Hacker Newsnew | past | comments | ask | show | jobs | submit | more marcrosoft's commentslogin

I thought we all quit when they disabled API access and ruined Apollo. People are still using Reddit?


tens of millions a day, I sometimes wonder why people take on these false "people are still using __________?" when they know that people are still using twitter and reddit and facebook. Do you have any explanation as to why one would act pseudo-shocked? Is there a point?


(Not OP) I legitimately thought reddit the site had died and gone full ghost town. My use of reddit has almost always been as a knowledge resource, and that had been totally destroyed with these API changes.

Like, very often I have a technical question, search it+reddit and find a seemingly helpful thread with most of the comments deleted. It's logical to assume that most people who also engaged with reddit like this have slowed their use considerably.

Hell, lately I'm more likely to slap "hacker news" onto my search query


reddit is more popular than ever to be honest. Sure they lost some people over the "no more free clients" massacre, but it was a miniscule amount.


Apollo can still be used with sideloading a patch that allows you to enter new API key. As a single user it is hard to get to the limit.


If their latest “REST” API is an indication of their tech you don’t want to work there anyways.


Oh dear, I haven't used any FedEx APIs in a while. but when I did it was XML based (maybe SOAP) and the ordering of some of the fields mattered!! This was of course not documented anywhere.

Ahh, here's a great example of the mess it was: https://stackoverflow.com/a/40315400/137695


I clicked the comments section hoping to see sivers response because I remember him mentioning it before. I’m glad I did.


Use virtual credit cards and privacy.com. It really helps with peace of mind that you can really have a free trial and not worry about it.


Recent story: I had Duolingo, for some reason I've forgotten. They sent me a "we are about to renew" message.

So I went there and cancelled my account, and they even sent me a "sorry to see you go" message. They definitely got it.

Then they went and charged my card anyway. But it was declined, because I'd closed the privacy.com virtual card I'd given them.

That's why you do this.


Potentially stupid question: if you cancel the virtual credit card, can't the company then send your debt to collections resulting in a bigger headache?


What debt? Typically when you sign up for a service with a free trial, they charge you before each paid period. You get X days for free as part of the trial, then at the end of the trial they attempt to charge you for the next month of service. When that charge declines, they just don't provide you with the next month of service, and all they've given you is the free trial which incurs no debt.

If you're receiving a service at cost but agreeing to pay for it later, then yes they could send you to collections if you refuse to pay it, but that model is only really used for a select few services (some phone plans come to mind). The vast majority of online services are prepay.


How does one get virtual credit cards? I remember some banks like Discover offered them, but I don't see them anymore.


with capital one they have a browser extension called Eno that remembers which card is for which domain and autofills it.



Apple Card does this: https://www.idownloadblog.com/2019/04/01/apple-card-virtual-...

It’s not unique to Apple though. I think it’s this standard from EMVCo

https://www.emvco.com/emv-technologies/payment-tokenisation/


Apple lets you replace your card number with a new one, but it's not quite the same as creating single-use virtual numbers for sketchy businesses.


The EMVCo link is actually more interesting. The payment tokenization scheme means that the merchant never gets your card number (the PAN), they get a token.

> EMV Payment Tokenisation enhances transaction security by removing the most valuable data to a fraudster within a transaction, the primary account number (PAN), and replacing it with a unique alternative value, a payment token.

> This reduces the value of payments information stolen in the event of a data compromise, as a payment token should not be able to be used beyond the environment in which it was intended. Payment tokens support both face-to-face (F2F) and remote payment transactions.

Basically, if Amazon leaks my credit card data, thieves can’t use it because the number is associated with my Amazon account only. That one token can be cancelled and the next time I buy something a new one is issued and I don’t have to replace my credit card just because one merchant leaked my info.


Citi still offers it


privacy.com


no, this is bad advice . Credit cards are better because the dispute process is more favorable to buyers and longer dispute window. A debit card has worse buyer protection.


The virtual cards offered by Privacy are credit cards (specifically, charge cards) and feature the normal credit card dispute process.


I was about to post this. The usual HNers with nothing better to do will warn about how you're still liable, they can come after you, yada yada yada.

Ignore them. Just give a virtual credit card to any subscription service, and set a credit limit on it. Problem solved. If they try to keep charging your card: too bad, the charges are declined.


This is close to my stack which is a single Go binary, SQLite db and either OpenBSD or Linux with a side of htmx.


Much prefer your stack, yours operates at a higher level of abstraction, the one which I would consider correct for web services (your website or REST api doesn't need to do syscalls, or allocate memory manually), while not sacrificing too much performance or simplicity.


This has become my go-to stack for playing around with for the last few months. Go/sqlite(bun)/templ/htmx with a sprinkle of proto-actors. Feels pretty close to phoenix framework with a few helper functions honestly but with the benefit of incredibly fast compilation of Go.

Single binary for distribution with assets/migrations embedded. Still need to build something substantial so I am sure there are edge cases/rough edges but so far it feels like a breath of fresh air compared to nodejs ecosystem.


Do you have any demo repos of this on gh


Very interested in this too.. not clear at what point something like Postgres will become necessary.

Just writing a simple hello world in node/express downloads a gazillion dependencies and code that'll all be points of failure or mystery for lack of understanding. To understand them all to be able to write non trivial stuff is likely no different from doing httpd in c on Linux.

I've done stuff in go and it makes it lot easier to code.


Take ruby vs go. Ruby can write expressions that resemble a sentence so you can skip commenting and it looks nice. Go is verbose and less English like. Which is easier to understand what the code actually does? Go may take a little extra time to digest but it is infinitely more clear what is going on. If you don’t need to debug the ruby code or optimize it ruby wins because it is easier to digest and reads like English. It is a trade off like many things in software.


This is actually one of the reasons I dislike cucumber so much. They try so hard to make it read like English for non-tech people but to really understand it you need to understand the underlying regex being used.


Hello! I'm Elise, the author of the post. Figured I'd jump in and say that in addition to Ruby, I really like Cucumber. But I think the non-technical people focus really cuts a lot out. For me, Cucumber gives the whole team(tech and non-tech) a shared language for what the software does.

I don't use it much anymore, because most companies seem to have moved away from it. But, I do miss it and wish we had something similar that could encouraged that level of shared understanding.


I spent several years working on production Go code and I never found it particularly clear or intuitive to read or debug compared to Ruby. This seems like a "your mileage may vary" situation.


CenturyLink advertises that they don’t raise the rate and they STILL do it. Mine went from $50->$60 with no speed change. I don’t have the energy anymore to fight it. My alternative is xfinity which is far worse.


In typed languages like go I don’t need to look at module imports linked to files for navigating. With the vim shortcut ‘gd’ I can immediately jump to the definition making this entire discussion about where to put things almost irrelevant. I would still try to organize for reading.


Have they fixed the macOS time drift issue yet?


No, they also haven't fixed the machine "failing to start" (but actually starting anyway) and not mounting volumes. I'm switching back to Docker.


Do you have an issue no.?



I will forward these to the team.

Sorry for the late reply.


You mean the time skew when you have your machine sleep (close the lid), right? We have fixed this a while back; this was part of CRC almost 2 years back and we backported this to Podman Machine. If you still see this, let me know and I'll have someone look at this again.


I've seen this as recently as a few weeks ago on Podman 4.7.1 on a 2019 16" Intel MacBook Pro running Monterey 12.5.1, although I hadn't recreated my Podman machine VM in a few months so it may have been running an outdated image; certainly not a 2 year old one though. In my case it manifested as TLS failures due to an "expired" certificate.


Yeah it is still happening on fresh podman install as of a month or so. So many of my coworkers ran into the issue there is an internal wiki about it. The hwutil sync command thing works but it’s annoying.


Why are we lumping a Linux kernel feature (cgroups) on top of macOS?


It is Go not golang. Golang is used for search it is not the name of the language.


How can you use Golang for search if no one is using it in the corpus?


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: