But..it's $1k. This is basically pocket change on an institutional level. I've been part of some very scrappy and poorly funded community organizations and even they took in more than $1k every year. Even if you don't believe NTP maintainers should be paid anything for their work (an opinion I don't hold), it's trivial to spend this amount on modest everyday expenses like renting a venue a couple of times, buying insurance, and paying for hosting and technical resources.
It looks like they took in more than $200k and spent $100k on "contract services" (I can't tell what that means) and somewhat modest amounts on other things. Unfortunately I need to exit the rabbit hole now.
It seems a big waste of effort to maintain -say- a damnable Trello board with upcoming priorities and roadmaps <strike>and Kickstarter stretch goals</strike> when their bug tracker and mailing list are visible to the public. (Though, it seems that they've put the list behind some broken moderation software, so you have to go to -say- the IETF's archive of the thing to read it. "AI" crawlers ruin everything.)
I'm not sure why they'd try so hard to keep bots from paying them anyway. If someone wants to write a bot that constantly pays me good money I'm fine with that. I might rate limit it if the stream of payments coming in can't cover the cost of keeping the server from being DoS'd, but that's not going to inconvenience a human trying to submit a payment one time.
If you have small payments that can be made by bots easily, then your service can be used by thieves as an oracle to determine which of their stolen credit card numbers still work. Then you get lots of chargebacks to deal with.
I wish when accepting donations, websites would stop caching the total collected amount or give it a super short TTL. I like to see the little progress bar get closer to the goal thanks to my couple of bucks.
So we have NTP begging to raise a grand yet we have hundreds of billions being spent on AI data centers.
NTP might not be able to generate AI cat videos full of hallucinations but it is a vital part of web infrastructure. The same can't be said about today's mega projects.
That might work, but the second order effect would probably be companies trying to do the work of time synchronisation themselves in case it happened again. That would lead to fragmentation and incompatibility.
I am surprised that NTP project is not funded, fully or partially, by larger organizations or governments, given the criticality of the project.
It's not really clear why they need this money either?
> It's not really clear why they need this money either?
Really? The sentence at the top of the Donate page seems pretty clear to me:
> Your donation helps Network Time Foundation maintain the NTP website and provide resources and support to NTP developers.
Is it unclear to you?
It is kind of vague IMO. Especially since most of the actual NTP infrastructure is run by governments, universities, and companies.
https://gist.github.com/mutin-sa/eea1c396b1e610a2da1e5550d94...
But..it's $1k. This is basically pocket change on an institutional level. I've been part of some very scrappy and poorly funded community organizations and even they took in more than $1k every year. Even if you don't believe NTP maintainers should be paid anything for their work (an opinion I don't hold), it's trivial to spend this amount on modest everyday expenses like renting a venue a couple of times, buying insurance, and paying for hosting and technical resources.
EDIT: Here is their 2024 tax return
https://www.nwtime.org/about/documents/2024_NTF_IRS_990.pdf
It looks like they took in more than $200k and spent $100k on "contract services" (I can't tell what that means) and somewhat modest amounts on other things. Unfortunately I need to exit the rabbit hole now.
> It is kind of vague IMO.
How much more clear can they reasonably be?
It seems a big waste of effort to maintain -say- a damnable Trello board with upcoming priorities and roadmaps <strike>and Kickstarter stretch goals</strike> when their bug tracker and mailing list are visible to the public. (Though, it seems that they've put the list behind some broken moderation software, so you have to go to -say- the IETF's archive of the thing to read it. "AI" crawlers ruin everything.)
Why is research into the protocol useful. Isn't it done?
Time is hard, time synchronization is arguably harder.
The project isn't about research it's about creating a reference implementation
The Network Time Foundation (which counts the NTP project among those it provides resources to) lists several corporate Members.
But yeah, critical infrastructure usually goes criminally underfunded.
I tried to donate, but apparently I am not human:
> 1 error prohibited this submission from being saved:
> Looks like you are not a human
Good to know.
I'm not sure why they'd try so hard to keep bots from paying them anyway. If someone wants to write a bot that constantly pays me good money I'm fine with that. I might rate limit it if the stream of payments coming in can't cover the cost of keeping the server from being DoS'd, but that's not going to inconvenience a human trying to submit a payment one time.
If you have small payments that can be made by bots easily, then your service can be used by thieves as an oracle to determine which of their stolen credit card numbers still work. Then you get lots of chargebacks to deal with.
Then when too many of the fradulent payments get charged back then your payment processor drops you
What if that money comes from fraud
Well fraudsters need to have their time in sync for their business right? Who are you to deny their donations?
Is there a problem with fraudsters donating to OSS projects?
Money is money.
How do you know the cash you are using is not "blood money"? Come on.
I wish when accepting donations, websites would stop caching the total collected amount or give it a super short TTL. I like to see the little progress bar get closer to the goal thanks to my couple of bucks.
Perhaps they don't have the funds to implement that feature.
So we have NTP begging to raise a grand yet we have hundreds of billions being spent on AI data centers.
NTP might not be able to generate AI cat videos full of hallucinations but it is a vital part of web infrastructure. The same can't be said about today's mega projects.
The folks who run the public NTP pool really ought not to make it easier to pay them money to use it commercially.
I submitted a request for commercial use via their online form but never received a response.
Confusing. On https://www.nwtime.org/ they use $11,000 as “November 2025 goal“, with $4,675 as current level?
Are these goals monthly goals, with the counter being reset? The sites don’t make that clear.
Why not just turn it off and say we need money to turn it in again?
That might work, but the second order effect would probably be companies trying to do the work of time synchronisation themselves in case it happened again. That would lead to fragmentation and incompatibility.
Turn what off?
Time itself maybe, I know I could use with a little bit of a pause.
Return to the basement now. No escape.
pool.ntp.org dns resolution and any servers that they control, presumably
It's sad that a project that literally every company in the world depends on is requiring donations to keep working.
https://imgs.xkcd.com/comics/dependency_2x.png
There are always money and resources in ad tech.