HNHacker News
TopNewBestAskShowJobs

ironman1478

1,250 karma · joined July 23, 2015

I write FW for camera systems.
submissionscomments
ironman1478··on Reverse-Engineering the Mali G78
You're probably right. But I'm just saying what ARM might be thinking.
ironman1478··on Reverse-Engineering the Mali G78
I haven't worked with their GPUs but I do work with their camera ISPs. I suspect that their reason might be that that seeing how to write software against the chip sort of gives away the secret sauce for how the chip works. I don't know if I would agree with that, but you can kind of intuit the design of chip based on its interface. How much values that provides is unclear though :/.

Also, they do charge customers for that code and it costs money to support/customize it work customer use cases. I suspect that's another part of the equation.

ironman1478··on On Leaving California and the Silicon Valley
I think the poor public transportation really hurts the city. The time to arrival is a crapshoot and discourages it's use. It's also not a walkable city or bikeable unless you stick to certain parts of the city. I think this + lack of density in certain parts of the city means that SF is largely a suburb with tall buildings on one side. It's not really a city, or at least not on the level of NYC. You are just surrounded by people in Manhattan at all times and I bet that helps promote socializing. You just don't really meet people in SF unless you stick to certain pockets of the city.
ironman1478··on No More Movies
I think if you only look at the biggest budget or most advertised movies you are right. But there are lots of great indie or lower budget, but not 'low budget's movies coming out. It's just hard to find these movies because they are drowned out by the noise of all the other ones. People said the same about music, but really I think things like discover weekly from Spotify and just general YouTube recommendations have proven that statement totally false and it's just people didn't have a way of finding anything.

Also, if you go back and watch some (not all) of those 90s movies some are pretty meh. Independence Day really stands out as being really boring and if anything the template for all the action movies you dislike. It has a very similar, shallow feel to it.

ironman1478··on Where are we going from here? Software engineering needs formal methods
Hmm. What if your unit tests are statements about your code? I view them as a way of locking in functionality and saying that the code does or doesn't do things for certain inputs. That's what unit tests are in my mind. I find it strange to view tests as only way a of catching bugs. You just use them to say what your code is/isn't capable of doing.

Using tests as a way of finding problems seems like not a great way of saying code has issues x,y, and z but they are a great of saying something does a, b, c. If you are going through a refactor you might not be implementing a, b or c correctly anymore and if your tests are designed well this will be caught.

If we are only talking about bug catching, you are right there are better tools. But I'll say if somebody says code does something and there are no tests demonstrating that fact, I don't trust it generally unless it's something like a game where it's hard to test certain things.

ironman1478··on Where are we going from here? Software engineering needs formal methods
How do you know it's working if it doesn't have unit tests that exercise edge cases? Or even basic functionality? Also, good unit testing can allow you to reduce the time it takes to write software by giving you an isolated environment to run your code.

Unit tests are a tool to get you to working code.

ironman1478··on Using CMake and Managing Dependencies
Python doesn't seem like a huge barrier to entry to me. CMake might be great in many ways, but the syntax is truly awful. It's just not intuitive to many people. I think due to how spacing works in CMake you can have incorrect things run and not fail. That's a huge issue, especially when onboarding people who aren't familiar with CMake.

Meson has some warts, especially around documentation but it's just easier and is just generally predictable. I've converted some CMake projects internally to meson and people were able to contribute and make changes with minimal learning and training, because the structure is intuitive. That means a lot.

ironman1478··on I Was Robbed in San Francisco While the Cameras Rolled
It's heavily resistant to change. It's really difficult to build new housing for example due to nimbyism. Many people are trying to remove the parklets and slow street programs. There is simply no progress in this city. It is literally conservative. I think you look at places like Miami and south Florida, they are a lot better at building new housing because they have vastly different policies. The really hip areas are definitely expensive, but a house is still generally attainable in the greater south Florida area.

I think SF wants itself to be perceived as an amazing place full of progress and inclusivity when it's really not that at all. It makes progress on civil rights only because the people of SF don't have to sacrifice anything to allow that to happen. But the second you want to change anything where they'd have to sacrifice for the collective good they fight tooth and nail. It's not all SF residents, but many are like this. You can hear it when the BoS accepts public comment.

ironman1478··on I Was Robbed in San Francisco While the Cameras Rolled
I think a big part of the SF hate is that it's ostensibly a liberal place, when it's really a conservative city. It has horrible housing policies and isn't actually racially equal when you look at the wealth distribution within the city. I think SF's external perception makes it a target for vitriol from right wing people on the internet.

SF has lots of problems to be clear, especially on the east side of the city.

ironman1478··on Colonial Pipeline Paid Hackers Nearly $5M in Ransom
Ha. The people that man the booths there are generally researchers for pen testing companies or universities.

Also, just a cool thing https://www.wired.com/story/evil-bubbles-industrial-pump-hac... There are like tons of attack like this that are possible and they demonstrate those at the booths

ironman1478··on Colonial Pipeline Paid Hackers Nearly $5M in Ransom
I don't have much to add here, but I've been going to Def Con and the other Las Vegas security conferences for a few years. Every year there is a section for infrastructure security (factories, refineries, etc). Its always the smallest section and the least populated. But its simultaneously the "most important" in terms of how much damage can be done from a single attack. Every year I went and was always terrified by all the stuff I saw because all the people hosting booths were like "yeah its dead simple to get in and break things." I feel like so many people could see this coming and there are just no consequences for the companies to incentivize them to do better.
ironman1478··on TimescaleDB raises $40M
Certain time series databases tend to me be optimized towards making the most recent data readily available and quick to fetch. There are also certain filtering / compression algorithms that are run on these time series databases that only make sense in a time domain.

Also, some of these time series databases have very specific use cases and you have to also think about the client tools associated with the database. Many of these databases sit in power plants, factories, etc. and they stream data to tools that are built to visualize or analyze the last few minutes of data and then trigger alerts based on patterns. Also, these database are very "device" aware and integrates with other systems that represent their data in a timeseries fashion already (like a sensor). A lot of customers who needed this type of database care only about this index because their concern is record keeping and monitoring. Not necessarily number crunching (this is changing though).

There are drawbacks to storing your data this way. If your primary index is time, it can be hard to merge that with some based on a coordinate system. So doing certain types of analysis is really difficult unless you replicate your data into some other database with a different index.

ironman1478··on TimescaleDB raises $40M
That's awesome. Y'all are going to crush it, especially since I saw that lots of companies have an appetite for adopting new backends, especially dbs that can work in both on-prem & in cloud.
ironman1478··on TimescaleDB raises $40M
I was employed there so I don't feel entirely comfortable stating their name. Just google around for "operational historian". It was really frustrating because we had these ideas years ago (and many other ideas), but due to reasons we were not allowed to actually make the product genuinely better. Also, the product is in more places than you'd think because they partner with vendors to sell the database as a component in a larger system that the vendor packages up and sells.
ironman1478··on TimescaleDB raises $40M
This is awesome. In certain industries (manufacturing, energy, etc) there are companies (really 1 company) that essentially have monopolies on time series databases / historians. There has been 0 competition in that space and as a result the databases and surrounding client tools are just so awful. It'll be interesting to see if timescaledb can really enter that market and force those companies to adapt.
ironman1478··on On incels, dead bedrooms and the hard problems of loneliness
I think the issue is how extreme the statement "It’s arguable she’s the only reason I’m able to be happy" is. I think there is nothing necessarily wrong with that if you find a partner who is comfortable with providing this author's needs. In my mind though that statement signals a codependent relationship which are generally not seen to be a great thing. I'm not saying they are in a codependent relationship as an FYI, but it sounds like something people in that type of relationship would say. Like can the person not be happy at all if they were not in a relationship? Can they not find satisfaction in their job, hobbies, volunteer work, etc? yes it might not be fulfilling as being in a relationship, but to have a binary happy / not happy state based only on being in a relationship seems extreme. This is just going based on how the author worded it. Maybe a better way to have written their statement would be "less happy".
ironman1478··on Google's push to bring employees back to offices, some say they'll rather quit
I really want to know this too. The traffic in the bay area is so awful and housing is so expensive that I imagine a lot of people are just brutalized by the commute, at least in the mountain view office (dunno what its like elsewhere). Even if you have access to the shuttle its still wasted time that a person will never get back and I think covid made that obvious to many people who might've just accepted the commute blindly (I did pre covid).
ironman1478··on Apple’s M1 Positioning Mocks the Entire x86 Business Model
Looking at core count and ram size isn't how to measure performance. You test and measure. But, the cores are just better on the m1 as has been shown through tons of benchmarking. Also, if the interconnects and busses are smarter (better cache coherency impl) then you'll be better in multithreading cases. Also because it's an SoC it has a way better implementation for the ddr interface (that's what they claim at least). I wouldn't be surprised if the latency hit for accessing RAM was way better than on x86 where you have to cross a long bus. For the workloads that people run on iMacs, light photoshop, some video editing, etc. We already know the new macbooks are amazing at it and this machine will have the same hardware, but with a dope screen (and cool colors. I want yellow >.>).
ironman1478··on Suicide rates are rising among U.S. nurses
Yeah I don't want to cast too wide of a net. My mom is a nurse in florida and isn't unionized and adjusting for cost of living, makes basically the same as a kaiser unionized nurse and has great job security. So it is not always necessary or they don't always work. But it seems like they are great for bringing the baseline work environment up.
ironman1478··on Suicide rates are rising among U.S. nurses
Because they are not the ones in power. The solution is to unionize. Teacher's unions seems to be unsuccessful as a whole (as an outsider at least). Nurses union's seem to be relatively successful in certain regions (the bay area seems like a good region. nurses make quite a bit of money here).
ironman1478··on Ask HN: How can a unhireable person get a job?
You've gotten a lot of advice and there is something I'll reiterate. Don't aim so high. I eventually got to the level of being pursued by FAANGs, but it took many years of working at a "boring" software company and just gaining skills. Also, on your resume try to make it outcome focused. Don't say what you did, say what the impact was. Everybody "knows" c++, docker, <insert technology here>, but what was the impact of what you did? Why do you know those things in the first place? That helps a lot when just sending resumes around and not having a connection. Also, you are studying all these algorithm text books, but first of all lots of companies do not go beyond simple graph theory questions. That was my experience at Apple (though Apple is very different from those other companies). If you are going for a specialized role like embedded systems or a backend software network engineer, the focus will be on things like "how does an OS scheduler work" not, "solve a dynamic memory problem". Also truly understanding the language you choose to program in is very important. It becomes immediately obvious when people don't really understand their tools, especially if they choose C++ (that's been my experience at least). You're over optimizing on the algorithms and not focusing on just software engineering. I am extrapolating this from what you said in the first paragraph.

Also, something that hasn't been really stated (and I feel weird saying this), but try to be funny (self deprecating is the safest way to go) and easy to talk to in an interview. Even though we all try to think of software as some pure meritocracy based on raw technical skill, it really isn't primarily because the technical part of the job is generally easy for the majority of positions and its communication where things break down. If you seem unapproachable or difficult to work with, then people won't want to work with you. I am noticing that at my current employer "soft skills" and ability to work with a team are becoming a bigger part of the interview process. When I do technical interviews if I can't converse with a candidate about how they got to their solution or their difficulties, its a bigger red flag than whether or not they could solve the problem or not (unless its clear that they really don't know anything).

edit: Also you mentioned something about long hair. I have long hair (a bit past shoulder length now, but its been a longer). Its something I worried about too. Nobody cares is what I learned. What people care about is if you are well kept. you can be well kept with jeans and a t-shirt and long hair.

ironman1478··on An RFC that adds support for Rust to the Linux kernel
To be fair, I think most C++ programmers are unaware of replacing global ::new and most don't know about placement new either (from what I've observed). I think lots of people who write library style code do (whether it be for boost, stl, or in their company) but outside of that most C++ devs are not library style devs and are a bit out of the loop on these topics.

Also, I agree with what you said early about RAII solving entire classes of issues that C (and C++ without RAII) has.

ironman1478··on An RFC that adds support for Rust to the Linux kernel
Personally, I don't think C++ would give a similar improvement to Rust or a future competitor to Rust. C++ has a severe useability problem. Many companies have to really restrict what subset of C++ is used based on use cases. I don't know what contributing to the linux kernel is like, but I imagine having readable code really helps. Paring down C++ to the subset that gives measurable improvements while also being readable seems daunting. Where do you draw the lines? Will concepts be allowed, what about template heavy code and SFINAE? If you go down the C with classes route (which doesn't provide too much advantage), it might be easier to for existing devs to work on it, but if you go too far the other way, you could alienate a lot of people who contribute to the kernel already and don't have the time or will to understand the complex template deduction system of C++.

Rust seems to be much more opinionated on how to do certain tasks, which might make it an easier to use for the kernel because there will be one way to do certain tasks. That's my read on things at least.

ironman1478··on Just Be Rich
You are truly out of touch with people who are in poverty or near it. My parents each work 60-70+ hour manual labor jobs not including brutal commutes. We were paycheck to paycheck for over a decade until my mom became a nurse (being able to pay for it was complete luck). My father didn't take a day off for over a decade include the weekends it felt like (obviously slight exaggeration) and injured himself multiple times on the job due to the stress. It's truly insulting and saddening to see this sentiment expressed. It is so out of touch.
ironman1478··on Just Be Rich
Sure. A quick Google search yielded the articles below. I don't have authoritative papers to reference, so you'll have to do your own digging too. But it also makes sense when you are aware of the realities of being in poverty. You also left out what I put in parentheses. Mental abuse is abuse and many children are subject to it. I explicitly called it out because I personally feel that people underestimate the damage it does primarily because it's something that is just completely alien to many people and cannot be related to. I also think people do not tend to realize what mental abuse is and in how many ways it can take shape.

https://stateofopportunity.michiganradio.org/post/poorest-ch... And this https://www.nber.org/digest/jan00/poverty-and-mistreatment-c...

Also, not to make excuses but I think in the mental abuse case people don't always know they are doing it. They do what their parents did or learned from those around them who also didn't know what they were doing.

ironman1478··on Just Be Rich
This. Poverty doesn't just cause immediate issues it can also cause long term issues that can effect not only adults, but the children who grew up in that environment. This is largely because families in poverty tend to also abuse their kids at higher rates (this includes mental abuse) or at least foster an extremely unhealthy environment due to the stress it causes. So even a generation later it can be difficult for a family tree to be lifted out of that situation. It's brutal not having a support system in place for people. What sucks is that it's probably bad for overall productivity to not help people in these situations, but the usa has chosen to say screw it for many years (hopefully that is changing).

Imo forget being "rich", just growing up in a relatively sane environment is an advantage.

ironman1478··on On Cache Invalidation: Why Is It Hard? (2018)
I think there are some embedded platforms that used message passing as a primary form of communication. However those are really out there and not at all mainstream. I think what you're talking about is coming. For example, for ARM there is a network on chip architecture that I could see being used as a back bone for this type of system. However, you need software support for it. It's not enough to just build the hardware.
ironman1478··on CEO of Waymo John Krafcik is leaving
Partnering with an existing car company helps a lot. They know how to build vehicles at scale and they also can provide a lot of support when it comes to interfacing with the control systems for the car. Controlling the car (engine, doors, wheels, wipers, lights, etc) is a surprisingly difficult task on its own. It requires a lot of control systems knowledge, a complex test infrastructure, and a lot of specialized talent that's probably hard to get. It could've been seen as too difficult to take on in addition to figuring out how to actually do all the self driving specific tasks (mapping, planning, detection, etc.).
ironman1478··on VW brand joins Audi in ending combustion engine development
That's extremely different. Those parts are made by company's that work with car brands to make compliant parts. Changing the software is a totally different problem.
ironman1478··on GitHub, fuck your name change
So, I was a part of a borderline poor family and somehow made it out into what is essentially an ivy but on the west coast (probably was a pity admission). Poor family's can have these values, however its simply the logistics and stressors. How do you take your kid(s) to a class outside of school hours to teach them how to memorize the answers to a test when you have multiple jobs, public transit sucks or non existent if you live in a place like florida (so your kid can't take themselves), and your kid is also probably ill equipped for that environment because your child's school is more of a day care and not a place to learn because we have given up on schools in the US. If you are a person of color, then its even worse because even getting into an OK school district can be extremely difficult due to Nimbyism which is a proxy for racism (and classism) basically.

Its not values. Poor people and especially poor people of color are setup to fail and its shear force of will that gets you out. I also don't want to discount the effort that middle class+ kids put in. They put a lot too and I don't want to say they aren't deserving. But the playing field isn't level.

← PreviousPage 8 of 11Next →