I hope you are not working on safety-critical systems then.
This is where things get really silly.
If an aircraft crashes because its software fails, 2-300 people may suffer a negative outcome.
If systemically important business systems fail, 2-300 million people will suffer a negative outcome, up to and including loss of property, health or even life. (Riots are indiscriminate)
Further, what other disciplines allow for the title "engineer" without a state certification/licensing board?
My country doesn't have states.
Anyway, software engineering is already a well structured and impressively defined discipline given it's existed for around 60 years. Civil engineers took more centuries than that to get to this stage.
I love that people can and do undertake this type of thing. It's good for the development and improvement of the state of the art, continued research is valuable purely as research and I don't really want it to stop.
It's also so far removed from pragmatic non-academic reality that it's insane. Three years, and found two bugs? Let alone the obvious risks around the complexity, completeness and accuracy of a model that took so long to develop.
The EAL link is interesting though, hadn't come across that before. I suspect most of the systems I work on would qualify at EAL4 but the cost of demonstrating it would kill their commercial value; we'd discontinue the products rather than waste the money.
EAL7? Comically that's more likely - but only for key algorithms we develop and use. A whole system? Could take down the company..
Oh, my mistake. I was unaware that only questions about the US presidency were allowed on this site.
In a discussion about Donald Trump's knowledge, the context is the fucking US presidency.
Or, alternatively, your response is the exact reason why the question is relevant.
Thirdly, you don't need a degree in biology to answer a very straightforward question.
I'm not sure a degree in biology would help know the scientific name for any chicken, let alone how to differentiate between several hundred fucking breeds.
Face it, you asked a stupid question and you got a more dignified answer than you deserved. Just what the fuck is your point, because it's completely passing me by.
constructed a formal model of the PikeOS separation kernel
How long did this take? Did it actually add any value? Would it have been quicker to just rewrite PikeOS? Were the people involved recruited in India and brought into the country on a dodgy visa scheme?
Sorry, just trying to translate this to the real world.
No, he's not. He's saying it's possible to create an entirely secure piece of software that's utterly fucking useless, and that it's possible to create a seriously awesome piece of software that transcends mere concepts like 'executing properly' without it needing to be secure.
You'll note that a correct program is necessarily secure in having no behaviours outside its specification.
Ok, write up the specification, in full.
When you're done we'll show you how much money your main competitor has made with their shitty implementation in the meantime, the three open source tools that replicate the key functionality and the two new industries that have been built up in response.
The market's already assessed 'secure' versus 'functional' and the decision remains a continuum between the two.
Or as I tell the fuckwits working in Infosec at my employer, if they want it secure I can hit the shiny red button in the data centre and shut the lot down.
No? Ok, you don't want secure, you want a compromise. Now we can discuss it.
So next time you hear Linus Torvalds rant against C++ tell him to fuck off and stop being an idiot.
Why would I want to? He's clearly not an idiot, I owe him for the work he's done to build a reliable kernel I use daily, and C++ is indeed a fucking stupid language.
I've yet to see any C++ code base that is bug free, mainly because inexperienced people write bugs and experienced people write so much fucking scaffolding code to try and prevent them that the code becomes unmaintainable anyway.
Good experienced people switch to better languages.
Sure. Lets ditch double-entry book keeping and make accountants add up properly. Lets ditch medicines and make doctors heal people with their own skills. Lets ditch case law and force lawyers to properly argue why their clients are innocent.
Or maybe, just maybe, professional software engineers take full ownership, full responsibility and exploit the tools and systems available to them to maximise the chances of good outcomes without adding unsustainable cost or time burdens on their work.
I'm not a biologist or a natural historian. Let me consult an appropriate reference or expert, and while I'm doing that, perhaps you could tell us all just what the fuck that has to do with being president of the US?
The Linux kernel has had 2 1/2 decades of continuous oversight, guidance, QA, design and discussion.
That it works at all is testament to the doggedness of the team that develops and maintains it. That it's the most used operating system on the planet is testament to their engineering excellence.
So, wondering when the masses stopped caring at all about how the music sounded?
When I was diagnosed with partial deafness at age 12. When I was issued hearing aids in my 20s. When I gave up on them and decided to live life in the quiet in my 30s.
Are Klipschorns real or was that an elaborate piss-take on audiophiles? Sorry, reality blurs with satire on this topic.
I lost patience with the NAT approach. I'm not a Windows admin, a network specialist or a virtualisation expert so I decided to defer the day or two of learning and experimentation for when I have energy and time.
Or Microsoft could fix the shitty hypervisor. Seriously, when it's easier to download software from Oracle you know there's something broken.
I've heard of Jello Biafra and Tipper Gore. I didn't know they were related.
Why would I be culturally unaware for not giving a flying fuck about some American punk band? We had The Clash, The Sex Pistols and The Wurzels. Wait, no, skip that last one.
You'll be telling me off for not knowing what sort of music The Stringcheese Incident play next.
Slashdot Mirror
I hope you are not working on safety-critical systems then.
This is where things get really silly.
If an aircraft crashes because its software fails, 2-300 people may suffer a negative outcome.
If systemically important business systems fail, 2-300 million people will suffer a negative outcome, up to and including loss of property, health or even life. (Riots are indiscriminate)
I work with the latter..
Further, what other disciplines allow for the title "engineer" without a state certification/licensing board?
My country doesn't have states.
Anyway, software engineering is already a well structured and impressively defined discipline given it's existed for around 60 years. Civil engineers took more centuries than that to get to this stage.
I love that people can and do undertake this type of thing. It's good for the development and improvement of the state of the art, continued research is valuable purely as research and I don't really want it to stop.
It's also so far removed from pragmatic non-academic reality that it's insane. Three years, and found two bugs? Let alone the obvious risks around the complexity, completeness and accuracy of a model that took so long to develop.
The EAL link is interesting though, hadn't come across that before. I suspect most of the systems I work on would qualify at EAL4 but the cost of demonstrating it would kill their commercial value; we'd discontinue the products rather than waste the money.
EAL7? Comically that's more likely - but only for key algorithms we develop and use. A whole system? Could take down the company..
Without knowing the software or its quality, it's hard to say. Taking the general case though, yes, you're right.
But you're merely adding even more reality into the conversation.
Oh, my mistake. I was unaware that only questions about the US presidency were allowed on this site.
In a discussion about Donald Trump's knowledge, the context is the fucking US presidency.
Or, alternatively, your response is the exact reason why the question is relevant.
Thirdly, you don't need a degree in biology to answer a very straightforward question.
I'm not sure a degree in biology would help know the scientific name for any chicken, let alone how to differentiate between several hundred fucking breeds.
Face it, you asked a stupid question and you got a more dignified answer than you deserved. Just what the fuck is your point, because it's completely passing me by.
constructed a formal model of the PikeOS separation kernel
How long did this take? Did it actually add any value? Would it have been quicker to just rewrite PikeOS? Were the people involved recruited in India and brought into the country on a dodgy visa scheme?
Sorry, just trying to translate this to the real world.
There are plenty of stupid programmers out there, and, sorry to be blunt, Dunning-Kruger means that you're probably one of them.
Nope, I've been saved from that by the Peter Principle.
Yeah, stack traces are so much worse than core dumps or a complete absence of information regarding system failure.
Shit, Java has many flaws but the average corporate software developer has many more, and at least Java protects you from several of them.
No, he's not. He's saying it's possible to create an entirely secure piece of software that's utterly fucking useless, and that it's possible to create a seriously awesome piece of software that transcends mere concepts like 'executing properly' without it needing to be secure.
You'll note that a correct program is necessarily secure in having no behaviours outside its specification.
Ok, write up the specification, in full.
When you're done we'll show you how much money your main competitor has made with their shitty implementation in the meantime, the three open source tools that replicate the key functionality and the two new industries that have been built up in response.
The market's already assessed 'secure' versus 'functional' and the decision remains a continuum between the two.
Or as I tell the fuckwits working in Infosec at my employer, if they want it secure I can hit the shiny red button in the data centre and shut the lot down.
No? Ok, you don't want secure, you want a compromise. Now we can discuss it.
So next time you hear Linus Torvalds rant against C++ tell him to fuck off and stop being an idiot.
Why would I want to? He's clearly not an idiot, I owe him for the work he's done to build a reliable kernel I use daily, and C++ is indeed a fucking stupid language.
I've yet to see any C++ code base that is bug free, mainly because inexperienced people write bugs and experienced people write so much fucking scaffolding code to try and prevent them that the code becomes unmaintainable anyway.
Good experienced people switch to better languages.
It means that when you eliminate the fancy words, the marketing, the ego, the bullshit and everything else.. programmers automate stuff.
Sure.
Lets ditch double-entry book keeping and make accountants add up properly.
Lets ditch medicines and make doctors heal people with their own skills.
Lets ditch case law and force lawyers to properly argue why their clients are innocent.
Or maybe, just maybe, professional software engineers take full ownership, full responsibility and exploit the tools and systems available to them to maximise the chances of good outcomes without adding unsustainable cost or time burdens on their work.
Maybe.
I'm not a biologist or a natural historian. Let me consult an appropriate reference or expert, and while I'm doing that, perhaps you could tell us all just what the fuck that has to do with being president of the US?
You appear to be conflating Linux with the shit that gets bundled with it.
Forget unit tests, testing your inputs has been a basic fundamental since the very first accidental bad data bug.
It would've been sooner but computing resources were too expensive so the inputs were validated offline first.
Voodoo programming can produce working software. Why not working hardware too?
The Linux kernel has had 2 1/2 decades of continuous oversight, guidance, QA, design and discussion.
That it works at all is testament to the doggedness of the team that develops and maintains it. That it's the most used operating system on the planet is testament to their engineering excellence.
Systemd hasn't yet proven either of these.
Not if you restrict the answer set to "things a traditional Unix init system should do".
Probably amazed he didn't know about Stack Overflow.
Just create a switch that is external and add that to yoru Vms and you should be good.
I should, but I'm not. Welcome to Hyper-V.
It's also pretty silly. There's masses of great music out there, more and more accessible than ever before.
He maybe just hasn't found it yet.
So, wondering when the masses stopped caring at all about how the music sounded?
When I was diagnosed with partial deafness at age 12.
When I was issued hearing aids in my 20s.
When I gave up on them and decided to live life in the quiet in my 30s.
Are Klipschorns real or was that an elaborate piss-take on audiophiles? Sorry, reality blurs with satire on this topic.
https://blogs.msdn.microsoft.c... is a useful resource, that includes "Unfortunately, this approach does not always work."
No, no it doesn't.
I lost patience with the NAT approach. I'm not a Windows admin, a network specialist or a virtualisation expert so I decided to defer the day or two of learning and experimentation for when I have energy and time.
Or Microsoft could fix the shitty hypervisor. Seriously, when it's easier to download software from Oracle you know there's something broken.
Does this mean they'll finally fix network access for hyper-v hosted VMs when the host system is connected via wifi?
Just that right now it's a fucking shitfest.
Or maybe they're creating a whole new hypervisor for Edge, that will actually work.
I've heard of Jello Biafra and Tipper Gore. I didn't know they were related.
Why would I be culturally unaware for not giving a flying fuck about some American punk band? We had The Clash, The Sex Pistols and The Wurzels. Wait, no, skip that last one.
You'll be telling me off for not knowing what sort of music The Stringcheese Incident play next.