Sometimes there is in fact a need for public shaming, I think. Kay asked for it, Kay got it delivered. It's as simple as that. Doing it in private would be a disservice to everyone. It's a developer community, there's no point for keeping this sort of thing private. It goes against the very grain of things, I think.
Of course these days all of this can be done, too, much faster, on off-the-shelf hardware. Just because the hardware doesn't have tag bits doesn't mean your compilers can't implement them. I'm running a bit of safety critical code on a bunch of ARM CPUs and all of the data RAM contents are tagged, pointers are tagged, and there is also software-driven error correction for RAM, execution log, restarts, those sorts of things that were en vogue at one point or another in the "hi-rel mainframe" market.
I have a couple of off-the-shelf servers from Dell that not only have error correcting RAM, but also have a spare memory stick and can cope with the failure of an entire chip on a RAM stick. So what you hail so eagerly is - who'd have thought - a standard feature on off-the-shelf hardware that can be had under $2K.
You're on to something here, but not for the reasons that you think. NASA has been releasing source for a long time. It's only that getting this source requires at least a mountainload of paperwork (U.S. citizens only, etc.), and it's usually costly. It's not like they don't have a catalog already. If it's going to be more of the same, then I'd call it outright deception. Note that nowhere it's stated that the code will be under a free source license!
How about you drive 65 miles an hour around a curve in traffic and have someone else tuun off the ignition? whats that? you won't do it?
Had that happen, had a service brake failure (pinhole in the flex line), and had an ABS failure in very slippery conditions. I'm here to tell the tale, and nobody was hurt, and no paint was scraped. Pretraining is key, and anyone who has a car and hasn't figured out for themselves how to handle such situations is playing with human life for no good reason. You're supposed to actually try things out. It's easy, if you're not an oaf it's safe, so what's the deal?
in automatics because in drive the wheels can't turn the engine, so the engine comes to a complete standstill
Stop making shit up. It doesn't work that way at highway speeds, and at least in my car, the automatic transmission stays engaged if I turn off the engine, all the way down to 30mph or so.
Yeah, sure, ha ha. Airbags often deploy so fast that by the time they are deployed, you are usually nowhere near the airbag. Even in fairly high-speed crashes you often impact an already deflating airbag. Never mind that modern cars have multi-stage airbags that inflate sufficiently for the severity of the crash.
Oh, and never mind that modern cars have many airbags that are nowhere near your front. I've personally checked out a side curtain, and it was quite nice. The T-bone felt like being pushed onto a bed by a rowdy kid jumping on you:) I didn't even get a headache.
If part number changes, how will the customer know there is now a new part?
Every online part retailer worth their salt has some indirect access to car vendor's part database that, surprise surprise, includes part substitutions. Had you ordered any car parts online, especially ones that fail due to bad design and get redesigns/upgrades, you'd have seen it. This ignition switch part is rather unusual in this respect - a redesign was done without retiring the old P/N and having a "new and improved" P/N as a designated replacement.
On some cars, I've seen parts go through 2 or 3 rounds of such upgrade cycles. When I eventually pulled the failed part off the car and went to buy a replacement, there was a substitute part #1. That substitute was substituted with #2. Eventually, that one had substitute #3, which was orderable.
probably do not have insight into any safety critical functions that it plays when integrated with the larger vehicle
That's what PPAP and APQP are dealing with. In practice, those processes are used to ensure that the vendors understand what is expected of their parts in their intended end-use.
Just a company that does not want to pay money to do what they believe is killing babies.
They should be rightly shunned for acting on fantasies, as it were. What they believe here is a figment of a collective imagination of a whole bunch of people, unfortunate enough to be misled to believe such nonsense. Just sayin'.
In the particular case here, I don't quite get how an engine stall can cause loss of control? Going 50+mph, you don't care about power steering. Loss of brake assist might be an issue, and loss of ABS too if that was the case, but even then, I think in this one the driver shares 50% of the blame.
I've had hydraulic brakes fail, I've had ABS fail (with no MIL coming on), and in both cases some forethought and pre-training has immensely helped me. Yes, in both cases those were close calls, but through no fault of mine. No, nobody had me do it. I simply figured: I must know what to do in such cases, since there's never enough time to think. For example, if my accelerator-by-wire car ever had an UA on a road with some clearance ahead of me, I know that it wouldn't cause anyone any harm - I know how to handle it, and I routinely test my service brakes to ensure that they will stop the car in such a scenario even if I were to forget to switch to neutral (or the switching didn't have effect, as it may). The ECU doesn't have a brake throttle override, BTW.
Yes. I've found that there's no way to use it without knowing this. It's as simple as that, and I don't consider it a deficiency of git at all. You're supposed to understand the tools that you use. The things you speak of are not implementation details, they are part of the semantics exposed directly to the user. They make the whole thing work and useful.
Yes, the command line syntax is abhorrent, but then I almost never use it. SmartGit/Hg has won me over.
The initial checkout from svn doesn't clone the entire history. With git, a shallow clone (--depth 1) doesn't either. Git's smart http transport is bound to be faster than svn, since compression can be applied across multiple files, and if there are similar files in the repository, their common chunks will always be factored out by design. With svn, it's not guaranteed and depends on how file copies/moves were managed.
First of all, I don't think that it's sane to use svn or git without a graphical front-end. That said, I find that git-svn works way better than svn itself. Subversion merging starts looking downright silly once you see how git does it. And I was one of those who actually argued that svn's merging is "good enough". Eventually I convinced myself, through experimentation, that it's nowhere near good enough. To me, subversion is a way of centrally storing git repos without having to configure server-side git, and also a way of doing partial checkouts when I don't need the whole thing.
Yeah, that's what I always tell people: the scale of things in the U.S. is quite hard to understand without actually being here. U.S. municipal budgets are like country budgets in eastern Europe. NYC tracks Poland, etc.
That's why I'm always smiling when people smirk at me when I tell them that a lot of firmware that I develop runs on devices with 64kb or less total memory (code + data, since it's Harvard). I'm not using frameworks with tens of thousands of lines of code, I can actually tell exactly how the whole thing works, top-to-bottom, and it can be explained in detail in about 150 pages of prose. A lot of that code is written in functional style (side-effect-free functions) and can be formally proven to agree with the specs. When you've got a lot more than that, it gets very, very costly to have formal proofs of anything, much less even formal specifications (formal not as in bureaucratic, formal as in using the language of mathematics).
Apple has integrated Total Information Awareness about the files and processes on our computers in that OS
Does that mean anything? An operating system by definition is totally aware of the files and processes that you run on it. So, the best I can make out of what you're saying, is just gobbledygook and wakalixes. Nobody forces you to use cloud backup of anything with Mavericks. I have no evidence that the app update mechanism is leaking contents of private files, but if you have some evidence to the contrary, I'm sure you could get some limelight time for it.
No, you don't have any rights to others' property just because it happens to be a camera pointing at you. You'll run into a lot of trouble with the law with your fantasies about what actually is in the law.
What's the difference between me looking at you doing something in public, and me recording the same? To me, there's none. Either you have no problem with what you speak and do, and others hear and see you do, or you do have a problem - in this latter case you should maybe, just maybe, refrain from saying or doing it. Easy.
Man, you must be a lot of joy to have around if you beat up everyone who dares to take out their cellphone with you around. Protip: it's way too late to be upset about cameras. Grow a pair. Cameras are everywhere.
Slashdot Mirror
Sometimes there is in fact a need for public shaming, I think. Kay asked for it, Kay got it delivered. It's as simple as that. Doing it in private would be a disservice to everyone. It's a developer community, there's no point for keeping this sort of thing private. It goes against the very grain of things, I think.
Of course these days all of this can be done, too, much faster, on off-the-shelf hardware. Just because the hardware doesn't have tag bits doesn't mean your compilers can't implement them. I'm running a bit of safety critical code on a bunch of ARM CPUs and all of the data RAM contents are tagged, pointers are tagged, and there is also software-driven error correction for RAM, execution log, restarts, those sorts of things that were en vogue at one point or another in the "hi-rel mainframe" market.
I have a couple of off-the-shelf servers from Dell that not only have error correcting RAM, but also have a spare memory stick and can cope with the failure of an entire chip on a RAM stick. So what you hail so eagerly is - who'd have thought - a standard feature on off-the-shelf hardware that can be had under $2K.
You're on to something here, but not for the reasons that you think. NASA has been releasing source for a long time. It's only that getting this source requires at least a mountainload of paperwork (U.S. citizens only, etc.), and it's usually costly. It's not like they don't have a catalog already. If it's going to be more of the same, then I'd call it outright deception. Note that nowhere it's stated that the code will be under a free source license!
How about you drive 65 miles an hour around a curve in traffic and have someone else tuun off the ignition? whats that? you won't do it?
Had that happen, had a service brake failure (pinhole in the flex line), and had an ABS failure in very slippery conditions. I'm here to tell the tale, and nobody was hurt, and no paint was scraped. Pretraining is key, and anyone who has a car and hasn't figured out for themselves how to handle such situations is playing with human life for no good reason. You're supposed to actually try things out. It's easy, if you're not an oaf it's safe, so what's the deal?
in automatics because in drive the wheels can't turn the engine, so the engine comes to a complete standstill
Stop making shit up. It doesn't work that way at highway speeds, and at least in my car, the automatic transmission stays engaged if I turn off the engine, all the way down to 30mph or so.
Yeah, sure, ha ha. Airbags often deploy so fast that by the time they are deployed, you are usually nowhere near the airbag. Even in fairly high-speed crashes you often impact an already deflating airbag. Never mind that modern cars have multi-stage airbags that inflate sufficiently for the severity of the crash.
Oh, and never mind that modern cars have many airbags that are nowhere near your front. I've personally checked out a side curtain, and it was quite nice. The T-bone felt like being pushed onto a bed by a rowdy kid jumping on you :) I didn't even get a headache.
If part number changes, how will the customer know there is now a new part?
Every online part retailer worth their salt has some indirect access to car vendor's part database that, surprise surprise, includes part substitutions. Had you ordered any car parts online, especially ones that fail due to bad design and get redesigns/upgrades, you'd have seen it. This ignition switch part is rather unusual in this respect - a redesign was done without retiring the old P/N and having a "new and improved" P/N as a designated replacement.
On some cars, I've seen parts go through 2 or 3 rounds of such upgrade cycles. When I eventually pulled the failed part off the car and went to buy a replacement, there was a substitute part #1. That substitute was substituted with #2. Eventually, that one had substitute #3, which was orderable.
probably do not have insight into any safety critical functions that it plays when integrated with the larger vehicle
That's what PPAP and APQP are dealing with. In practice, those processes are used to ensure that the vendors understand what is expected of their parts in their intended end-use.
Just a company that does not want to pay money to do what they believe is killing babies.
They should be rightly shunned for acting on fantasies, as it were. What they believe here is a figment of a collective imagination of a whole bunch of people, unfortunate enough to be misled to believe such nonsense. Just sayin'.
In the particular case here, I don't quite get how an engine stall can cause loss of control? Going 50+mph, you don't care about power steering. Loss of brake assist might be an issue, and loss of ABS too if that was the case, but even then, I think in this one the driver shares 50% of the blame.
I've had hydraulic brakes fail, I've had ABS fail (with no MIL coming on), and in both cases some forethought and pre-training has immensely helped me. Yes, in both cases those were close calls, but through no fault of mine. No, nobody had me do it. I simply figured: I must know what to do in such cases, since there's never enough time to think. For example, if my accelerator-by-wire car ever had an UA on a road with some clearance ahead of me, I know that it wouldn't cause anyone any harm - I know how to handle it, and I routinely test my service brakes to ensure that they will stop the car in such a scenario even if I were to forget to switch to neutral (or the switching didn't have effect, as it may). The ECU doesn't have a brake throttle override, BTW.
OK, but should I need to know any of that?
Yes. I've found that there's no way to use it without knowing this. It's as simple as that, and I don't consider it a deficiency of git at all. You're supposed to understand the tools that you use. The things you speak of are not implementation details, they are part of the semantics exposed directly to the user. They make the whole thing work and useful.
Yes, the command line syntax is abhorrent, but then I almost never use it. SmartGit/Hg has won me over.
The initial checkout from svn doesn't clone the entire history. With git, a shallow clone (--depth 1) doesn't either. Git's smart http transport is bound to be faster than svn, since compression can be applied across multiple files, and if there are similar files in the repository, their common chunks will always be factored out by design. With svn, it's not guaranteed and depends on how file copies/moves were managed.
The shallow clone have fixed this, mostly. So far, I find no downsides to shallow clones, other than the loss of history.
First of all, I don't think that it's sane to use svn or git without a graphical front-end. That said, I find that git-svn works way better than svn itself. Subversion merging starts looking downright silly once you see how git does it. And I was one of those who actually argued that svn's merging is "good enough". Eventually I convinced myself, through experimentation, that it's nowhere near good enough. To me, subversion is a way of centrally storing git repos without having to configure server-side git, and also a way of doing partial checkouts when I don't need the whole thing.
Is "tough" a technical term?
Fuck yes!
Man, that MS Word source should be perfect for an allnighter port to modern winapi :)
Yeah, that's what I always tell people: the scale of things in the U.S. is quite hard to understand without actually being here. U.S. municipal budgets are like country budgets in eastern Europe. NYC tracks Poland, etc.
That's why I'm always smiling when people smirk at me when I tell them that a lot of firmware that I develop runs on devices with 64kb or less total memory (code + data, since it's Harvard). I'm not using frameworks with tens of thousands of lines of code, I can actually tell exactly how the whole thing works, top-to-bottom, and it can be explained in detail in about 150 pages of prose. A lot of that code is written in functional style (side-effect-free functions) and can be formally proven to agree with the specs. When you've got a lot more than that, it gets very, very costly to have formal proofs of anything, much less even formal specifications (formal not as in bureaucratic, formal as in using the language of mathematics).
Apple has integrated Total Information Awareness about the files and processes on our computers in that OS
Does that mean anything? An operating system by definition is totally aware of the files and processes that you run on it. So, the best I can make out of what you're saying, is just gobbledygook and wakalixes. Nobody forces you to use cloud backup of anything with Mavericks. I have no evidence that the app update mechanism is leaking contents of private files, but if you have some evidence to the contrary, I'm sure you could get some limelight time for it.
I think that you'll find that Windows 8.1 will perform like shit on those machines... Not only because all of the rendering will be software-based.
Care to elaborate? (seriously)
Again, what's wrong with it? You don't want to be seen, don't go out in public, it's that easy.
No, you don't have any rights to others' property just because it happens to be a camera pointing at you. You'll run into a lot of trouble with the law with your fantasies about what actually is in the law.
What's the difference between me looking at you doing something in public, and me recording the same? To me, there's none. Either you have no problem with what you speak and do, and others hear and see you do, or you do have a problem - in this latter case you should maybe, just maybe, refrain from saying or doing it. Easy.
Man, you must be a lot of joy to have around if you beat up everyone who dares to take out their cellphone with you around. Protip: it's way too late to be upset about cameras. Grow a pair. Cameras are everywhere.