Live data from Hacker News

This Blog is now hosted on a GPS/LTE modem

nns.ee

171–180 of 212 posts

Re: This Blog is now hosted on a GPS/LTE modem

#171
post #80

So a modem on a pine phone can survive the HN hug of death (and the number one post no less)? Am I missing something here? Because there's no way I believe that!

Yes it can survive the HN hug of death.

I speak as a software engineer whose blog has hit the HN frontpage multiple times, and who has optimized my blog to be static & lightweight, just like the author. In my experience, a front-page HN post (in the top 3) leads to 30-40 page hits per second. The page at https://nns.ee/blog/2021/04/01/modem-blog.html weights 44.4 kB transferred (HTML, CSS, images, fonts, etc, all compressed over HTTPS). So this is 11-15 Mbit/s of peak bandwidth, below the maximum throughput the author measured on the Pine phone modem (20.7 Mbit/s).

And the bottleneck is going to be purely network. Not CPU. Not disk. Just serving a small set of files cached in RAM by the pagecache.

Re: This Blog is now hosted on a GPS/LTE modem

#172
post #18

According to this[0] which is a neat read anyway, it "has 256 MiB DRAM, 256 MiB NAND, and a single Cortex-A7 CPU clocked up to 1.3GHz." You can run DNS, a VPN, or a small web server even on a VPS with 32MB of RAM so those are pretty cool specs. [0] https://xnux.eu/devices/feature/modem-pp.html

Wow, that's probably about the same ram & cpu as my desktop 16 years ago.

I bought my first PC in the late 80s. It was an 8086 by Amstrad, with a 4Mhz processor, 512kb of RAM and no HDD. It was running MS DOS 3.30 Plus and I made my first programs in GW-Basic.

Re: This Blog is now hosted on a GPS/LTE modem

#173

Earlier quoted context omitted.

Could be. The statuspage from the device shows darkhttpd as the server, not nginx: https://nns.ee/blog/status.html

Looks like darkhttpd's website: https://unix4lyfe.org/darkhttpd/ is NOT hosted on a GPS/LTE modem and has been hugged to death.

I think I saw it mentioned (possibly also linked) in a comment on another article here today. So it's not just us, it's also us!

Re: This Blog is now hosted on a GPS/LTE modem

#174
post #35

This is realy cool! I wonder if Justine's Redbean[0] webserver, which got recently posted on HN[1], would support ARM in such use case. Its executable size is 460KB, it even allows Lua scripting, and benchmarks 1 million pages per second on desktop PC. [0] https://justine.lol/redbean/index.html [1] https://news.ycombinator.com/item?id=26271117

Possibly not. I believe that runs natively on (any!) x86 system, but it uses QEMU to run on ARM, which might make things a big too big here.

You are right. Indeed it falls back to QEMU on non-x86_64 https://github.com/jart/cosmopolitan/issues/73#issuecomment-...

Running QEMU on a modem would be another hack for a day :)

Re: This Blog is now hosted on a GPS/LTE modem

#175

Earlier quoted context omitted.

Possibly not. I believe that runs natively on (any!) x86 system, but it uses QEMU to run on ARM, which might make things a big too big here.

You are right. Indeed it falls back to QEMU on non-x86_64 https://github.com/jart/cosmopolitan/issues/73#issuecomment-... Running QEMU on a modem would be another hack for a day :)

Don't give me any ideas...

Re: This Blog is now hosted on a GPS/LTE modem

#176
post #38
post #24

Earlier quoted context omitted.

Do those regulations explicitly say that the RF device is not allowed to be programmable by the user? And isn't the user legally forbidden from substantially modifying the RF behaviour in any case?

I can't find it now but there was an article posted a while ago that explained the existence of just such rules back when the very first mobile phones used analogue transmission, which of course was easy to intercept. So the FCC forced radio manufacturers to prevent users being able to tune to those frequencies and additionally specified that manufacturers be responsible for making this limitation very hard to defeat…

https://news.ycombinator.com/item?id=25403163 "Why radio receivers won’t tune 800-900 MHz"

Re: This Blog is now hosted on a GPS/LTE modem

#177

Is there something like the PinePhone that is manufactured in a place with substantial political freedom (however you wish to define that)?

Depending on your definition of political freedom, USA might fit: https://puri.sm/products/librem-5-usa . It's expensive, since practically no phones are produced in USA. Alternatively, https://puri.sm/products/librem-5 has its components produced in China and assembled in USA.

Holy crap, $2000! Literally 10x the price of PinePhone. Still, that’s exactly what I was looking for; thank you!

Re: This Blog is now hosted on a GPS/LTE modem

#178
post #43
post #24

Earlier quoted context omitted.

Do those regulations explicitly say that the RF device is not allowed to be programmable by the user? And isn't the user legally forbidden from substantially modifying the RF behaviour in any case?

Requiring a device to not allow illegal RF behavior is significantly easier to require and audit than policing the actions of millions of consumers.

Off-topic, abstracting laws away from the harm they aim to prevent has always bothered me a bit. By punishing behaviors (e.g., making a device with a certain set of capabilities) other than the ones people care about (e.g., careless or malicious RF activity) you're nearly guaranteed to introduce unwanted loopholes and second-order effects.

E.g., CA requires tags for your vehicles (fair enough -- gotta charge people for the privilege of driving at all, not just incrementally through gas taxes and whatnot /s), but the law isn't that your vehicle be registered and have tags purchased; it's that valid tags must be displayed on the vehicle.

That makes enforcement super easy because an officer can just walk up and down a line of cars handing out tickets, but it also means that people can break the law for no fault of their own even after having taken the action the law was intending to promote (paying CA in this case). If you tried to renew your tags at the right time in 2020, even months before they were set to expire, you couldn't pick them up in person (covid) and wouldn't receive them in the mail till after the governor's order temporarily allowing you to drive with expired tags had expired. Moreover, at least one police department was excited to cash in on that discrepancy :)

The problem in that example is that the behavior the law is supposed to encourage (paying CA if you own a car) only partially aligns with what the law punishes (not displaying a valid tag on your car). That lack of alignment leaves room for exploitation.

That's just a small example (intentionally so -- hopefully it's simple enough so as to not be controversial), but it's a constant pattern in our legal system:

- Taser-immune meth-heads robbing people at knife point is bad, so let's make a hint of a whiff of any kind of drug anywhere near an individual for any reason a felony.

- Child pornography is bad, so let's (try to) make encryption illegal.

- Pump-and-dump scams are bad, so let's make [thousands of pages of financial legalese] illegal.

- (intentional hyperbole on this point) Low quality medical devices hurting people is bad, so let's give a private corporation the privilege to write up a set of standards, never change them no matter how far the world progresses, and charge people for the privilege of reading them to know what's legal or not.

Re: This Blog is now hosted on a GPS/LTE modem

#179

I have been using these modems at work, specifically the EC25/EC20. They come with linux with some limitations. These modems are typically meant for IOT devices and you can get them for ~ 16/18 USD (or even lower). 4G/3G/2G, with around 128/256M flash and 128M RAM and 1.3GHz Arm Cortex A7. They are pretty powerful for their size, cost, current consumption and usability. Almost all such parts can be hacked across manu…

I'm a bit confused: does the manufacturer provide a toolchain to compile and run binary code on something that is not advertised as a SOM/SOC, but only as an lte modem?

Yes that is true in fact even the most basic modems in their line up for 2G can be used as an Application processor rather than as a plain modem.

In this case however it is more like a complete build environment that includes the Linux distribution, GCC and other relevant Linux libraries etc for the distribution. The real FW for cellular modem is likely never exposed since that may run on a DSP core or an ASIC sub system

Re: This Blog is now hosted on a GPS/LTE modem

#180
post #171
post #80

So a modem on a pine phone can survive the HN hug of death (and the number one post no less)? Am I missing something here? Because there's no way I believe that!

Yes it can survive the HN hug of death. I speak as a software engineer whose blog has hit the HN frontpage multiple times, and who has optimized my blog to be static & lightweight, just like the author. In my experience, a front-page HN post (in the top 3) leads to 30-40 page hits per second. The page at https://nns.ee/blog/2021/04/01/modem-blog.html weights 44.4 kB transferred (HTML, CSS, images, fonts, etc, all com…

It's above the max, no? The author measured 20Mbps between the phones main OS and the rest of the world, but 10Mbps between the phone and the modem (over the adb bridge).
Post reply on HN