[gregmac]

I just switched ISPs, and the new one has one of the most obnoxious phone processes I've ever interacted with.

I go through the usual hoops: press 1 for English, "we detected an account linked to the number you're calling from, is that that you're calling about?" ... Press 1 for support, press 1 for Internet, "no outages detected in your area. Most problems can be solved by rebooting your modem. Press 1 if you want to try rebooting." (Pause)... "thank you for your call click"

First off, rebooting doesn't solve my problem. But I guess I have to try anyway?

So I call back, this time I do pick to reboot, and get "your modem will reboot in the next few minutes, and could take up to 10 minutes to come online. If things still aren't working, try our online support chat"

So, basically there doesn't seem to be any phone technical support (with a human), at all.

Also, rebooting is offensive to me as a programmer. Kernel updates and memory leaks are the only reason you need to reboot. How absolutely shitty is modem firmware that the ISP actually spent the time to build this reboot system out??(Never mind that I personally don't feel like I've ever had a modem/isp actually problem solved by rebooting)

Made me wonder if I should have switched.

[yetihehe]

> (Never mind that I personally don't feel like I've ever had a modem/isp actually problem solved by rebooting)

I had problems solved several times by rebooting modem. One time it was "reboot modem and access point in proper order", me naively rebooting them both at the same time didn't help, only phone support solved this problem.

> Also, rebooting is offensive to me as a programmer.

Hmm, I might be desensitivised from too much programming in erlang. It's implied that your program will encounter bugs or strange data and parts WILL be restarted, better account for that and plan on what to do on restart of each small part at the start of writing your program.

> So, basically there doesn't seem to be any phone technical support (with a human), at all.

Because it's cheaper. Those who don't have support can offer lower prices. When people search for trinkets, they only have information about what is supported, there is no good information about quality of device and support, high price also not always means better support. SO they just go for lower price and hope not to suffer too much.

[gregmac]

> Hmm, I might be desensitivised from too much programming in erlang. It's implied that your program will encounter bugs or strange data and parts WILL be restarted, better account for that and plan on what to do on restart of each small part at the start of writing your program.

That's different from having software that can get into an unknown broken state, where physically power cycling it is the only fix.

It's better to crash and ensure there's a process that automatically restarts. Or use a watchdog process. Just breaking, to the point companies build external systems to reboot them, is just bad.

[eru]

Yes, see also https://news.ycombinator.com/item?id=40212967

[ginko]

>If things still aren't working, try our online support chat

>So, basically there doesn't seem to be any phone technical support (with a human), at all.

I wish everything had support chat. IMO it's much less hassle than having to call. It's usually trivial to get through the first layer of automated support and get a human on the line.

[maccard]

I agree, but;

Support chat is universally shitty. My mobile provider's website only works if you keep the browser window open, and times out if you go away for 2 minutes. The replies often take more than 2 minutes. I can only access "certain" information about my account if I'm on mobile data, except my carrier's website doesn't work if I am out of data. (granted, I am on a super budget mobile network, but still). In the last 18 months, the chat experience has been taken over by LLM's which are just acting as full text search for the doc pages that don't solve my problem.

I still choose web chat over any other method of interaction though.

[auggierose]

Why did you switch? Can't have been their reputation for support.

[eru]

> Also, rebooting is offensive to me as a programmer. Kernel updates and memory leaks are the only reason you need to reboot. How absolutely shitty is modem firmware that the ISP actually spent the time to build this reboot system out??(Never mind that I personally don't feel like I've ever had a modem/isp actually problem solved by rebooting)

Why is rebooting offensive to you? State is hard; resetting your system to know state can fix many issues.

See also https://news.ycombinator.com/item?id=40212967

[sethammons]

If your microwave had an error, you would be put off if you had to power cycle the whole house. Installed a new receptacle, sure, but operating an appliance? No way. Now you would have to reset your clocks everywhere at a minimum.

I have a linux computer running a public server that has not be restarted in three and a half years. This is what I expect.

Every time I have to reboot my work laptop due to work pushing some updates or that I have to reboot my windows machine because it is running unreasonably slow, I am reminded that inconsiderate assholes have become more lazy and are ok with polluting the whole system, mismanaging state and resiliency, and when the equivalent of the microwave has an error, the only solution is rebooting my house. We can do better.

[eru]

> I have a linux computer running a public server that has not be restarted in three and a half years. This is what I expect.

I restart my Linux desktop every few weeks, when the kernel updates.

For a reliable server, you want to exercise the restart ritual somewhat regularly, because when anything goes wrong (eg with the hardware), you might have to restart anyway, so you want to be sure that this works.

[bornfreddy]

Windows is awful at this. Completely weird problems with many apps (especially VPN) which get resolved with a reboot. Seriously? It is the whole culture around this OS which finds this acceptable.

[chgs]

Often this will be resolved by the network being quiet long enough for a reboot - with connection tables in intermediate firewalls timing out etc - rather than the actual reboot.

[gregmac]

It's a sign of bad code.

Yeah state is hard, but "else" statements are a thing. Even if a state is not "expected" if the system can represent it, it should be treated as possible.

Crash if you have to, but don't do unexpected things or just stop executing.

[maccard]

> Also, rebooting is offensive to me as a programmer. Kernel updates and memory leaks are the only reason you need to reboot.

This surprises me - as a programmer you should realise that reboots can often help. Cache invalidation is one of the notoriously hard CS problems and an awful lot of systems will start fresh on reboot.

> (Never mind that I personally don't feel like I've ever had a modem/isp actually problem solved by rebooting)

My current ISP is better, but my previous ISP cycled IP addresses at 2am (and lost connectivity for about 30 seconds at the same time) on a Friday night. I would semi-frequently be up playing games at that hour, and it was about 50/50 as to whether devices on my network would survive the blip. Rebooting the router had a 100% success rate.

I currently (unfortunately) have a google wifi mesh system. It works great, except about once a month it reports that absolutely everything is fine, all tests pass from my mobile device, but my laptop has no internet connectivity. Rebooting fixes it just fine.

> How absolutely shitty is modem firmware that the ISP actually spent the time to build this reboot system out?

Firmware is still software, like it or lump it. Modem firmware has been shitty for a long time. A major ISP [0] in the UK had an issue with their firmware that caused massive latency spikes under load. Alsom Power loss happens sometimes. The modem/router has to be able to turn on in the first place, so a "reboot" is just going through that process again. It's attempting to return to a "last known good".

[0] https://community.virginmedia.com/t5/Forum-Archive/Hub-3-Com...