It sounds to me like the only reasonable definition of compromise between someone who wants a tax increase of x and someone who wants nothing is a smaller tax increase.
Re:Data, Images, Binary builds etc.
on
The Rise of Git
·
· Score: 1
Your images used in the app are part of the source. While there's a place for storing data elsewhere, it still has to be controlled in a way that you can get it back out again for a particular version.
Indeed it does. But that storage has a very different set of requirements from a VCS - it makes very little sense to do a diff between two versions of an image file, and even less to talk about branching and merging for them. Use another piece of software for that (heck, use subversion - its limitations don't really apply for this use case). Git does one thing and does it well, which is as it should be.
Moreover, you don't need to type in the entire SHA1; any abbreviation will do as long as it is unique among the other SHA1s known in the repository.
This is really useless though. How can I know a given start will be unique? So how do I know how long is long enough to be safe to email to bob (who has his own set of local commits, whose SHA1s I don't know)? So I end up just using the whole thing every time.
Sure, but you can't write and release an interpreter for a turing-complete language for other people to use. And an awful number of useful things are turing-complete (Heck, if I've read the app store guidelines correctly, a postscript renderer would be in violation of apple policies and they would have every right to remove it).
So your definition of smartphone isn't the commonly accepted one where users are allowed to do those things. Instead your defintion is one where users have to things they no longer have to do because of something called progress.
Look, if you can't run arbitrary programs it's not a smartphone. That's the only meaningful feature distinction we can make. Otherwise my old nokia was a smartphone - I could use it to make calls and text and browse the web and play nokia's games and...
The KH-11 spy satellites that have very similar dimensions and exactly the same optics as Hubble were flown into space using a Titan IIIE missle - which could have brought the telescope into a much higher and reasonable orbit.
It bears remembering that if this had happened we could never have installed the correction lenses for the misshapen mirror, and Hubble would've been pretty much completely useless.
It is impossible for a photon to decay into an electron-positron pair while traversing vacuum. Only when the photon is scattered by a massive, charged particle, an electron-positron pair can be produced (provided it has enough energy).
Can't two photons colliding in a vacuum "decay" like that? It should be possible to time-reverse what happens when an electron and positron collide in a vacuum.
If you gave out proportionately more BTC as the number of users grow, the inflation would be insane.
Would that actually be a problem? If you buy bitcoins as/when you need them, inflation wouldn't bother you when using them. It would only prevent people from hoarding them - which, guess what, is a good thing.
So it's 500 BTC per hour, divided among everyone
Would that it were. But no, it's far worse than that, the amount given out per hour decays exponentially. Which is insane.
Why is it shocking to have to earn your currency?
It's not. It's shocking that early adapters don't have to. Why should I consider bitcoins worth my work, when other people got them for free?
Mandela was far from civil in his disobedience - the ANC was blowing things up. I love how we lionize a terrorist. Your sibling post says Rosa Parks was fined the equivalent of $120.
The people insisting so loudly that 'Dalvik is not Java' are those that feel that creating an incompatible Java is 'wrong'... but they still want it in their phone.
Having multiple almost-compatible versions of java that users could be confused between would be bad. If Dalvik let you take random jar files and try and run them, and some would work and some wouldn't, that would be bad. To my mind Dalvik is distinct enough that this problem won't arise, and there's absolutely nothing wrong with using an existing language as a foundation for yours, as long as there's no potential for confusion. (I only wish they'd gone with C#)
Taking that weakness and calling it a strength just doesn't make sense.
But it is - the only thing it removes is a false flexibility that will always bite you in the ass sooner or later if you try and use it. The MS way forces you to reboot your system when it's updated, which is the only reliable way to ensure the update actually gets applied (if you try and do this yourself, by restarting only the things that are needed, you will get it wrong sooner or later). And it forces you to architect your services such that any given machine can be rebooted at any time without impacting service, which again is something that you need to be doing already or it will bite you.
They ought to follow the apache versioning standards (major version for completely incompatible changes, minor for backwards- but not forwards-compatible changes, patch for bidirectional-compatible changes). Of course, given the kernel's attitude to compatibility that would probably lead to every release being a major version.
That java itself was GPL'd prior to the oracle acquisition and teh terms of that license provides blanket patent coverage over java itself
No they don't - GPLv2 mentions nothing of patents. (I think possibly the CDDL does, making it in some ways better - was java ever released under the CDDL?). This has been an issue for Apache Harmony, as sun's patent grants only apply to software that passes those tests that they're not allowed to have.
Imagine if a company sued wordpress for patent infringement and then claimed that every ad shown on their commercial wordpress.com free blog hosting site is revenue that should count toward the calculation of patent damages ?
I think that's perfectly sensible (or rather, as sensible as patents ever are). You shouldn't be able to get free reign to infringe on anyone's patents just by making the particular infringing part of your software open source.
The same goes for fake RAID (ie, software RAID driven by the BIOS), but s/controller card/motherboard/g;
Not really; linux's dmraid can read all the common fakeraid formats from any controller. You may not be able to boot from your disks in another machine, but you'll be able to read them.
Mdadm has lost me large amounts of data in the past; after a while you will have isolated bad sectors on all your drives, but linux believes the correct way to deal with a single URE is to boot that drive out of the array. Of course this doesn't happen until you try to read it. So to my mind the only option is ZFS's raid-z, which handles this case correctly, and has a way to verify your data before it breaks.
Sure. But even taking that into account I suspect that for equal-entropy passwords, an English sentence is going to be easier to remember. (In fact if I remember correctly English averages about 3 bits per character, so it's only going to be about twice the length of a symbolful password.
This is why I prefer to use English sentences as passphrases - if you're a decent typist you can type those perfectly accurately, however long, and the extra length more than makes up for using a smaller range of characters. (And I don't even get to use muscle memory, since I'm frequently typing them in on an unusual (to me) keyboard layout).
I've implemented a related, but certainly not identical, system in my home with two wireless APs running two independent networks feeding a single cable connection.
Huh? Surely that's backwards from the useful way to do it - all your client machines would need to be able to connect to two wireless networks simultaneously and sensibly failover between them.
Is the Hurd's design still relevant? Yes, and it's motivated many improvements that have made their way into linux and other systems; as a testing ground for microkernel technology it will probably continue to do so (it's probably the biggest open-source microkernel out there).
But as for actually running it day-to-day, the Hurd never was relevant simply because it never had the broad driver support that users need. That won't change unless and until Hurd attracts a substantial developer base - but this is a good step in the direction of that.
Slashdot Mirror
Fusion is a specific nuclear reaction... they have achieved that already...
Sure, but we'd achieved that back in 1970. When we talk about fusion being 20 years away, we're talking about useful energy from it.
It sounds to me like the only reasonable definition of compromise between someone who wants a tax increase of x and someone who wants nothing is a smaller tax increase.
Your images used in the app are part of the source. While there's a place for storing data elsewhere, it still has to be controlled in a way that you can get it back out again for a particular version.
Indeed it does. But that storage has a very different set of requirements from a VCS - it makes very little sense to do a diff between two versions of an image file, and even less to talk about branching and merging for them. Use another piece of software for that (heck, use subversion - its limitations don't really apply for this use case). Git does one thing and does it well, which is as it should be.
Moreover, you don't need to type in the entire SHA1; any abbreviation will do as long as it is unique among the other SHA1s known in the repository.
This is really useless though. How can I know a given start will be unique? So how do I know how long is long enough to be safe to email to bob (who has his own set of local commits, whose SHA1s I don't know)? So I end up just using the whole thing every time.
Except that it can you are a developer.
Sure, but you can't write and release an interpreter for a turing-complete language for other people to use. And an awful number of useful things are turing-complete (Heck, if I've read the app store guidelines correctly, a postscript renderer would be in violation of apple policies and they would have every right to remove it).
So your definition of smartphone isn't the commonly accepted one where users are allowed to do those things. Instead your defintion is one where users have to things they no longer have to do because of something called progress.
Look, if you can't run arbitrary programs it's not a smartphone. That's the only meaningful feature distinction we can make. Otherwise my old nokia was a smartphone - I could use it to make calls and text and browse the web and play nokia's games and ...
The KH-11 spy satellites that have very similar dimensions and exactly the same optics as Hubble were flown into space using a Titan IIIE missle - which could have brought the telescope into a much higher and reasonable orbit.
It bears remembering that if this had happened we could never have installed the correction lenses for the misshapen mirror, and Hubble would've been pretty much completely useless.
Democrats made a stupid mistake prosecuting the court case, and it cost them. Shows you the problem with an adverserial justice system.
It is impossible for a photon to decay into an electron-positron pair while traversing vacuum. Only when the photon is scattered by a massive, charged particle, an electron-positron pair can be produced (provided it has enough energy).
Can't two photons colliding in a vacuum "decay" like that? It should be possible to time-reverse what happens when an electron and positron collide in a vacuum.
If you gave out proportionately more BTC as the number of users grow, the inflation would be insane.
Would that actually be a problem? If you buy bitcoins as/when you need them, inflation wouldn't bother you when using them. It would only prevent people from hoarding them - which, guess what, is a good thing.
So it's 500 BTC per hour, divided among everyone
Would that it were. But no, it's far worse than that, the amount given out per hour decays exponentially. Which is insane.
Why is it shocking to have to earn your currency?
It's not. It's shocking that early adapters don't have to. Why should I consider bitcoins worth my work, when other people got them for free?
The amount that early adopters will get out of this utterly pales in comparison to what the big financial corporations are raping you for.
Bollocks. The big financial corporations haven't taken 80% of all the wealth in existence.
It will hurt its supporters, sure. But it will benefit the party, politically. Just like every rightist party ever.
Sounds more like "we will have fusion in the next year save for mechanical malfunctions" to me.
Mandela was far from civil in his disobedience - the ANC was blowing things up. I love how we lionize a terrorist. Your sibling post says Rosa Parks was fined the equivalent of $120.
The people insisting so loudly that 'Dalvik is not Java' are those that feel that creating an incompatible Java is 'wrong'... but they still want it in their phone.
Having multiple almost-compatible versions of java that users could be confused between would be bad. If Dalvik let you take random jar files and try and run them, and some would work and some wouldn't, that would be bad. To my mind Dalvik is distinct enough that this problem won't arise, and there's absolutely nothing wrong with using an existing language as a foundation for yours, as long as there's no potential for confusion. (I only wish they'd gone with C#)
Taking that weakness and calling it a strength just doesn't make sense.
But it is - the only thing it removes is a false flexibility that will always bite you in the ass sooner or later if you try and use it. The MS way forces you to reboot your system when it's updated, which is the only reliable way to ensure the update actually gets applied (if you try and do this yourself, by restarting only the things that are needed, you will get it wrong sooner or later). And it forces you to architect your services such that any given machine can be rebooted at any time without impacting service, which again is something that you need to be doing already or it will bite you.
They ought to follow the apache versioning standards (major version for completely incompatible changes, minor for backwards- but not forwards-compatible changes, patch for bidirectional-compatible changes). Of course, given the kernel's attitude to compatibility that would probably lead to every release being a major version.
That java itself was GPL'd prior to the oracle acquisition and teh terms of that license provides blanket patent coverage over java itself
No they don't - GPLv2 mentions nothing of patents. (I think possibly the CDDL does, making it in some ways better - was java ever released under the CDDL?). This has been an issue for Apache Harmony, as sun's patent grants only apply to software that passes those tests that they're not allowed to have.
Imagine if a company sued wordpress for patent infringement and then claimed that every ad shown on their commercial wordpress.com free blog hosting site is revenue that should count toward the calculation of patent damages ?
I think that's perfectly sensible (or rather, as sensible as patents ever are). You shouldn't be able to get free reign to infringe on anyone's patents just by making the particular infringing part of your software open source.
At this stage pretty much any TLA is being used by someone as a file format name.
The same goes for fake RAID (ie, software RAID driven by the BIOS), but s/controller card/motherboard/g;
Not really; linux's dmraid can read all the common fakeraid formats from any controller. You may not be able to boot from your disks in another machine, but you'll be able to read them.
Mdadm has lost me large amounts of data in the past; after a while you will have isolated bad sectors on all your drives, but linux believes the correct way to deal with a single URE is to boot that drive out of the array. Of course this doesn't happen until you try to read it. So to my mind the only option is ZFS's raid-z, which handles this case correctly, and has a way to verify your data before it breaks.
Sure. But even taking that into account I suspect that for equal-entropy passwords, an English sentence is going to be easier to remember. (In fact if I remember correctly English averages about 3 bits per character, so it's only going to be about twice the length of a symbolful password.
if you look up "Tiny7 Rev 09" you'll find a version of Windows 7 that uses less memory than XP
Compare like with like. I'll bet it uses more memory than comparable minimal repacks of windows XP, of which there are plenty.
This is why I prefer to use English sentences as passphrases - if you're a decent typist you can type those perfectly accurately, however long, and the extra length more than makes up for using a smaller range of characters. (And I don't even get to use muscle memory, since I'm frequently typing them in on an unusual (to me) keyboard layout).
Facile? Did you really mean that?
I've implemented a related, but certainly not identical, system in my home with two wireless APs running two independent networks feeding a single cable connection.
Huh? Surely that's backwards from the useful way to do it - all your client machines would need to be able to connect to two wireless networks simultaneously and sensibly failover between them.
But as for actually running it day-to-day, the Hurd never was relevant simply because it never had the broad driver support that users need. That won't change unless and until Hurd attracts a substantial developer base - but this is a good step in the direction of that.