HN user

siffland

560 karma
Posts0
Comments166
View on HN
No posts found.

My issues is with false arrests due to people blindly following AI. People pulled out of their cars are gunpoint due to the AI misreading a single letter on the plate, the lady who was arrested for stealing amazon packages just because flock saw her enter and exit a neighborhood and she had to get doorbell camera to prove her innocence.

Also i think it should be public record about every single point of data flock collects and retention period. I am not saying they have to release the data itself, but I want to know what my tax dollars are collecting, how long they are keeping it, who can access it and who they are selling, er i mean sharing it with.

The issue is a lot of times it takes a senior level programmer to reason with a LLM to get the results needed. What happens when there are no Juniors to replace the Seniors. I guess by then AI will be able to program efficiently enough.

So far AI seems to be a great augmentation, but not a replacement.

When I saw Loki as the name, I instantly thought of Grafana Loki for logging. I click on the GitHub and get Libr-AI and OpenFactVerification.

I am not commenting on the actual software and I know names are hard and often overlap, but with something as popular as Loki already used for logging I think it might get confusing.

I remember play a game years and years ago (cannot recall the game), you come out of a mountain cave halfway up turn left and then enter another cave. However i stopped and you got a view of a desert landscape (i remember it as awe inspiring, maybe breathtaking, but it was years ago so in reality it was probably pixelated), it looked like someone put a lot of effort into the view and wondered if anyone else stopped and looked or just kept going.

Point is, if i work on code, i refactor and refactor and try to make it perfect even though it already worked and wonder how many people working on games spent a ton of time on a scene in a game or background or sprite to get it perfect and no one pays attention. I have a friend who plays Super Mario Bros all the time just to beat it and has NEVER seen world 5, 6 or 7 because he always uses the warp zones (i am not implying those are breathtaking worlds, if they were he would be missing out).

You would think a company would like to negotiate and be seen by a community as a positive company. I would not buy a product from them on principal after their statement. myQ could have engaged the home assistant maintainer and worked out, less API calls or something.

On a side note, i do love my home assistant, but ANYTHING that has to do with entry into my house is not and will not be automated, garage doors, door locks, etc. However that is my personal paranoia talking.

I think one of the biggest fears from ARM would be the popularity of the Raspberry Pi and their community.

There are better boards than the Raspberry Pi (strictly speaking specifications here). I took that path of playing with a lot of alternative boards, my biggest issue is lack of support, some boards had kernels never updated, etc...(YMMV).

If Raspberry Pi released a RISC-V board I have no reason to believe the community would not be just as strong. Sure in the beginning they would support both, but eventually the ARM support would wither.

I have no idea why this is fun, but on AI chat-bots, i always test

2 + 2(2-2)

Should be answer of 2, however on different bots (not just GPT) i have gotten 0, 2, 4 and 6 (all i can understand except the 6).

So yeah.......math messes with some bots, who would of guessed.

Working in Tech, I love a lot about electric cars. I just have a few issues I cannot seem to overcome (this is a personal ME thing about EV's and some new gas vehicles, your mileage may vary....pun intended).

All new cars spy on you (well most do), but EV's are completely software driven and seem to be more intrusive. I want a car I can connect to the internet when I want to for updates and that is it. I also want android auto and the apple one for my wife, I do not want a proprietary (Tesla) entertainment system.

I want right to repair. It is my car if I spend that much on it, I want to do to it whatever I want.

I want all the equipment in the car to work when i purchase it and no monthly fees (I get it for XM Radio, I am talking like seat heaters and such).

For some reason it also irritates me Tesla can increase mileage with a flip of a switch (Like they were going to do in Florida for a hurricane a while back). So you are telling me they could of been build with smaller batteries and save money and the environment, but they were not so you could have a software upgrade (or was that a monthly payment as well). Also the fact Tesla can remote disable my vehicle, and purchased upgrades do not transfer to the new owner.

I can probably go on and on with my whining, but that is why i still have 2 gas guzzlers and a motorcycle (newest vehicle is 2008, 2005 Honda van has 270,000 miles on it). They are paid for an yes a little maintenance here and there but that is not enough to make me spend $30k + on an electric vehicle.

I have been using FreeBSD since 1999, it was version 3.somthing, cannot remember. Since then my main server/development box has always been FreeBSD (I have a Linux one as well for some software that just will not run). Somewhere along that path i became a Linux admin (and HPUX but that was long ago), starting with RHEL3 (I was using gentoo at home). I have a bunch of servers, FreeBSD, Linux and yes windows. They each have their place.

FreeBSD for me is simple. I have 2 Raspberry Pi 3's with 512MB of ram running FreeBSD 13 and Unbound and a Pi 4 Running FreeBSD and Asterisk. I forget they exist. They literally just keep running, I update them and if needed reboot and they just keep going. You can use mtree to verify the OS, Security is good. But some things like DNS, i do not play with, just let it run (and keep it patched).

I have a full blown Ubuntu Kubernetes cluster and a few RHEL and OEL vm's running various tasks in the lab. Most are the right tool for the job i need it to do.

I have never understood the flame wars for my distro (Linux vs BSD) is better than yours and my license is better (or allows more freedom) than your. They each really do have their places with some expected overlap. I am just happy with the variety that allows me a ton of choices and usually drives innovation.

I think where the difference comes in with vinyl is the mastering process. CD's/Digital and Vinyl usually go through two entirely different mastering processes. This will make listening to the two sound different even though the mix is the same, kind of like listening to your favorite song at home from CD, then hearing it on the radio with a more in your face sound because of brick wall limiters in the sound chain.

I am not going to debate if one is better than the other, but they will sound different just due to mastering. So it is not a direct apples to apples comparison.

I think most people buy video game vinyl just to display them, haha.

I hate the network switch debate. Most center around noise shaping the sound.

It is a one or zero, it will checksum and TCP either gives the correct packet or drops it. Also I do not know many audio streamers that do not buffer since TCP works way faster than a song can play. Plenty of time to make sure the packets are correct.

I have no idea why the audiophile network switches irritates me the most. At least with audio cables I guess you can do measurements that show the signal is cleaner, even though there is zero chance of you hearing a difference most of the time (I just make sure cables have good shielding and if the cable has to be seen it better look cool too, haha).

Ya know there are even audiophile USB cables.......not a joke google it.

When I am writing code for something it is fun to see what it comes up with as opposed to what I did. I also find it fun for bouncing ideas off of while brainstorming (it does not judge me for stupid ideas). Most importantly I taught it the proper response for when I type.....who ya gonna call.....

You are probably correct. Again people will bet on current system not being in use 15 years from now and it is not a problem until it is a problem.

On a better note Y2K brought us the movie "Office Space", so perhaps Y2K38 will give us another equally awesome movie.

Hindsight is always 20/20, we all know now that it would of been a bad idea. I do think this, along with Project Orion, are great thinking by people to utilize terribly destructive weapons for outside the box peaceful solutions.

Again we know now this would be bad and the consequences would not outweigh the benefits. I do think it is vital to have this kind of thinking (I mean outside the box). I think some people may find it crazy that this was even thought of, however I find is an interesting read.

At my first UNIX IT job, my team lead told me:

Listen when you break something tell us. Explain what you did and we can get it fixed. You will make mistakes, we all still make mistakes.

That is what i always do and expect my coworkers to do. Yeah we talk a little smack afterwards (only jokingly), but were able to resolve most self inflicted wounds quickly. Nowadays what scares me more is a coworker who hides mistakes rather than one who confesses to them.

From the beginning of Byte magazine until about 1984 are really fun to read. The ads were probably annoying at the time of release, but not they are fun and interesting to read. I like the 1983 issue with imaging where they are talking about movie effects for "Tron" and "Revenge of the Jedi".

I wished they would of had more UNIX in the issues, however it was usually unaffordable. But a fun look, at least for me, to the history of computing. And who does not like an ad for a 8MB, $3999 disk drive subsystem.

GitHub was down 6 years ago

Every project i have been on where we built for availability and resilience has inevitably had at least one single point of failure. Usually it is something deemed non critical, but somehow can still bring the infrastructure down (A single DNS server at one of our production sites did this, we have 2 more accessible via a VPN tunnel, it was deemed if the production DNS went down the other two were still reachable, to bad the day it happened the tunnel was down too).

Also you have to deal with sysadmin error, i know us sysadmins are practically perfect in every way, but occasionally we make mistakes....big mistakes. ;)

So redirecting might not always be possible.....

Funny story, about 6 years ago we got a HP DL980 server with 1TB of memory to move from an Itanium HP-UX server. The test database was Oracle and about 600GB in size. We loaded the data and they had some query test they would run and the first time took about 45 minutes (which was several hours faster than the HPUX), They made changes and all the rest of the runs took about 5 minutes for their test to complete. Finally someone asked me and i manually dropped the buffers and cache and back to about 45 minutes.

Their changes did not do anything, everything was getting cached. It was cool, but one needs to know what is happening with their data. I am just glad they asked before going to management saying their tests only took 5 minutes.

You can lock down YOUR infrastructure, but then are 100% dependent on the cloud environment to maintain and patch theirs. Since you have no insight to what the underlying infrastructure consists of you really have no way of knowing if they are secure. Do their storage arrays have open CVE's? are they employing people who are mentally sane? You just need to trust them.

So in the cloud just migrates the a lot of the security to another team. I do not know for a fact, but I am pretty sure the DoD cannot just show up at the AWS or Azure facilities and start auditing them (maybe they can and it is in a contract, someone else might know).

This was an accident, they meant to buy 2 blades for more SAN ports and they ordered the wrong parts.

They did this with a few database servers as well, they were supposed to order 8 core processors, but they ordered 10 core instead (it was a typo), they had 4 processors per machine it was a $40k extra mess not including the extra Oracle licenses needed to buy since Oracle licenses by cores (8 extra cores per machine and 3 machines).

But i do understand the spend it or lose it, which is another reason no one tries to be efficient.

After 15 years of being a DoD contractor, it is frustrating to see yet another sole source entity getting the contract. Prices will inflate and there is no competition.

HP and Oracle have been reaping the benefits of this for at least two decades now. We put databases on Oracle that should be handled by Postgres or MariaDB since DoD prefers Oracle. We would buy useless HP software because we were a HP shop. I fought to get a non HP solid state array for our data, it was an epic battle (in the end I won, on the extreme we had 6 to 7 hour processes cut down to under an hour, the HP equivalent could not replicate that at the time).

So i can see DoD moving to Azure and then get the vendor lock in and in 10 years if they want to move the cost will be so extreme it will either cause taxpayers a ton of money or not be realistic.

As impossible as it sounds, and somewhat impractical, i would rather see a vendor agnostic approach and DoD spread across multiple gov clouds. i guess years has gotten me jaded with government spending (wait, what, how did we buy 2 extra $50k Cisco chassis and then keep them in storage for 3 years....).

I agree Nomad is probably a better solution, however my friend is an Oracle DBA who now manages the servers and docker since their System Administrator left and he was thrown into it.

I was addressing the:

"The takeaway is: don't use Kubernetes just because everybody else does."

line in the article, and agreeing (did not mean to start a holy war about swarm).

On a side note, I used to run a docker swarm at home up until about 4 months ago when I switched it to K8s, I really didn't have any bad mesh routing issues, and it was pretty stable. But to be a hypocrite I switched to k8s because everyone else DOES use it and I wanted to kind of stay relevant.

A good friend and old coworker has been hitting me up for how to do stuff with docker since his job now uses containers. He has about 15 containers (on 2 hosts) that do not change to much and was asking about setting up k8s (it was a buzzword his manager heard). i talked him into just setting up swarm.

It took all of about an hour (over text messaging) to get it set up and all stacks/services running. They could not be happier.

It comes down to the right tool for the job. If you don't need all the bells and whistles, then keep it stupid simple. I realize swarm is not a 100% "enterprise" solution, however before they they were just issuing docker command after each reboot.

My wife drives for Lyft, she does not have to, she is an extrovert and loves to talk to other people. She puts up with her share of crap, people puking in her van (a lot) lots of drunks, guys hitting on her (usually drunk), you name it it happens. (disclaimer: she drives weekends from about 9pm to 4am in a van for plus size rides (rides over 3 people), so some could consider a lot of this self inflicted on her part, but someone has to drive the inebriated).

She drives to talk to people and make a bit of extra money, it was never meant to be a job to support us. I personally think she is crazy for putting up with all she does. Anytime she is waiting for passengers or going through a drive through (for the passengers) for 15 minutes she makes NO money, because the ride price is fixed (and yes after events passengers will get in the van and one will be outside talking to someone else and she has to wait for them to get in, not getting paid for her time). She tells me all the time, a few riders are really grateful, but most don't care and the driver is just some person they hired (she has hard candy and the large things of bubble gum for riders, a few weeks ago a group of guys stole one of the things of gum, she saw them walking away with it and laughing, $6 down the drain).

There are also the passengers who order the Lyft, she shows up and they are 2 streets over from where they said and tell her to go to a different location. Again, she could cancel the ride, but unless someone is in the same location waiting for a ride she would be losing more time and money.

So i think charging for ride time is fine. If the driver does take extra turns and time for more money you can always give them a lower rating, i am unsure about Uber, but the driver rating at Lyft really does mean something.

I would not get it for this exact reason. However to counterpoint your argument this will work for some people like my high school kids.

They stream spotify because they don't want to buy songs and after that current song is no longer popular they move on to another one and forget the first song ever existed. They play video games and when they are done they sell them and buy a new one, never wanting to revisit the old one. They spend $50 on the new game and sell it for $15ish. For that mindset of people this is a great product, along with the digital only xbox one.

I am not implying all kids or youth are like this, mine are and a bunch of their friends are. I prefer gog or a physical copy, but then again i don't buy that many games.