HNHacker News
TopNewBestAskShowJobs

shandor

693 karma · joined June 3, 2016

submissionscomments
shandor··on Rust and the Future of Systems Programming [video]
Well, one might notice "everybody" has to be a rhetoric exaggeration, since there's absolutely no way even remotely close to "everybody" would have any codebase at all whatsoever, not to talk about 20 year old legacy. Right?

Second, I was responding to a comment talking about "existing, critical, real time applications". Of which a huge number of cases do have existing, very old legacy codebases.

Third, I fail to see what you tried to bring to the conversation. If your only problem was with my rhetoric, see above.

shandor··on Rust and the Future of Systems Programming [video]
> IMO it would be great to get folks who write the enormous base of existing realtime apps driving critical devices everywhere to sit up and take notice of Rust.

It would, and definitely should, move into the direction of safer languages than C.

The biggest problem I see is tooling and legacy. Tooling, because there's a ginormous amount of testing and design software that "works with C" (whatever that means in the context of the tool). Legacy, because everyone already has their 20 year old codebases and it's just not convenient to start focusing on two languages and switching the old code to Rust is just plain impossible economically.

Third problem is compiler. LLVM (rustc is a frontend to it, no?) is really good choice, but gcc has enormous advantage in supporting so many small platforms and it's very significant in this area of SW dev.

On the plus side, I've really gotten the impression that the Rust folks are truly trying to make adoption as smooth as possible. If that works, and Rust proves to be much better than C for these kinds of systems, I'd guess adopting Rust rather than not would start looking economically viable to companies. I mean, in the end that's what matters to them the most, and it's not easy to replace all you C ninjas with competent Rust writer.

shandor··on DDoS attack halts heating in Finland amidst winter
Officials reported (sorry, no link) that they had "heavy reasons" to think that the attack was

1) done by actual criminals 2) pointed elsewhere, and the heating system was just one of the many systems used to initiate the attack.

They weren't more specific as to who attacked and who was the actual intended victim, but they were pretty sure the heatimg systems going down was just collateral.

Welcome to the Internet of Shit.

shandor··on A Method I’ve Used to Eliminate Bad Tech Hires
> we encourage candidates to use all resources on our take home interview test

This sounds great. In my response to another comment I was worried if accepting-but-not-encouraging this kind of behavior would put the interviewees on uneven ground. But if you tell them in advance that basically anything goes, it's fair for all parties.

> The real surprising part is how many take home problems we have received that either do not compile, do not pass the test data we give, or look like the instructions were completely ignored.

Sounds horrible. Though I guess that's only a good thing. Seems like the assignment is doing its job in those cases pretty well :)

shandor··on A Method I’ve Used to Eliminate Bad Tech Hires
You and GP both bring up a really mind-opening view into this. "Conversation catalyst" indeed, sorry but I'm definitely stealing that one.

> Outside of very specific roles, non-trivial software development is a team effort.

Couldn't agree more, and I have to say I'm rather baffled that so few recruitment processes are fine-tuned for this. Or maybe I should say that the image one easily gets is so. I myself have had pretty well balanced interviews that also tried to assess things like this, but I've always kind of considered myself just lucky.

When one takes your points into consideration, I can definitely see how having the take-home assignments offer value above the usual whiteboarding+bullshit bingo.

shandor··on A Method I’ve Used to Eliminate Bad Tech Hires
> We have an extension to the take home done paired on site that usually involves significant refactoring

That sounds actually brilliant. While some might argue that this comes back to the "whiteboarding under pressure", honestly if you have already written the code from scratch you definitely should be in a much better position than with the usual "implement a red-black tree insert, you have 15 minutes" whiteboarding tasks.

shandor··on A Method I’ve Used to Eliminate Bad Tech Hires
Sure I would, I have the $X/h to divide between them for helping me ;)

Seriously though, I guessed that many companies probably must have this kind of laid-back attitude on the matter, since it would be stupid to presume they are ignorant of my original point. Still, I think it might be a little unfair to the people going by the rules-as-written, which somewhat diminishes the consensus of awesomeness of this approach.

That is, if the interview process is not actively trying to choose for hustle... :)

shandor··on A Method I’ve Used to Eliminate Bad Tech Hires
This seems like a pretty popular take on the interview process here on HN. However, there's one pretty glaring flaw that I've never seen discussed: if the interviewee can take the problem home, what prevents them setting up a hackathon with their friends at their place, and making the interview problem a team effort?

A couple of knowledgeable friends can do wonders, and if you also prep with them a little you can probably answer all relevant questions and beyond about the code on the Monday on-site interview.

Sure, that's cheating. And even if it wasn't, I would never do it for the wrong expectations it would set upon me if I actually got the job. But the ease of "embellishing" one's CV always comes up in these threads/posts, and while I wouldn't do that either, I kind of fail to see how this home project approach solves that problem.

Anyone able to ELI5?

shandor··on Blockchain Healthcare 2016 Report – Promise and Pitfalls
I hear this from all over the world, and I honestly don't understand it.

Or rather, the only way I can see this happening is that everyone, everywhere on the buying side is either actively doing harm or so incompetent those are indistinguishable.

The arguments that "butbutbut, all vendors require us to do X and never let us do anything" got to be bullshit. If you have sexy multimillion medical project that needs vendors, and some of them won't take it if they can't force you into submission, I have no doubt that the million other companies in the world would take the project more than gladly.

shandor··on L4 microkernels: The lessons from 20 years of research and deployment
I won't pretend to having any kind of statistics on the matter.

But the whole point is that usually in embedded systems, there is no separation between "application" and "kernel", at least on the low-end of CPU power scale. To be able to isolate application level problems from kernel would already be a huge boon.

As a highly publicized anecdote, the Jeep hack of Miller and Valaseck was done by attacking through wireless, and replacing the CAN driver code to suit their needs. Not possible with proper isolation between critical system drivers and application layer.

shandor··on L4 microkernels: The lessons from 20 years of research and deployment
Sure, it's not a panacea. But at least it gives a much better foundation than what we usually are having now.
shandor··on L4 microkernels: The lessons from 20 years of research and deployment
I'm having really big expectations on seL4. The modularity of a microkernel with the security of formal methods is something that would give hope of solving some really fundamental problems with the "IoT scene". I.e. having good security out of the box (which is horrible in IoT as of now), while still maintaining good flexibility regarding platforms (which will become more and more of a problem in the future when IoT devices start actually proliferating).

Edit: oh, and you can still run seL4 as hypervisor and run Linux on top of it to get the usual Embedded Linux stack where your CPU has enough juice. And still keep the critical systems safe behind seL4's capability system. Best of both worlds? Dunno, hopefully seL4 will be there someday in production quality.

shandor··on Why are Adults so busy?
I have a feeling that we're talking about a little different things.

My meaning was that I rigorously try to put a price tag on my free time, just like my minutes are counted and paid for at work. This really helps me to put many priorities straight immediately. Like GP's "oh hey let's fetch gas from 39 mins away to save a penny" example, it is usually very easy for people to give zero value for their own time when making such decisions. Putting a virtual price tag for that extra 25 mins helps to prioritize those things in life that are all lumped under the category of "free time"; family, kids, friends, just relaxing at sauna, whatever.

shandor··on Why are Adults so busy?
Oh my god yes. I generally put a price tag on everything I do. Depending on ones salary it might be anything from 20 to 300 per hour. Then, if I see myself "making a loss" according to that calculation, I start paying for someone else to do the thing. Or, like in your gas example, definitely fill up on my side of the border.
shandor··on Some bad Git situations and how I got myself out of them
Yes, just rebase your whole feature branch on top of origin/master. Long-running branches with huge amounts of commits are bad anyway for multiple reasons, so it's not like you're bringing too many commits at once with any given rebase.

I'm not sure why you think this is worse to coordinate than merge commits? I think encouraging rebase-heavy workflow where feature branches are put into origin/master as often as possible actually improves the coordination quite a lot. When working that way, origin/master is always pretty recent, and anyone starting a new feature branch from top of it has really small risk of missing any big change sets from coworkers. On the other hand, I feel that merge-heavy workflow encourages people to have long-running branches other than origin/master, which hides commits from others.

shandor··on Some bad Git situations and how I got myself out of them
This is why rebase is so, so much better than merge. And it's infinitely better for bisecting, too.

Simple, completely linear history for origin/master is just so powerful in many that aspects that I'm constantly baffled why merging seems to be the flow mostly being talked about.

Now someone usually comes and says that merges are superior for long running branches. Which may be true in some aspects, but when you have more than one "running" branch and you got to try and find which commits have the code changes that only together seem to break and this is after 2 non-trivial merge commits, you really start to wish you'd gone with rebase.

shandor··on Some bad Git situations and how I got myself out of them
> Actually the easiest thing is simply not to care about how your log looks.

Yes, that sounds as easy as anything that accepts the idea of "let's do this the shitty-but-quick way now, and pay the technical debt later". So yes, it is "easier" to work with in the beginning.

On the other hand, if I had to inherit your repository that was developed by "not caring about history", I would probably have to hunt you down and kill you. So there's that.

shandor··on Why GitHub Finally Abandoned Its Bossless Workplace
This.

I previously worked for a company that does purely outsourcing work; they nowadays have a bossless hierarchy (apart from project leads, who are named anew for every project, and basically anyone willing can volunteer). It seems to work well for them, and I don't see why not.

Now I'm in a "product company", and the needs are completely different. Like has been stated elsewhere in this thread, when making products, there just has to be someone making the final decisions, i.e. being the product owner.

shandor··on Python vs. C/C++ in embedded systems
Another problem is power consumption.

Wit the current, appallingly bad battery technology the only way off-the-grid IoT can hope to proliferate is by keeping the power consumption as small as possible. This is of course a problem for a rather specific use case of IoT/WSNs, but one that gets completely destroyed if not taking this into account.

shandor··on In rich and poor countries, researchers turn to Sci-Hub
>there's absolutely no moral or philosophical justification for the ownership of information

Are you sure that the whole thing is really that simple? If information cannot be owned in any way, in any case, then don't the Udemy-using pirates from the othe HN story[1] have all the right in the world to do what they are doing? And surely it also means that for example all so called pirate version "China copies" of brands like Nike etc are also completely fine?

And it goes deeper than the megacorps. If no one "owns" any information, then no one can dictate how any single piece of it is used. So you don't get to say that NSA/Facebook/some random hacker has "no right" to hack into your emails and sell them to highest bidder. Or if you think you can, please lay out the logic chain leading to that, I wasn't able to come up with one quickly.

Finally, blanket statements lile this don't help anyone. I mean...your bank account and its contents are "info" nowadays, bits on some computers somewhere. Bits just like any other piece of digital information in today's world. I think we can safely say that is some information that is only fair to be owned by some, preferably the individual who earned that money.

[1]https://www.troyhunt.com/the-piracy-paradox-at-udemy/

shandor··on Why GNU grep is fast (2010)
Thanks for the really detailed reply.

I guess that explains my experience really well, since I'm always searching from a big amount of small files like huge codebases and config files, and almost never from files bigger than several megabytes.

ag has tremendously improved my quality of life in that regard, so please accept my sincere Thank You for making ag :)

shandor··on Why GNU grep is fast (2010)
Really? My completely anecdotal experience is that ag kills grep. Stuff like searching usages of variables in Linux kernel seems to take ages with grep, whereas ag search on the whole repository doesn't take annoyingly long.

I'll need to test this, just to see how wrong my "feeling" on the matter may be.

shandor··on GitLab 8.11 Released with Issue Boards and Merge Conflict Resolution
Thanks for the reply!

I'll see if I have time to evaluate yet more tools very thoroughly , but if I look into GitLab I'll definitely hit you with something for that page :)

one thing that jumps at me from those comparisons is that mostly all of them are against other tools that are mainly for repository hosting. Is it intentional, and similarly the main idea of GitLab itself?

shandor··on GitLab 8.11 Released with Issue Boards and Merge Conflict Resolution
I'm glad there's starting to be a good amount of choice in this realm of tools!

I've actually just lately been looking into a new all-in-one package like this for our company. Git hosting, code reviews, issue tracking, wiki, and nice project management features all in one place. My current systems-under-test are Phabricator and Jira.

Anyone have an opinion on how GitLab compares to those two?

shandor··on Int 80h (2001)
I came to post this exact thing. I guess the article is actually quite old. It talks people "migrating from MS-DOS to Unix/BSD" among other things, which sounds more than a little odd to be even from this millennium.
shandor··on Less stress, more productivity: working fewer hours is good for you and your boss
> turnover rates start to creep in.

This is the biggest thing baffling me on this whole thing.

I mean, it is absolutely critical for a startup to get their product out ASAP, right? And everyone can probably agree that having a high turnover of engineers is absolutely devastating to that goal, since one leaver's knowledge is a hefty amount of the overall know-how in a small company, right?

So,shouldn't these startups be holding on to their good hires like a drowning man to something floating, instead of whipping them to yet another death march?

shandor··on Think before you Mongo
> Why PostgreSQL instead of MySQL?

Why, is there something inherently better in MySQL that makes their second choice also suboptimal? Genuine question, I have very limited knowledge on DBs.

shandor··on A New Wireless Hack Can Unlock 100M Volkswagens
> Seriously... Why do so many people not care about security?

Seriously?

That's because as of today, not having security in this domain hasn't bitten anyone hard enough, ever.

Just think about it. At least I haven't heard that despite all these car key hacks any significant number of people would've had their cars and/or belongings stolen. Nor have I heard that any one car manufacturer's sales have significantly plummeted because their cars "can be hacled".

After that, it's just pure economics. As of now, the ROI for implementing good security in cars (or anything IoT for that matter) just doesn't pay itself back. At all.

Now, I'd like for the car makers to actually do something before chaos-by-hacked-cars ensues as much as anyone and people have to actually die. But it won't happen before the incentive is there. This is one place where government regulation really, really should come into play. There should already be laws for massive fines plus complete obligations to fix every security and safety related bug in all cars sold. Sound too hard on the makers? I don't think so, just look at the mechanical side of the cars: call-backs to fix those are happening all the time, because everyone understands that you can't drive a car which may hit you with an airbag at 60 mph. Or have a broken airbag system.

shandor··on Why is a Rust executable large?
> I've used Rust on a variety of ARM Cortex-M based microcontrollers including Atmel SAMD21, NXP LPC1830, and STM32F4

That's an encouraging start.

Having been there, and as reported on weekly basis by numerous people like @internetofshit, much of the code that runs the current IoT hype is utter and complete tripe. I have a faint hope that Rust could be something to help in that regard, even though at the same time I recognize that much of the badness in IoT has to do with financial considerations (too costly to make a good quality talking coffee-maker). Maybe it's especially because IoT seems to be all about getting cheap stuff out cheaply, I have this hope that at least having the language and development ecosystem help devs instead of shooting them in their foots would be a good start.

← PreviousPage 7 of 7