Fork me on GitHub
#off-topic
<
2020-02-22
>
seancorfield00:02:36

@andy.fingerhut Last year a lot of people were receiving that as a Facebook Messenger post (and the non-tech folks were suitably freaking out).

seancorfield00:02:10

It always shocks me how easily people fall for any of these scams...

andy.fingerhut00:02:36

Her email mentioned a password they claimed to have gotten, which she wasn't sure whether she had ever used before, but sounded like one she might have used on some web site 15 years ago (from her memory, no written record to check). If it is, I am guessing they might have gleaned that detail from a data breach at that company, and certainly raises the level of concern when reading it. Implying they know more than that is wrong is easily checked by her in this particular case, but if you guess that some non-0 fraction of people might have done embarrassing things in front of their laptop, you will be right for some of them 🙂 Email is easier to reflect on such things with time, so they included a 24-hour deadline. Reminds me of phone calls where they claim to be a creditor/bank/credit-card-company/whatever, and it is so easy in the moment to think "oh, I'm so glad that a person who can handle the problem is already right here on the phone with me who can fix it".

manutter5100:02:13

That is definitely a scam as far as the blackmail goes. You can look up your email address/user ID on https://haveibeenpwned.com/ and see if your login credentials appear on any published list of leaked credentials (which is where the blackmailer got it). I've gotten several of these scam emails over the past couple years and they've all used passwords that I changed long ago because I knew about the databreaches.

👍 4
seancorfield00:02:23

I've had both the email and the messenger post with different supposed passwords in them. Sometimes it is a password I used on one of the sites that had breaches, sometimes it is a password I have never used anywhere... which is kinda weird.

seancorfield00:02:20

And then of course there are all the people who get these scams, who have never had a camera on any computer, and they still believe the scam and freak out...

andy.fingerhut01:02:15

Huh. It's been so many years since I bought a computer without a built-in camera...

andy.fingerhut01:02:08

I actually bought one of these for a Mac PowerBook circa 2004-5: https://en.wikipedia.org/wiki/ISight#/media/File:ISights.jpg

seancorfield01:02:40

Ah, memories! I had an isight too, back in the day!

hindol02:02:31

FWIW, I use a password vault to store most (but not banking) passwords and all my passwords are from the vault's random secure password generator. No two passwords are same.

hindol02:02:01

This way, the password from one breach cannot be applied to another site.

andy.fingerhut03:02:36

Yes, very useful technique, agreed, and I use it, too. I am not so sure I would trust that entire collection to a personal computer with a rootkit installed on it, that was quiet about its presence for long enough for me to start that program and enter my master password on it once.

andy.fingerhut03:02:04

Or stated more accurately, I would definitely not trust the security of all of those passwords if such an event occurred.

hindol10:02:50

Yes, agree. Security is hard. Between many imperfect approaches, I like this one better.

👍 4
sogaiu23:02:39

i try not to store the most important stuff digitally

kulminaator11:02:07

has anyone done something slightly bigger than a demo with cljs + threejs ?

Saikyun09:02:40

sure, here's a wip browser release of a game I'm building with expo: http://animal-capital.surge.sh/

Saikyun09:02:48

initial load time is pretty harsh 🙂

Saikyun09:02:34

it's a few iterations old, so don't really need feedback on it.

kulminaator07:02:16

just show a loader or some funny cheap animation while the loader runs on the background 🙂

kulminaator07:02:01

also these days i suggest you grab a free https certificate from lets encrypt , so your page looks more legit in browser of 2020 🙂

kulminaator07:02:20

but cool idea 🙂

kulminaator07:02:47

i was a bit puzzled on how the programming model of three and immutability minded clojure actually match up

Saikyun09:02:06

haha yeah, need to fix loading screen 🙂 web isn't the primary target though (will mostly run on app or electron), so load times aren't normally as dramatic

Saikyun09:02:39

thanks for the tip about https certificate.

Saikyun09:02:31

wrt three / immutability -- have you looked at arcadia? clojure on unity. I basically copy their approach: use clojure for set up and data-driveness, use mutable built-in tools for "frame-to-frame" business

Saikyun09:02:13

e.g. I just create normal three.js objects, but I have an immutable data structure representing the game, then I use add-watch to add/remove "game objects", add animations etc 🙂

Saikyun09:02:15

if one really digs into it though, I think there is an interesting beast lurking in the combination of immutability + gpu programming

Saikyun09:02:35

currently the model is something akin to cljs immutable -> three.js mutable -> webgl immutable

Saikyun09:02:13

if one could afford to ditch three.js (I currently can't since it'd take too much time), you could probably do some very interesting things from cljs -> webgl

Saikyun09:02:22

I've played around a bit with three.bas, which does the animations on the gpu: https://github.com/zadvorsky/three.bas if you build your game around that, you could most likely get surprisingly good performance from cljs + webgl alone

Saikyun09:02:59

I don't know what restrictions apply though. often you get the problem of "positions of objects on the cpu", i.e. the data is stuck on the gpu. however with raycasting I think you can solve some of those issues 🙂

Saikyun09:02:17

either way, this is not something I know much about, but more have a feeling that there are interesting parts to explore here

kulminaator09:02:22

compared to webgl - threejs definitely has the ease of use in it's pockets.

Saikyun09:02:49

exactly! and in my case, I'm taking on a lot of unexplored areas by trying out cljs + three.js + expo

kulminaator09:02:53

if you have ran into the performance issues, have they been on javascript level ?

kulminaator09:02:28

i imagine webgl itself, as it's pretty close to opengl layer, should perform unless the graphics card is absent on the host

Saikyun09:02:10

hm, so far the most sensitive part has been the drag'n'drop on mobile. in that case I solved it by using cljs-functions mutating a THREE/Vector2 🙂

Saikyun09:02:27

yeah you're absolutely right, webgl works great

Saikyun09:02:57

the performance overall is good imo. but I've erred on the side of cautions a lot of times, opting for js-objects when I'm unsure

Saikyun09:02:56

I think one could get good performance by doing everything in cljs, then using defrecord as needed. in my (limited) testing, a defrecorded cljs-object performs as well as a js-object when being read

Saikyun09:02:03

not too sure about write operations

Saikyun09:02:42

also worth noting my game isn't very sensitive overall. if you make a real time game you'd probably stumble into problems I haven't ran into

Saikyun09:02:48

also worth noting I haven't managed to compile three.js with :advanced optimizations on. there's another rabbit hole one has to explore 🙂

Saikyun09:02:27

when I'm more done with my game, I plan on releasing the engine part as open source (for expo + cljs + three.js development). but currently the code is a real mess, and there are some issues to iron out first

dominicm19:02:20

Don't suppose anyone here knows USB well enough to tell me why I can't find a usb OTG C male to A male, only to female?

p-himik19:02:16

I have such a cable, it went with my phone.

p-himik20:02:15

Wait, I just now realized what OTG is. If you don't mind me asking, what's the device that has a female USB type A?

kulminaator20:02:37

i dont think there's a technical reason, just nobody needs otg-c to a male . i have just resoldered usb cables as i have needed to modify them

p-himik20:02:40

Well this one claims to be OTG, but I wouldn't trust it: https://www.aliexpress.com/item/33040564573.html

kulminaator20:02:51

until usb-c spec usb cables were super simple

dominicm21:02:08

I have bought a usb kvm switch, for moving between desktop and laptop easily for my keyboard, mouse, webcam. The switch requires OTG though, and I'm struggling to find something really short (like 10-15cm). My current plan is to buy a male-to-male usb adapter (not cable, which makes it tiny). But I haven't figured out quite how to make sure it's OTG compatible from the pictures/description yet.

dominicm21:02:26

I found a guide on soldering my own, but it obviously didn't cover USBC.

Andy Wootton11:02:32

OTG has 5V output, so it sounds like a feature of the device rather than the cable, as long as that doesn't burn out the cable.

kulminaator12:02:16

yeah you can't get any more than 5V out of a smartphone 🙂 , even that 5V is not guaranteed to have many milliamps behind it (despite what usb spec recommends or requires)

kulminaator12:02:38

regarding usb-c data connectors - you can only connect the cables that usb-a port has anyway ...

dominicm14:02:29

Wikipedia suggested (and looking at my usbc->female A otg connector) that otg has a fifth pin.

Andy Wootton15:02:19

This guy lights up a LED with 2 (of 4) wires https://www.youtube.com/watch?v=YxFemTqSA4I

dominicm15:02:21

I think the fifth pin is connected to ground (fourth wire)