lsofzz

> Thank you. I use most of this, I've been using it for years and I just don't talk about it because it's hard to argue when people just want to force an idea that k8s is "really the best way of doing things".

I wouldn't call it the best way; Rather a good way because Kubernetes does encapsulate the really good bits from scalability, development, security and reliability aspect. It's not a panacea but if you have team bandwidth to run k8s cluster, it's definitely worth a look.

<3 HAProxy. It's a solid piece of software tested to the teeth over the years and is great; Here's the thing: You can run it as your preferred ingress controller too :) https://www.haproxy.com/blog/dissecting-the-haproxy-kubernet...

> So if you'll indulge me -- this list is exactly why a system like Kubernetes is valuable and why I think personally that it contains a lot of essential complexity.

Yes. I would agree to your statement precisely as an answer to @jen20.

Some things such as getting stateful systems, HPAs and persistent storages were a little tricky initially but a breeze after.

But I do want to mention that you really really need a team to look after it. Without it, it'll bound to be another snowflake.

[edit]: <sigh/> i meant to say stateful when i wrote stateless.

> No. We can fix things one at a time. Master/slave doesn't upset me but it does upset some people. Ok, if they're willing to put in the effort to update the code and all the docs then I'm willing to adopt the new terminology.

I have nothing against change but _context_ matters. When I read about `pty` masters or slaves, I think of pseudo-terminals not humans.

> Good thing I don't have a master's degree.

Hah. Exactly.

I'd also say `man` from the unix manual <- yep, would you consider that to be the elephant in the room?

If I had to hazard a guess, we are all a bunch of technologists/engineers/tinkerers who deal with technical stuffs; not some societal driven definitions and constraints that supposedly define what `man` is in this context. We bloody well know what `man` is in this context - manual pages. For some odd $deity fore shaken reason, please don't try to bend the meaning of it.

> That's the beauty of identity politics. No matter what you say, you are always going to encroach on someone's identity.

> Whoever promoted it, was a machiavellian genius. People can't rebel if they keep each other in check.

That is exactly what I wanted to say too but you said it better.

I'd point out that there are possible hundreds if not thousand other software that use the terminology. It doesn't offend me at all.

$> man -k slave grantpt (3) - grant access to the slave pseudoterminal grantpt (3p) - grant access to the slave pseudo-terminal device jack_netsource (1) - Netjack Master client for one slave kdeinit5 (8) - Launcher for applications built with kdeinit support, ... ptmx (4) - pseudoterminal master and slave pts (4) - pseudoterminal master and slave ptsname (3) - get the name of the slave pseudoterminal ptsname (3p) - get name of the slave pseudo-terminal device ptsname_r (3) - get the name of the slave pseudoterminal Tcl_CreateSlave (3) - manage multiple Tcl interpreters, aliases and hidden c... Tcl_GetSlave (3) - manage multiple Tcl interpreters, aliases and hidden c... unlockpt (3) - unlock a pseudoterminal master/slave pair unlockpt (3p) - unlock a pseudo-terminal master/slave pair

$> man -k master agentxtrap (1) - send an AgentX NotifyPDU to an AgentX master agent getpt (3) - open the pseudoterminal master (PTM) gnutls_session_ext_master_secret_status (3) - API function gnutls_session_get_master_secret (3) - API function gnutls_session_set_premaster (3) - API function jack_netsource (1) - Netjack Master client for one slave mmafm (1) - creates AFM font metrics for multiple master fonts mmpfb (1) - creates single-master fonts from multiple master fonts ptmx (4) - pseudoterminal master and slave pts (4) - pseudoterminal master and slave pulseaudio-ctl (1) - Control pulseaudio's basic functions such as the maste... RAND_DRBG_get0_master (3ssl) - get access to the global RAND_DRBG instances SSL_get_client_random (3ssl) - get internal TLS/SSL random values and get/set... SSL_get_extms_support (3ssl) - extended master secret support SSL_get_server_random (3ssl) - get internal TLS/SSL random values and get/set... SSL_SESSION_get_master_key (3ssl) - get internal TLS/SSL random values and ge... SSL_SESSION_set1_master_key (3ssl) - get internal TLS/SSL random values and g... Tcl_GetMaster (3) - manage multiple Tcl interpreters, aliases and hidden c... Tk_GetImageMasterData (3) - define new kind of image unlockpt (3) - unlock a pseudoterminal master/slave pair unlockpt (3p) - unlock a pseudo-terminal master/slave pair WildMidi_MasterVolume (3) - sets the overall audio level of the library. xapian-replicate (1) - Replicate a database from a master server to a local copy

> Unix is seriously uncool with young people at the moment. I intend to turn that around and articles like this offer good material.

I offer you my well wishes on that.

I think this reply needs more up votes! ;)

Why not a littel bit more? ;)

  set -euf -o pipefail

> I strongly disagree although, just like you, I have no evidence to support my claim

Haha. Fair enough!

> This is where you need to use awk.

I think need is a strong statement in this context.

  echo $(( 1 + $(grep 'cpu cores' /proc/cpuinfo | head -1 | awk -F ':' '{print $2}') ))

Choosing shell vs. [insert modern] programming is a matter of trade-off of taste, time, $$$, collective consensus (in a team setting) and so forth.

[edit]: Forgot code syntax on HN.

> - Bash is awful (should be replaced with Python, and there are so many Python-based shells now that reify that opinion)

You know that no one uses Bash these days right?

Fish shell, Zsh are the much newer and friendlier shells these days. All my shells run on Fish these days.

I have to tell you that my shell ergonomics has improved a lot after I started using fish shell.

But to your point, _why_ is Bash awful?

> http://langsec.org

The maintainer apparently has no transport security concerns. It takes literally two minutes to set up TLS these days

flatpak/snap applications can run on its own confined sanboxes

COBOL

From watching the presentation on Youtube (https://youtu.be/sQCgtCoZyFQ), it seems NiFi is geared towards acquisition of data, gluing/batching/massaging the flow between systems and providing the necessary interfaces to downstream systems;

- Would love to see the ability to develop custom NiFi processors in Go/Rust/Elixir etc.

- XML is a big pain in the rear.

- Being container-aware is big win. Stateless is even better.

I see a good opportunity there for users like me to explorer NiFi's capability in the future.

> They both schedule jobs and move data according to control flow topologies that you build in a GUI, right?

Airflow on the other hand is designed to run scheduled jobs (whether it be batched or otherwise). The 'job' can really be anything - build / data processing pipelines, system configuration management pipelines and so on. In Airflow parlance, one can create connected DAG's as pipelines that massage the data in a way you intend it to.

They both share some commonalities but I do gravitate towards their use cases being subtly different and an important one highlighted above.

> This time, AMD may have stepped up with a steel pipe. We'll see how well they can use it...

:-))))

Expanding on it further, what I mean is that our BT stack should be built from ground up with security in mind. We cannot fight every attack surface but we probably can get to all the low hanging ones that the crims can exploit.

One day in some utopia probably.

I as a user of BT stack do not have to worry about whether it is switched on or off. This is what I mean by _secure by default_.

Downvoter - thank you. I took time to explain what I mean this time :)

> You can disable Bluetooth by default and enable it only when used.

You are missing the point.

Ever heard of secure by default?

> Thousands do care. So here's what I mean: When you take the ratio of the thousands that do compared to the millions who don't, you get a number that's pretty close to zero (one tenth of a percent, back of envelope).

Fair enough.

> And it certainly isn’t dead.

Or to put it another way, it's alive and kicking ass :-)

> If you define hacking as controlling your computer hardware at the lowest level, then hacking is dead because no-one wants to write device drivers.

This is absurd. Dude, let me be clear - Just because you write Java all day long does not mean that there are hundreds if not thousands of embedded system developers who engineer (quite passionately!) the thing that powers your SSD or your wireless card or the thing that heats your home.

We often take these substrates (in this context - the firmware and device drivers etc.) for granted in the modern hardware and often fail to even acknowledge them. I am not one of them and I vehemently defend these silent heroes of our generation.

> There was a term from the early 00's for malicious "hackers", which was crackers.

Yes.

> But if the media is unwilling to budge should we just make a new word?

Not sure how that will pan out :-)

> 1. I have no idea why "hacking" aka "going fast and breaking things" is so glorified while "building good reliable programs" aka "good programming" is not.

The way I see it, `going fast and breaking things' is an iterative process. IMHO, it should not be viewed as `go fast, break things and _never_ iterate to fix things'.

Just thought I'd toss my two cents on it.

> 2. How come that someone thinks of himself/herself that he/she has the right to say who is a hacker and who is not.

No idea why you feel binary. If you like to call yourself a hacker - fuck it - go for it.

The connotation in the beginning were that hackers were pioneers, thinkers, achievers and great motivators. Over time, the meaning has been mutated and engineered by the media to mean `malicious' and `conniving' thieves/criminals.

To this day, I still stick to the original definition of a `hacker'. I don't give a two fuck about what the media likes to call it.

> OK: it's fairly trivial for someone like me to do this, and you don't even need "ML" to do it. It's just counting, binary trees and simple models.

Would you be interested in PoC'ing it out for such a trivial project?

> I do not have direct ML experience at the moment. I’ve only run simpler regressions. I was making an educated guess as to the tech which may emerge, and it doesn’t seem far-fetched since it probably would not require language comprehension or higher order reasoning. I would love someone who’s worked with ML to weigh in.

Aha. Indeed. I would prefer to have someone actively working in ML to weigh in as well.

> Sherlocked :-)

Hehe

> Train it on content that you know was produced by someone, then unleash it on the rest of the internet to find content that they wrote anonymously.

I apologies but that's disappointingly vague..

> What next? A publicly available ML web crawler that analyzes speech and belief patterns, triangulates them with metadata, and returns someone’s identity with 99% confidence?

Close but no cigar.

But seriously, correct me If I am wrong but if I were to propose a solution to defeat this `ML web crawler' confidence in it's results, then the first thing I would do is to feed the internet with invalid data. How does ML-based analyzers deal with this kind of data set?

> Which is basically TOR

Last time I checked Tor's technical paper (probably been 10 or so years), it stated the fact that a relay? node in between can only decrypt the information required to route to next hop and not the actual packet's payload. Is that correct or am I dreaming?

Also, I recall that if someone malicious flooded the tor network with malicious exit nodes, then all traffic details can be `inferred' right? i.e., the assumption was that exit nodes need to route the packet to destination thus it needed to look at every packet but couldn't infer the originating IP address (based on my dated knowledge :))

If a malicious person now floods the tor network with bazillion exit + relay nodes - then essentially all contents (payload + IP src/dst) can be aggregated. Is this still problem in tor network?