Hacker News Reader

Comment by keithpeter | Hacker News Reader

The Rob Pike approach https://usesthis.com/interviews/rob.pike/

I'm coming round to it (at a considerably less exalted level).

>> Twenty years ago, you expected a phone to be provided everywhere you went, and that phone worked the same everywhere. At a friend's house, or a restaurant, or a hotel, or a pay phone, you could pick up the receiver and make a call. You didn't carry a phone around with you; phones were part of the infrastructure. Computers, well, that was a different story. As laptops came in, people started carrying computers around with them everywhere. The reason was to have the state stored on the computer, not the computer itself. You carry around a computer so you can access its disk.

In summary, it used to be that phones worked without you having to carry them around, but computers only worked if you did carry one around with you. The solution to this inconsistency was to break the way phones worked rather than fix the way computers work. <<

What I do is the VDI approach. I have one AWS Windows instance with and I RDP to it from any of my computers at home, the office or any of my second homes, which happen to be iMacs but could be anything.

I'd love to do something like this but I'm 150ms away from cheap VPSes..

Do you even watch Youtube (say) inside the instance??

If I do not have fiber or my ISP is having a bad day, using my phone G4 as a personal Hotspot I get some 80 ms. I use USB instead of wifi to save 5ms. To save some money, I shutdown the AWS instance when I do not need it because it starts up in seconds.

For web browsing I often use the local browser because with Macs swiching the current desktop is just a key stroke away (which is not the case for W10 if one of those desktops is the RDP client.

How's the latency for text editing? I tried doing something like this with *nix tools but found it unbearable even when connecting to localhost.

In my case it is indistinguisable from local text editing. I remember however that I chose to turn off SublimeText scrolling animations because it was a way worse than locally. My fiber latency to my AWS instance is about 60ms and with phone 4G about 80ms.

What does that run you per month? What are you doing on it compute wise?

I do not have exact costs at hand because it is just one of several AWS instances we have, but probably my fully loaded costs on a 160 hour month for this instance would be something like:

- 150GB SSD: $15 - 150GB 30 last days of daily snapshshots as a backup: $3.45 - 160 hours of t2.medium (4GB) Windows: $8.64

The typical applications I use are Visual Studio, Delphi, Office, Chrome, and some domain specific apps.

Admittedly, you can reduce further the SSD cost of $15 to 160/744*$15=$3.22, by snapshotting and deleting the SSD volume each time you shut down but I never did that optimization because startup time would then not be seconds but something probably in the range of 5 minutes, as you need to have some lambda funciton create a new instance, create a volume from the last snapshot, shutdown the newly created instance, replace the boot volume and finally start the instance with the right boot volume.

What I also like of this approach is that I do not have to overprovision disk or instances, if I ever need a larger drive, I just modify the volume, if I need a bigger instance, I just shutdown and start with a bigger instance size.

It's not really true, since he wrote his OS, text editor, programming language. Now he just installs the language and the editor/IDE wherever he goes and uses that.

Reply on news.ycombinator.com