you create a debt item or financed asset "X", there should be an "Extra X Payments" financial goal automatically generated, which you can click and drag higher up the list if you care more about allocating extra available yearly income towards that than certain other priorities. Is this feature working for you currently? And how close does it come to addressing / approximating your use cases above?
