ChipTuner

https://www.vaughnnugent.com

Building software they don't like.
Low-level and server engineer: libnoscrypt, NVault, vnlib.

Public Key

npub1qdjn8j4gwgmkj3k5un775nq6q3q7mguv5tvajstmkdsqdja2havq03fqm7

NIP-05 Address

ChipTuner@vaughnnugent.com

Profile Code

nprofile1qqsqxefne258ydmfgm2wfl02fsdqgs0d5wx29kweg9amxcqxew4t7kqpr4mhxue69uhkummnw3ezucnfw33k76twv4ezuum0vd5kzmp0qythwumn8ghj7un9d3shjtnwdaehgu3wvfskuep05lmmn4

Publishing to

nostr.bitcoiner.social

relay.nostr.band

Author Public Key

npub1qdjn8j4gwgmkj3k5un775nq6q3q7mguv5tvajstmkdsqdja2havq03fqm7

Show more details

Last Notes

2024-11-05 14:32:16

- reply

Please do! I want to have support for many languages in the future
The issues is that on my most *nix platforms the `memset(ptr, size);` call gets optimized away because there is no read after write. So it's considered insecure. Along with that both the cl.exe and cc compilers treat volatile argument pointers differently and they even disagree heavily about when to use volatile pointers either as arguments or even in the procedure body. Really the only right way is to look at the assembly after you compile your project to see if there are instructions to wipe the buffer correctly.

2024-11-04 23:02:49

- reply

Thank's for the info I'm going to consider it. I don't have any strain now but I'm interested in optimisation lol

2024-11-04 18:25:32

- reply

Yes I remember your keyboard posts, I haven't actually tried using anything other than a "qwerty" setup. How long did it take you to switch? I'm currently on brown keys, i used to be on red which is very light. This dell has very, very heavy keys. I do really like the reds but most of the leds are burn out and it bothers me so I use a cheaper mech with the brows now.

2024-11-04 18:10:08

- reply

Wow, yeah I don't think i've really ever had a keyboard failure, maybe a junky membrane I got in parts boxes

2024-11-04 15:14:18

- reply

How often do people usually purchase keyboards??? I didn't know this was a thing.
I usually have more issues with USB2 devices in USB3 ports than anything. I have to purchase motherboards with at least two USB2 ports.

2024-11-01 15:48:00

- reply

What he said.

2024-10-29 16:11:50

- reply

Then it will just be California where you aren't allowed to register old vehicles anymore.

2024-10-27 00:31:53

- reply

Thank you, but node is up, channels are a pain, and lightning is kind of silly imo so exploring other options too :/ I have an on-chain donation address for those that are highly interested :)

2024-10-26 22:17:30

- reply

Don't worry there are still lots of us out there!
https://www.vaughnnugent.com/resources/software/modules/noscrypt

2024-10-26 20:03:37

- reply

2024-10-25 18:44:21

- reply

*Half duplex mode for a given duration or signal. Essentially master/slave

2024-10-24 17:52:43

- reply

I appreciate it! I had an offer a couple days ago as well for smaller channels so I'm taking offers for sure!

2024-10-23 17:16:20

- reply

Yeah I'd like to get there too! Also I'm glad to be one of your 108 follows :)
I should probably go for some pruning, I know I'm missing a lot of dev discussion I should be part of.

2024-10-23 17:05:38

- reply

Completely understand, good luck and I'm sure you'll be fine for now! It's a shame these things have to be such a trap for newbies. Just reach out if you get stuck with anything!

2024-10-22 15:32:36

- reply

Oh yeah baby is that a 710 at the bottom? I still have one of those collecting dust as a cold fallback. I just got rid of all my r910 space heaters
Also just setup nodes myself if your in need of some assitance.
I chose the Bitcoin Core, CLN, RTL stack. I went with a container setup I can share my compose file if you are interested!

2024-10-21 23:25:56

- reply

So nothings changed really XD

2024-10-19 17:11:00

- reply

If that's the case then sure, maybe make it clear the size of your company if it's that tiny.
When we were that small (just 1-2 people) of course you have to handle it individually, but you still attempt a support funnel so you can scale the support side of the business. Even when we were over 10 staff, and handled support ourselves we did so as the company and reached out as the company. Unified support image. You don't want to drive support issues to your handle. Because you're the same guy, if you're lazy on the offical support channel, people will funnel their requests to your personal channel. You have the incentive to funnel people away from your personal channel.

2024-10-19 00:38:57

- reply

Well I've never conferenced so I see!

2024-10-16 22:13:31

- reply

So how do?

2024-10-16 21:44:23

- reply

Right, but I need to have the UTXOs available for the liquidity right

2024-10-16 21:29:49

How to lightning for poor people? It looks like most popular peers are requiring anywhere from 1Ms to 5Ms or more to open a channel. I don't have that laying around. :/
Maybe I'm misunderstanding? Still learning about channels.
#lightning

2024-10-15 20:28:54

- reply

I'm doing it for repeatability and reduced dependence on systems like Jenkins, Github, etc. I think it scales very well. The point was to allow other developers to have access to the same pipeline I do without all of the GB of server software and required knowledge to manage it.

2024-10-14 21:25:39

- reply

On second thought I probably already have it. Silly me

2024-10-09 04:13:29

- reply

I owe you a response! I don't know really anything about Toyota's from that generation, especially Land Cruisers. That seems like it should be old enough to avoid the phone-home tracking systems. Likely new enough to still get parts for if needed. Take a look around at your favorite parts catalog website (Rock Auto here in states is good) and see if they have a good number of parts still available, ideally OEM from the manufacturer, aka "genuine parts"
Also you may be able to use https://vehicleprivacyreport.com/ to see what it says. It might not have much from that era

2024-10-08 21:11:17

- reply

@npub1q6y…t3sh Was careful to remind me of that earlier XD

2024-10-08 18:22:51

- reply

I have.
LN is a dumpster fire of complexity and overhead. As you said, well it's a Layer 2 system so that's how it goes. I just mean to say it's going to take far more work than nostr to reach maturity. Why does it require curl to interact with my Bitcoin Core node and Core Lighting to access all features? In some cases it feels like I would be better off and safer doing it myself.
But seriously, there are like 4 developers working on anything at the core. @npub179e…lz4s being of the few.
When I was invited to look at Umbrel, I realized they built their own, and with no fault of their own, it's a fairly complicated mess that the average self-hosted is unprepared to manage correctly.

2024-10-04 06:37:09

- reply

Yeah. When I have time to get around to blossom, I'm hoping we can make a difference there :/
It's obvious big data is a difficult problem to work with

2024-10-03 18:00:26

- reply

Big fan of incandescent lighting. I still have a bunch of bulbs stashed (most are outlawed now here in the US). It's just about the feel and comfort. Consumer LED and florescent bulbs give me headaches and makes me dizzy, for a lack of a better word. It's that I can see the 30/60 cycle flicker. Especially when things are moving in a room with flickering lights.
I use 12v DC lamps and led strips in my office and bedroom so I don't see the flicker, I power off an ATX psu lol. My ceiling lamp has "specialty" halogen floods in it (probably can't get them anymore) and my reading lamp is incandescent.

2024-10-03 17:04:23

- reply

I would fucking pillage android if I had the motivation. Smart phones do way too much these days

2024-10-03 05:27:31

- reply

I'm building a "dedicated" VM for GC stuff, so I haven't got status page back up yet :)

2024-10-02 20:19:35

- reply

Wouldn't know, hanging out here on TheForest ;)

2024-10-01 21:25:29

- reply

But yeah now I see where the newsletter thing would come in.
It's probably better the other way around, or maybe require both, if either is missing it's assumed the user should not be tagged.

2024-10-01 21:24:21

- reply

I'm thinking the npub should exist and be authoritative, otherwise alias could be created everywhere.
I suppose you could announce: Hey I'm part of @gc-dev-team but I think it's probably best to have the npub actually exist, and sign a list of npubs that are aliased.

2024-10-01 21:06:20

- reply

The beard comes with the GNU standard libraries

2024-10-01 18:35:54

- reply

I think it can be done really easily with maybe some metadata and client resolution

2024-10-01 18:34:01

#nostr #dev question
Something I've wanted but haven't figured out how to formulate.
Title: alias tagging?
Description:
A way for users to tag a group of users (presumably kind 01 to start?) with a single @ in their client. When a user @npub1s3h…75wz for instance, we could configure a way for all "members" to be tagged instead.
Use case:
Teams to be tagged on a note via a single reference by the user, but are tagged individually so they can all get notifications.

2024-10-01 18:14:23

- reply

Yeah, I simply avoid them. I used macros and tiny structs to track length. In noscrypt I replicated C# spans with inline functions and small structs, the same way literally every other language does because most of them are written in C anyway lol.
To be fair though, if people didn't keep ignoring the language requirements, ub, and ignoring compiler manuals, storing the termination with 0 at the end of a buffer is arguably the most efficient way to handle that. Strings are just a sequence of memory, and literals are stored in data segments. Most efficient way to handle that imo.
https://git.vaughnnugent.com/cgit/vnuge/noscrypt/.git/tree/src/nc-util.h#n96

2024-10-01 18:08:08

- reply

But why?
It works and I have control over my CPU instructions if I want it :), because it's worked for 40+ years in the same way over and over again, and is available on every machine and developer and ABI layers, the same way it's always done. You can save your breath on zig I'm very aware of it's capabilities and I'm not against it, right tool for the right job, but for the reasons stated above, I think C is the largest hammer in the tool belt and should continue to be used when necessary.

2024-10-01 18:02:26

- reply

Okay, I will admit I'm not the most satisfied with libsecp256k1. BUT don't let that reflect on classic GNU software, despite my also not following GNU/FSF development best practices XD

2024-10-01 18:00:01

- reply

Despite my grump attitude and attraction to technology from the 80s and 90s, I'm likely of the youngest demographic on nostr XD

2024-10-01 17:58:00

- reply

2024-10-01 17:56:06

For when yall give up and realize C is the proper language to build in.
https://www.vaughnnugent.com/resources/software/modules/noscrypt
#nevent1q…a7gy

2024-10-01 17:52:49

- reply

Pretty sure were the only ones building nostr tools in battle tested langauges, with official gray beards, and grumpy attitudes. The correct way to build software.

2024-10-01 05:38:19

- reply

For shipping my library to public repos?

2024-09-30 22:17:49

- reply

I've also been thinking of borrowing a friend's welding rig and making some up myself, but steel prices are out of my budget, and good scrap steel is pretty hard to come by these days.

2024-09-30 22:15:43

- reply

You read my mind lol

2024-09-30 21:56:38

- reply

Started collecting Craigslist equipment in my teens, really paid off now. Only thing I'm missing is a squat/lift rack. Maybe a leg press would be nice :/

2024-09-30 17:28:48

It's so fun being able to reference my source code files, in my code documentation, from my self-hosted git browser, on my own domain :)
https://git.vaughnnugent.com/
https://www.vaughnnugent.com/resources/software/articles

2024-09-30 16:44:56

- reply

Kind of off topic. All of my close college friends work for US Military contractors (and all of their family members). The stuff they tell me consistently blows my mind. One works for an engineering contractor that has no specialty, they just bid on whatever the teams say they can do. Recently they were contracted to build and equip off-the-shelf, enterprise server hardware, into portable racks with other custom radio/sat tech that could be fitted onto planes, and some air dropped and survive. Presumably somewhere in the middle east were my other friends are currently deployed.

2024-09-29 21:53:30

- reply

Currently Chameleon on "Firefox". I have not fully vetted the source, just a perusal a year or so ago.
I use it to the fullest extent. Many (Including Brave) believe that fully hiding your fingerprint makes you more visible, which is why Brave discouraged and reset the defaults to be less "private" for the case of less visible.
The important part with Chameleon is that you can create whitelists so you can follow whatever best privacy practices you believe help you.
- User Agent
- Language
- Spoof X-Forwarde-For Ip addresses for misconfigured proxy servers
- Extra feature disable like websockets, CSS, canvas, camera and mic stuff
- Font fingerprinting (better to disable remote fonts anyway using uBlock anyway)
- Extra cookie and fingerprinting features that Brave and Firefox still have but disable by default and I think have now hidden from the UI but can still be enabled in the about:\ pages

2024-09-29 21:45:25

- reply

Yeah and that's a free market, big fan, incentivized to follow the user's wishes when you are beholden to your customer base, not higher powers.

2024-09-29 21:43:52

- reply

And I have the ability/power to do it myself and not at the mercy of a corporation that is beholden to may more powers than I

2024-09-29 21:13:06

- reply

Very true. However I believe the protocol still needs the ability to regularly store targeted npub notes forever imo. What public officials say on their public accounts should be permanent public record. I'm sure many would feel that way even about many public figures.
> is not the same as demanding that relays store every single one of your chat messages forever
yes and all clients and relays always have the ability to say GFY I'll do what I want at any time

2024-09-29 20:36:00

- reply

#6 imo. NVault (optionally) archives everything it signs in your accounts database (regardless of npub but can be filtered)

2024-09-29 06:19:43

- reply

I'd say without the gamification, but that would be false. Quite easy to get attention in the same ways.

2024-09-29 05:03:04

- reply

>The users are comfortable, and even objective improvements introduce an unwelcome disruption and a new learning curve.
Yup! I missed that point, but I'm sure that's one of the largest. My experience is with customers who generally dislike technology and especially change. A difficult demographic to please.

2024-09-29 04:51:24

- reply

Almost every UI I have done, I redo from scratch because it's harder to refactor design, and nearly impossible in production imo. Which is why I think products that have objectively bad UI stick to it (*cough* Google) because it's too expensive to redesign.

2024-09-24 21:04:46

- reply

Tag me any time for stuff like this!
Step #1 care that your car spies on your whole life even when you aren't driving it
Step #2 should use an older vehicle that needs maintenance
Step #3 care that cars have become as disposable as tin cans and parts are hard to purchase on purpose
Step #4 Get past MFGs promoting disposable vehicles
Step #5 understand that incentives only align with moving new vehicles and only ones that the MFG recommends because MFGs control what dealers can purchase by adjusting purchase prices
...
Most people haven't made it past step 1 yet.

2024-09-24 05:48:10

- reply

So since my major project is an HTTP/Web server, the configuration is meant to mirror that of nginx. A variable array of virtual-hosts transports, and so on. A single app (with all plugins loaded) is somewhere between 250-300 variables the user would want access to. Each virtual host is a minimum of 30 variables with only a single transport loaded (with tls)
If the user inputs the wrong value, the app needs to crash and tell them, no fallback, the user needs to trust that when they input a value the application does what they told it to do

2024-09-24 05:32:18

- reply

I can only speak to US vehicles mostly. I know GM started using 3g cellular connections (that no longer work) somewhere around 2008 and later. For repair-ability and parts availability somewhere under 2010 I think is a good spot. You also get some of the newer safety features if that's on your list. But most cars in this era are heavily computerized, except for your more budget model stuff.
Fully computerized engine management (throttle is fully electronic) electronic traction control with differentials and braking systems that are expensive when they break.
90s into 00s is the newest without computers being too intrusive, but more likely to fail due to age, and parts are harder to find, especially computers. I have hoarded at least 1 spare electronic module for my truck for this reason, they are so hard to get PCM's for, and I usually have to program them because the correct ones are too hard to find.

2024-09-23 17:57:58

- reply

> in my 8 years working in this field my biggest take-away, as of recent days, is that you make so much trouble adding features
You're right.
My big thing is unless you require the user to add another system like dotenv it's not simple to store state unless you write a script that exports variables.
My other point, I remembered was, I don't like env variables for system configuration, tuning maybe. I use them like many other libraries do, for adjusting startup library behavior. For instance, enabling debugging, tracing, specifying the heap binary to map into the process, things that are meant to be transparent to the application developer at startup. If the library is used widley, it needs global state, the best way to ensure that variable is set when the process starts is with an env.
Other than that, strict typing and language enforcement that I don't have to write parsing code is good. Errors in configuration data is easier to validate and give clear feedback is also a big deal.
I was going to get into secrets, but its kind of pedantic, I don't like secrets in environment variables.

2024-09-23 15:02:41

- reply

My bad! I swore I saw env in earlier versions. I recently took a liking to yaml over json as config files grow to over 300 lines it gets really messy

2024-09-23 15:01:30

- reply

I can't get behind that personally. I hate that node aps use env variables. I think it's extremely messy, I dislike containers for the same reason. I understand the utility, but for an application (especially a server application) a typed and documented configuration file is ideal. Most of my apps would need 100s of variables and if I'm not mistaken there is an os limit that smaller than that.
I had another good point but I forgot it in the making of this node XD

2024-09-22 19:41:53

- reply

Whoooo! Config files! Please tell me you moved away from ENV variables for application specific config?

2024-09-21 16:24:32

- reply

I'm not well versed on relay coms, but I believe it's per session when the socket is opened. It would get expensive otherwise. However I think a relay can send an auth message whenever it feels like, but it's attached to a session id.
The only code I'm most familiar with is Grain, and he uses per kind limiting iirc. Strfy defers it to the handlers I believe, so it depends on the message kind. I would assume strfy also has per-req limiting as well.

2024-09-21 16:07:12

- reply

Resource exhaustion, so were just asking for a whitelist for sessions I suppose. It's still possible to overwhelm a server by repeatedly requesting auth upgrades too, deplete the RNG pool requesting challenges. So rate limiting still applies, and can mostly solve the exhaustion problem.
I think it's fair to just say whitelist may be useful for reading. I don't think it helps any more than rate limiting for resource exhaustion.

2024-09-21 15:31:55

- reply

I'm starting to think AUTH is utterly necessary for writing. I think Digler is correct in letting users read without proof, it doesn't risk any harm to the network to read. It's just going to be a relay architecture change. That and clients will just need to accept the reality and handle auth better.
At first I was sort of against auth, like who tf are you to tell me "papers please", just be a dumb relay and handle it for me wouldya chap.

2024-09-21 15:13:02

- reply

So basically issuing a noise protocol for message data. I suppose that could be done with proxies as some sort of aggregation. I think we're still fighting the war of my IP may be linked to my npub and were fighting networking protocols here.

2024-09-20 18:36:02

- reply

Dedication. That's pretty cool and I really hope you can continue to make that work for you and your customers! I'm happy I got out of auto for now, but I do miss all of the problems that need to be solved and no one to solve it. Just no really money investment, and it's super dangerous running up against so many law enforcement agencies and regulations.

2024-09-20 17:00:25

- reply

More people need to build HTTP from scratch again.
https://github.com/VnUgE/VNLib.Core/tree/master/lib/Net.Http

2024-09-19 21:14:05

- reply

It's got a lot of attention, it seemed pretty neat when I took a look at it a while go. Syntax is a big deal for me, old dog I guess.

2024-09-19 20:32:57

- reply

Then you got people like me that think C is the way it should be. Specifically C99 imo. That is neat btw, adding it to bookmarks

2024-09-19 16:11:04

- reply

As mentioned downloading APKs are the first-hand way to install apps directly from the author.
Obtainium is a 3rd party app store taht allows you to fetch APKs from their source. That's my go-to for Android now and it's easier to get updates automatically rather than remembering to go to the authors website and download the new update. Obtianium has some quirks but for most FOSS android apps it works great. It relies on GitHub releases, F-Droid and a couple other common places, but it also allows scraping of html websites for APK links from the author.

2024-09-18 21:41:46

I knew LinkedIn was a giant piece of shit, so this just confirmed it. You get locked out if you don't verify identity in a certain period of time. A smart phone and app is required in order to upload verification data (government id and a selfie). You cannot contact support unless to can sign in.
So to recap, I cannot access, edit, or delete my account without sign in, I cannot sign in unless I verify my identity, I cannot verify my identity without a smart phone and their fucking app, I cannot contact support to help me without access.
This is by design. Instagram and Snapchat also require the app to be installed on a smart phone before web access can be temporarily authorized.

2024-09-17 22:41:10

- reply

Yeah that's for mobile unfortunately and I don't use a smartphone. I also connects directly to Youtube servers which I cannot do on my network :(

2024-09-17 21:28:00

It looks like Google (YouTube) has successfully blocked privacy video proxy services Invidious and Piped completely at this time.
Any recommendations for accessing YT without contacting resolving Google domains is welcome.

2024-09-16 17:50:34

- reply

Uh sounds kind of silly, but also maybe consider building a team if possible? Then get crack smoking engineers like myself to do it for you?? Less time implementing and more time designing? That's how @npub1m3x…a5sf tricked us XD

2024-09-15 18:46:59

- reply

So I should keep multiple btc instances running? I assume they could share the blocks/ directory also or will it get all corrupted? Also should I put a load-balancer between the ln node and the btc nodes?

2024-09-15 18:04:53

- reply

For me, i felt like it was pretty dumb to have to dedicate crappier hardware to a node. I have a massive enterprise stack ready to run these things, but the size of the blocks forced me to move it away from my precious SSD SAN. So I just stuffed the entire .bitcoin dir onto the network HDD array. It's the latency that seems to be the major issue. So I just bind mounted .bitcoin/blocks to the HDD and left everything else on the SAN array.

2024-09-15 17:55:46

So TIL that mounting the .bitcoin/chainstate to a high IOPs storage back-end seems to speed up the IDB by quite bit. I don't know that was my only issue, but it seems to be the major one. I also increased peer counts, and cache size (was never hitting a limit before anyway).

2024-09-15 05:15:02

- reply

Yeah that explains it thank you. So is there any way to acheive this while also having the squash feature, effectively 2 linear branches?

2024-09-14 14:48:23

I have a git question. I like using a development branch that tracks all commits individually, then I squash merge into master, then tags and CI and all happen on master. So this is basically the flow
git switch master
git merge --squash develop
git commit
git switch develop
git merge master
My issue is that the compare history does not collapse after merging master into develop. I don't want to lose the history of develop but I want git to ignore the branch history before the merge if that makes sense?

2024-09-13 14:20:00

I don't want nips to tell me how to design my data structures and API. Too many of you OOP devs leaking code abstractions into the protocol specs IMO. nip44 was a lot of work to abstract away from the API into an architectural blueprint. I enjoy reading scientific white papers and RFCs way more because it doesn't tell me how to write code. Even NIP01 couldn't explain the spec without code.
I'm just grumpy when reading
https://github.com/nostr-protocol/nips/pull/1494

2024-09-13 14:09:03

- reply

Can you see it connecting to relays?

2024-09-11 18:26:39

- reply

For nearly 7 years, every day myself and other company members woke up to emails, forum posts, facebook posts whatever calling us shit. Right now, if you Google search my name, one of the first results that comes up is a powerstroke forum "Extreme Frustration with (my old company)" calling the owners, myself, and our product shit an incompetent.
We knew what we were selling was pretty much a black box. We combated it with transparency, tons of money spent on podcast production, newsletters, and long form explainer episodes. I wrote tons of articles about how diesel engines function and what we do with them.
Beyond that transitioning to extreme support and raising the price of the product to explicitly compensate for getting me as your personal diagnostic assistant for hours per day.

2024-09-11 16:17:48

- reply

How all-in-one are you looking to get? And how much IT knowledge/staff do you have? Also what is the budget?

2024-09-11 15:38:42

- reply

> It’s all about tradeoffs.
For sure I just wonder how hard we should be relying on nip44 (I wrote the reference C impl) to protect us from ciperhtext hoovers and leakage, since it should be assumed the instance you publish your encrypted note its available to anyone in the world within seconds probably.

2024-09-11 15:15:27

- reply

Totally down to join a call on this!
For #2 I think all content should be in a uniform place, ideally that would be relays, but of course blob-sizes would hit a maximum, I'm not sure what limits are but I have dozens of source files over 100KB for what it's worth. As much as I like dumb clients (especially browser based), I understand the idea of reducing server dependence, specifically non-standardized communications.
I have a lot to chat about but my lack of in-depth knowledge of git makes it difficult. So many ideas, likely so few of them relevant XD

2024-09-11 15:00:48

- reply

So, I wonder over the long term how much information can be leaked because it's publicly available on relay-servers?
I've shared my concerns with nip46 signing with public relays and clients hoovering decryption requests to see when and where you are decrypting your secrets. Even cryptographically how much noise might be generated when the whole world has your ciphertext message and can see you encrypt/decrypt your "passwords". You can hope relays require auth, but should trust them.
Of course there really is no way around it in your app, but for me, the less personal information stored on relays the better. I deny all client requests to even store my preferences because just store them locally unless I ask otherwise.

2024-09-10 21:15:24

- reply

Any recommendations for desktop/server apps? I have yet to find any. I was hoping for a straightforward desktop app but that doesn't seem to exist, and likely most LN apps don't index well in search results. The GH awesome list is super outdated and mostly abandonware :/
Currently hoping to get a proper BTC node up along with cln

2024-09-10 00:53:33

- reply

Oh yeah! I would suggest anything to keep that drive pressure down (EGT primarily) to help keep them from cracking heads! I believe cracked heads (and valves) were more common in our shop than head gaskets. Cracked pistons are the next most common behind head-gaskets probably. I'm not sure what powerstroke is worse for piston failure but they were up there. Funny thing is, we used to dyno a few high horsepower (800+) 6.4s and they would lift heads (push cylinder pressure into the cooling system) when loaded with stock bolts or studded, and the gaskets would seal right back up at lower horsepower and be just fine.
I don't think it was uncommon to see over 80 psi of drive pressure, but it's been a while since I was looking at logs.
I usually say, I wouldn't expect to pickup much HP if any, but usually efficiency!

2024-09-10 00:41:07

- reply

Yeah the BD kits are good if you are staying close to stock power. They are one of the more professional companies we worked with in that space. Anything past like 100hp over stock on 6.4s I usually suggest a T4 setup if you can afford it to avoid excessive drive pressure from the stock compound configuration.
Are you running factory heads?

2024-09-10 00:16:37

- reply

Using a stock charger or going T4?

2024-09-09 22:42:41

@npub1f49…zpez How does your website follow button work?
I'm quite exited to listen to more episodes of bitpunk.fm! I can understand how much effort you have to put into each episode. Highly technical stuff is not easy but it's my favorite to listen to and was/is my goal with IO Bound. ! When I learned OTP it was a majority of the course, but basically learned that it's completely impractical and has many other drawbacks so, hopefully I can make time to read up.

2024-09-09 21:43:15

- reply

"Your mom's a hoe" I believe is the correct phrase

2024-09-09 19:14:33

- reply

OMG 100%. Everything I apply to, I'm like I'm only staying here to fund my ability to end this companies business model XD

2024-09-09 18:58:17

- reply

NP! The ELF stuff is way over my head, but they have done some pretty wicked OS "hacking"

2024-09-09 18:52:01

- reply

CC: @npub1f49…zpez I'm not sure if this is your space but figured I'd tag you!

2024-09-09 18:51:05

- reply

Okay, let me do some searching and digging and Ill see what else I have saved that might help!
Have you looked into FreeRTOs? I had to dabble the tiniest bit when working on a school project a while ago. Then I got into directly working with Atmel hardware bare metal. From 2018-2023 I spent learning early 8065 (pre x86 but tightly coupled to early stuff)
Something also worth looking into is the Cosmo LibC project https://github.com/jart/cosmopolitan. Justine is awesome and so is her discord! I mention this because most of her library is highly hardware optimized and plenty of asm to go around.
There are a couple other folks in the discord like ahgamut and paulclinger along with some other peeps from NIST and other orgs that are just awesome to read and chat with too

2024-09-09 16:23:20

- reply

They have and should have been for a long time.
1. Non-secret chats are not encrypted (maybe tls between devices and servers)
2. Messages are stored on servers in plaintext (or easily reversible encryption)
3. Over the past (almost decade) they have complied with governments by supplying chat data when requested (with some notable pushback, but yeah)
4. It's a free service (how are they funding running the service)? Most suggest that Durov was just paying for it out of his pocket but I highly doubt that math
5. It's a central point of failure, and relies and company hosted servers with proprietary software.
Like I've mentioned before, a company in order to exist must often have to comply with government requests. To me that's the reality. Those who say otherwise imo have never been told by a government attorney that "they don't want you in business by any means necessary".
Alternatives like signal, still mostly rely on a single companies existence, but offer
- e2ee private and group chats
- store little to no information on users or their chats
- open source a portion of their software
- Have a better history of non-compliance or more transparent compliance procedures
Telegram relatively convenient which I think aided in it's popularity, along with it's ability to have a big public API for bots and other fun things, that no other platform can do with encryption.