> can there be any good argument against greater energy efficiency?
Quite possibly, yes, locally.
For example, my water heater (not that old, not that new) is about 85% efficient. It's expected to have a useful life of another 10 years or so. I could get that up to 90% or so by getting a brand-new high-efficiency one. My net energy savings would be on the order of $50 a year at most, which works out to $500 over those ten years. A new water heater costs more than $500.
So the right thing to do in this case is to keep using the old heater for those 10 years. Note that this also avoids the energy use of _producing_ an extra water heater, so it's less of a loss to others than it would appear at first glance. It's clearly a win for me personally.
Now obviously this calculation depends on the price of energy, the price of water heaters, and the efficiency gains to be had. The last of these starts hitting diminishing returns quickly once what you have is anything resembling non-crappy; for example it's hard to reduce your energy usage by 2x if you're already 60% efficient.... Much easier to do if you're 10% efficient, of course.
. GP _asked_ what the plural is. So I told him, along with a good way to look up that information in the future (on the "teach a man to fish" premise).
The correct plural is "nemeses" (it's a Greek word, not Germanic or Latin as suggested respectively by your two proposed plural forms). Similar to how one pluralizes "axis", "synthesis", "analysis", "genesis", etc, and for the same reasons.
I should note that any sane dictionary will tell you what the plural form of a noun is. Or heck, googling "plural nemesis" in a pinch (first two hits are dictionary entries for "nemesis" that include the plural form). Just for future reference.;)
The US government "outsources" _all_ its military procurement, in the sense that they don't operate their own manufacturing facilities, don't necessarily do all the military R&D, etc.
So if, say, you're working for Northrop Grumman (the only manufacturer of US nuclear aircraft carriers and one of the two manufacturers of US nuclear submarines) and you're on a classified project and you leak the details of that project to some other country... then that's espionage, period.
It's receiving updates to the apps. The last update to the core OS itself was in September 2009, right when 10.6 shipped. They never announce end of support, so we won't know whether they've dropped it until they just haven't shipped any security fixes for a while.... and since they don't ship those on a regular schedule, no one knows what the right "while" is.
Pretty hard, it turns out, unless you don't care how slow it is.
> I'm surprised you don't already have a generic text-rendering wrapper.
Not sure what you mean by that. There's a generic text-rendering API that is implemented via either ATSUI or CoreText right now, selected at runtime; that introduces a lot of complication and slows things down on Mac.
> You have the OS version right there in the user-agent string.
. I'm just passing on what I was told by the people working on the download stuff.
> Yes there is. Ship two binaries.
Is this the best course of action, though? Is that better than spending the same resources on making the browser work better on 10.5 and 10.6 instead?
Apple's normal security update cycle seems to be that they stop doing security updates to an OS about 2.5 years after last shipping it (which happens to correspond to when they start shipping the next+1 OS; in this case 10.6). They dropped 10.3 support pretty much when 10.5 came out, and that was also about 2.5 years after 10.4 appeared.
Well... The question is not about current releases; the current release of Firefox supports 10.4. The question is about _next_ releases. The next release of Safari... who knows. They don't publicly talk about their release planning 18 months in advance, unlike Mozilla.
Note that Safari right this second ships different binaries for 10.4, 10.5, and 10.6. Of course this is made easier by the fact that most users get it via software update, not by having to download the right thing.
As for Camino, it's going to end up dropping 10.4 support once it switches to Gecko 1.9.3 or equivalent, just like Firefox.... It's Gecko that will probably drop 10.4 support, not the browser UI.
> Or you could have a legacy 10.4 build separate from the 10.5/6 build
Possible, at a fair amount of cost in maintenance and code complexity and a good bit of cost in user confusion when downloading builds.
> but it's probably still worth it IF there are sufficient users
That's the big question. Right now about 25% of the Firefox 3.5/3.6 Mac users are on 10.4. Where will it be a year from now? Who knows. In general, about 1% of current Firefox 3.5/3.6 users across all platforms are on Mac OS X 10.4.
> It's certainly more recent than the start of Windows XP, which is still well supported
XP is also a heck of a lot more like Win7 or Vista than 10.4 is like 10.6. That is, it's a lot easier to support XP. And yes, the large number of users probably doesn't hurt.
> Some older iMac users already have been shafted by Apple
They're shafted by Apple period. Apple has typicall been dropping support of its operating system versions within about 4 years of initially shipping them, and within about 2 years of last shipping them. As in, you buy a computer and 2 years later security updates for your OS stop. That's about the point at which Mozilla happens to stop security updates for your browser on that OS too, on the premise that you're insecure no matter what.
> The last 'general' security update for 10.4 was on 10 Sept 2009
Which was right about when 10.6 shipped. At that point 10.4 was in fact the "previous OS".
The question is whether there will be more 10.4 security updates. And since Apple never announces EOL for its OSes officially, there's no way to know for sure.
The question is what Firefox whatever-the-next-version-is will be offered on, and the right comparison there is to Safari 5. Which will be offered.... somewhere. Who knows where.
Microsoft takes backwards compatibility a lot more seriously than Apple does. Supporting both 10.4 and 10.6 well at the same time is somewhat comparable to supporting both Win98 and Vista at the same time (have to use different text rendering APIs, different graphics APIs, etc, etc).
Unknown. The last Tiger security update was in Sept 2009, as I recall. Apple never announces official end-of-life dates for its OSes as far as I can see; it just silently stops shipping security updates at some point. Since those don't ship on a predictable schedule to start with, one never knows whether you've seen the last one or not.
> Maybe Mozilla's done the user research and they know that they're not dropping many users,
The second link in the summary has the data on that. In brief, as of end of January 1010, 25% of Firefox 3.5 Mac users (about 1.4 million users, or about 0.5% of total Firefox users) are using OS 10.4. 12% of Firefox 3.6 Mac users (about 36,000 users) are using OS 10.4.
The big question mark, of course, is what those numbers will look like about 15 months from now, which is the earliest that Firefox 3.6 might be going out of support...
> Mozilla should have a very clear policy about backwards compatibility and follow it to the > letter.
The basic setup is:
1) Once an OS vendor drops support for an OS, support for it will not be maintained unless
it's really easy to do. 2) Whether an OS is supported depends on whether there are resources to support it and on
how many users are using it.
It's not exactly a clear policy, but the important part is that support decisions are pretty complicated and involve a lot of factors.... it's not clear to me what a sane policy would be that would not lead to dropping support in some cases when there's no real reason to do it.
> Until Apple actively does something to break the older "deprecated" code in Firefox, > they should support older OSes
10.6 dropped ATSUI support. 10.4 doesn't have Core Text. So the only way to support both is to have codepaths to use both text rendering backends and switch at runtime. Does that count as "does something to break"?;)
Thing is, it's all software. Everything can be worked around. The question is the cost (to users, in the end, either in terms of money or in terms of things users want that don't happen).
> and you deal with the slightly older APIs/compilers to serve your users
Not that simple. You have to use gcc 4.0 if you're going to run on 10.4. So doing that serves the 10.4 users. But on 10.5 and 10.6, using gcc 4.2 gives a pretty significant across-the-board speedup. So to properly serve those users, you want to be using gcc 4.2. Where that leaves you is either underserving 10.5/10.6 users to better serve 10.4 users or vice versa (at which point relative numbers of users start to matter), or shipping separate binaries with the ensuing user confusion during downloading, etc. So there's not an obvious course of action here that best serves "the users". It's a matter of compromise.
> But most importantly, because that's the law in a huge part of the world, the US included
Ok, so you're saying the existing law already has a double standard and your quibble is with where the boundary the law draws in applying this double standard lies? I can buy that.
That said, I see no inherent issue with being able to patent a clever way of combining mathematical operations to, say, extract a signal from a noisy data stream, if we're ok with being able to patent combining spiky bits and a roller to separate cotton seeds from cotton fibers.
See http://www.paulgraham.com/softwarepatents.html for some more thoughts on this (I just ran into it when I searched on "patent algorithm", and it eerily echoes exactly what I was saying earlier).
Now there is an important point here: If you're going to have patents, they must be granted for specific inventions, not overbroad descriptions. Patenting "a roller with teeth" would never fly; patenting a roller with teeth being used in a particular way to get a particular effect is a different issue entirely. Similarly, patenting a mathematical technique may or may not make sense depending on what exactly is being patented. This is, sadly, somewhat subjective, but that's inherent in the patent system.
> polluting mathematics in particular and science in general with patents goes against the > whole concept of collaboration and open share of knowledge.
Does it, though? If I come up with a new way of doing signal processing that has interesting applications, I have two obvious options for monetizing it: keep it a trade secret and start a closed-source company to sell products based on it, or get a patent on the method, which involves sharing the method with others. That's the premise of the patent system, in fact. Perhaps you feel that this is not a good enough argument for patents; I might buy that. But then that's true for all patents, not just algorithm patents (whatever one might want to call them).
> I specifically said that a physicist should answer that question,
I do have a BS in physics. Doesn't make me a physicist, of course.
> to state that if we hold mathematics to be unpatentable, then it logically follows that > software should be equally so
Why are we holding "mathematics" (whatever that is in this case; the definition is pretty fluid, and I say that as someone with a good bit of experience with the subject) to be unpatentable?
> if a piece of software, *ANY* piece of software, can theoretically infringe on your > patents, your whole patent should be thrown out as far as I'm concerned.
So you keep saying, but WHY? I realize this is how you personally feel. What's not clear to me is what the argument for your feelings is that would actually convince someone, as opposed to you just repeating that you feel that way over and over again.
As far as I can tell, either you're saying we should have no patents at all (why or why not?) or you're applying a double standard based on some criteria you haven't bothered to define.
Slashdot Mirror
> can there be any good argument against greater energy efficiency?
Quite possibly, yes, locally.
For example, my water heater (not that old, not that new) is about 85% efficient. It's expected to have a useful life of another 10 years or so. I could get that up to 90% or so by getting a brand-new high-efficiency one. My net energy savings would be on the order of $50 a year at most, which works out to $500 over those ten years. A new water heater costs more than $500.
So the right thing to do in this case is to keep using the old heater for those 10 years. Note that this also avoids the energy use of _producing_ an extra water heater, so it's less of a loss to others than it would appear at first glance. It's clearly a win for me personally.
Now obviously this calculation depends on the price of energy, the price of water heaters, and the efficiency gains to be had. The last of these starts hitting diminishing returns quickly once what you have is anything resembling non-crappy; for example it's hard to reduce your energy usage by 2x if you're already 60% efficient.... Much easier to do if you're 10% efficient, of course.
. GP _asked_ what the plural is. So I told him, along with a good way to look up that information in the future (on the "teach a man to fish" premise).
The correct plural is "nemeses" (it's a Greek word, not Germanic or Latin as suggested respectively by your two proposed plural forms). Similar to how one pluralizes "axis", "synthesis", "analysis", "genesis", etc, and for the same reasons.
I should note that any sane dictionary will tell you what the plural form of a noun is. Or heck, googling "plural nemesis" in a pinch (first two hits are dictionary entries for "nemesis" that include the plural form). Just for future reference. ;)
> why the f--- would you outsource it
The US government "outsources" _all_ its military procurement, in the sense that they don't operate their own manufacturing facilities, don't necessarily do all the military R&D, etc.
So if, say, you're working for Northrop Grumman (the only manufacturer of US nuclear aircraft carriers and one of the two manufacturers of US nuclear submarines) and you're on a classified project and you leak the details of that project to some other country... then that's espionage, period.
It's receiving updates to the apps. The last update to the core OS itself was in September 2009, right when 10.6 shipped. They never announce end of support, so we won't know whether they've dropped it until they just haven't shipped any security fixes for a while.... and since they don't ship those on a regular schedule, no one knows what the right "while" is.
> How hard can that be?
Pretty hard, it turns out, unless you don't care how slow it is.
> I'm surprised you don't already have a generic text-rendering wrapper.
Not sure what you mean by that. There's a generic text-rendering API that is implemented via either ATSUI or CoreText right now, selected at runtime; that introduces a lot of complication and slows things down on Mac.
> You have the OS version right there in the user-agent string.
. I'm just passing on what I was told by the people working on the download stuff.
> Yes there is. Ship two binaries.
Is this the best course of action, though? Is that better than spending the same resources on making the browser work better on 10.5 and 10.6 instead?
What OS do you have installed on it, though? Apple supports its hardware a lot longer than it supports its software.
The information you seek is linked from the article summary.
Apple's normal security update cycle seems to be that they stop doing security updates to an OS about 2.5 years after last shipping it (which happens to correspond to when they start shipping the next+1 OS; in this case 10.6). They dropped 10.3 support pretty much when 10.5 came out, and that was also about 2.5 years after 10.4 appeared.
Mozilla's OS X usage statistics are in the second link from the summary....
Well... The question is not about current releases; the current release of Firefox supports 10.4. The question is about _next_ releases. The next release of Safari... who knows. They don't publicly talk about their release planning 18 months in advance, unlike Mozilla.
Note that Safari right this second ships different binaries for 10.4, 10.5, and 10.6. Of course this is made easier by the fact that most users get it via software update, not by having to download the right thing.
As for Camino, it's going to end up dropping 10.4 support once it switches to Gecko 1.9.3 or equivalent, just like Firefox.... It's Gecko that will probably drop 10.4 support, not the browser UI.
> Or you could have a legacy 10.4 build separate from the 10.5/6 build
Possible, at a fair amount of cost in maintenance and code complexity and a good bit of cost in user confusion when downloading builds.
> but it's probably still worth it IF there are sufficient users
That's the big question. Right now about 25% of the Firefox 3.5/3.6 Mac users are on 10.4. Where will it be a year from now? Who knows. In general, about 1% of current Firefox 3.5/3.6 users across all platforms are on Mac OS X 10.4.
> It's certainly more recent than the start of Windows XP, which is still well supported
XP is also a heck of a lot more like Win7 or Vista than 10.4 is like 10.6. That is, it's a lot easier to support XP. And yes, the large number of users probably doesn't hurt.
> Some older iMac users already have been shafted by Apple
They're shafted by Apple period. Apple has typicall been dropping support of its operating system versions within about 4 years of initially shipping them, and within about 2 years of last shipping them. As in, you buy a computer and 2 years later security updates for your OS stop. That's about the point at which Mozilla happens to stop security updates for your browser on that OS too, on the premise that you're insecure no matter what.
Do they? Will Safari 5 (which is the relevant comparison to the next version of Firefox) support 10.4?
> The last 'general' security update for 10.4 was on 10 Sept 2009
Which was right about when 10.6 shipped. At that point 10.4 was in fact the "previous OS".
The question is whether there will be more 10.4 security updates. And since Apple never announces EOL for its OSes officially, there's no way to know for sure.
Mozilla offers Firefox 3.6 on Tiger too.
The question is what Firefox whatever-the-next-version-is will be offered on, and the right comparison there is to Safari 5. Which will be offered.... somewhere. Who knows where.
> Firefox is still supported on Windows 2000,
That's because stuff that worked in Windows 2000 mostly still works in Vista or Win7.
The story is a bit less happy with 10.4 as compared to 10.6.
Microsoft takes backwards compatibility a lot more seriously than Apple does. Supporting both 10.4 and 10.6 well at the same time is somewhat comparable to supporting both Win98 and Vista at the same time (have to use different text rendering APIs, different graphics APIs, etc, etc).
Unknown. The last Tiger security update was in Sept 2009, as I recall. Apple never announces official end-of-life dates for its OSes as far as I can see; it just silently stops shipping security updates at some point. Since those don't ship on a predictable schedule to start with, one never knows whether you've seen the last one or not.
> Maybe Mozilla's done the user research and they know that they're not dropping many users,
The second link in the summary has the data on that. In brief, as of end of January 1010, 25% of Firefox 3.5 Mac users (about 1.4 million users, or about 0.5% of total Firefox users) are using OS 10.4. 12% of Firefox 3.6 Mac users (about 36,000 users) are using OS 10.4.
The big question mark, of course, is what those numbers will look like about 15 months from now, which is the earliest that Firefox 3.6 might be going out of support...
> Mozilla should have a very clear policy about backwards compatibility and follow it to the
> letter.
The basic setup is:
1) Once an OS vendor drops support for an OS, support for it will not be maintained unless
it's really easy to do.
2) Whether an OS is supported depends on whether there are resources to support it and on
how many users are using it.
It's not exactly a clear policy, but the important part is that support decisions are pretty complicated and involve a lot of factors.... it's not clear to me what a sane policy would be that would not lead to dropping support in some cases when there's no real reason to do it.
> Until Apple actively does something to break the older "deprecated" code in Firefox,
> they should support older OSes
10.6 dropped ATSUI support. 10.4 doesn't have Core Text. So the only way to support both is to have codepaths to use both text rendering backends and switch at runtime. Does that count as "does something to break"? ;)
Thing is, it's all software. Everything can be worked around. The question is the cost (to users, in the end, either in terms of money or in terms of things users want that don't happen).
> and you deal with the slightly older APIs/compilers to serve your users
Not that simple. You have to use gcc 4.0 if you're going to run on 10.4. So doing that serves the 10.4 users. But on 10.5 and 10.6, using gcc 4.2 gives a pretty significant across-the-board speedup. So to properly serve those users, you want to be using gcc 4.2. Where that leaves you is either underserving 10.5/10.6 users to better serve 10.4 users or vice versa (at which point relative numbers of users start to matter), or shipping separate binaries with the ensuing user confusion during downloading, etc. So there's not an obvious course of action here that best serves "the users". It's a matter of compromise.
> It would be very ironic if Chrome running under proprietary Windows and OS X could play
> Theora, while Chrome on Linux would only support H.264.
Chrome supports Theora out of the box natively, so I'm not sure what you're talking about...
Ah, indeed. With that correction you may well be right.
> You'll note nothing seems to get cheaper to the end user.
Since we're talking about computer equipment, this is demonstrably false.
> But most importantly, because that's the law in a huge part of the world, the US included
Ok, so you're saying the existing law already has a double standard and your quibble is with where the boundary the law draws in applying this double standard lies? I can buy that.
That said, I see no inherent issue with being able to patent a clever way of combining mathematical operations to, say, extract a signal from a noisy data stream, if we're ok with being able to patent combining spiky bits and a roller to separate cotton seeds from cotton fibers.
See http://www.paulgraham.com/softwarepatents.html for some more thoughts on this (I just ran into it when I searched on "patent algorithm", and it eerily echoes exactly what I was saying earlier).
Now there is an important point here: If you're going to have patents, they must be granted for specific inventions, not overbroad descriptions. Patenting "a roller with teeth" would never fly; patenting a roller with teeth being used in a particular way to get a particular effect is a different issue entirely. Similarly, patenting a mathematical technique may or may not make sense depending on what exactly is being patented. This is, sadly, somewhat subjective, but that's inherent in the patent system.
> polluting mathematics in particular and science in general with patents goes against the
> whole concept of collaboration and open share of knowledge.
Does it, though? If I come up with a new way of doing signal processing that has interesting applications, I have two obvious options for monetizing it: keep it a trade secret and start a closed-source company to sell products based on it, or get a patent on the method, which involves sharing the method with others. That's the premise of the patent system, in fact. Perhaps you feel that this is not a good enough argument for patents; I might buy that. But then that's true for all patents, not just algorithm patents (whatever one might want to call them).
> I specifically said that a physicist should answer that question,
I do have a BS in physics. Doesn't make me a physicist, of course.
> to state that if we hold mathematics to be unpatentable, then it logically follows that
> software should be equally so
Why are we holding "mathematics" (whatever that is in this case; the definition is pretty fluid, and I say that as someone with a good bit of experience with the subject) to be unpatentable?
> if a piece of software, *ANY* piece of software, can theoretically infringe on your
> patents, your whole patent should be thrown out as far as I'm concerned.
So you keep saying, but WHY? I realize this is how you personally feel. What's not clear to me is what the argument for your feelings is that would actually convince someone, as opposed to you just repeating that you feel that way over and over again.
As far as I can tell, either you're saying we should have no patents at all (why or why not?) or you're applying a double standard based on some criteria you haven't bothered to define.