I guess when the FSF talks about "free software", they don't mean "free to you", but that the software it self is somehow "free" -- i.e., not locked up in a proprietary product. The limitations in the GPL as to what you can do to the software isn't to keep you free (you're already in a free country) but to keep the software free -- i.e., to protect it from you.
Yes, I totally agree with you here. I fear that most people don't understand this though. The 'Free Software' slogan sounds good. I figure it makes people think they are gaining some sort of freedom by using it, silimar to living in a free country. But it is the *code* that is ensured this freedom. If the term 'Protected Software' or some other more descriptive term was used I don't the GPL would have nearly as many die hard supporters.
Once you assign the freedom to an non living object, a lot of warm fuzzy feeling goes away. The moral high ground isn't so obvious anymore.
Anyways, it's unlikely to change given how effective the 'Free Software' slogan is. That slogan is simply good marketing.
On a side note, another reason people release BSD code is to promote quick adoption of a proposed standard. Probably the best example of this is the TCP/IP stack. The success of the BSD based TCP/IP stack is evidenced by the fact that almost all operating system's TCP/IP stacks are based off of the BSD TCP/IP stack. Even Windows uses the BSD TCP/IP stack.
Of course, I'm not trying to suggest the GPL doesn't have it's uses. I've even released a project as GPL. But I did not do it out of kindness. Rather I did it out of greed.
I understand the reasons behind the GPL, but the analogy with a free country is completely wrong. This is my main gripe with the FSF dogma. Unlike in a country where anarchy is in conflict with the freedoms garunteed, public domain code is not in conflict with any freedoms. For instance, people who write software are free to keep it closed, open, or put it under a license somewhere in between.
Public domain code can be considered free code since all of the above freedoms are preserved when the code is extended. The author of the extensions is free to do whatever they please with the code.
The GPL on the other hand takes away my freedom of choosing a license. It is therefore not free as in a free country. It is freer than proprietary software and freeware, but it is not as free as BSD, which is not as free as public domain.
I don't really believe that GPL code has the rights to use the "Free Software" dogma. I'd have no problem if it was advertised as "Share Software" or something similar. But the analogy with a free country is not correct.
When I write code it's either under the BSD or a propietary license. GPL sofware is just as closed to me as free to use proprietary code. I can't use the source since I don't use the GPL.
Most people who release BSD code have benefitted from BSD code in the past. In these cases GPL software was likely too restrictive to use. There are many exaples, grad students working on code with a different license, inexperienced developers making corporate code (remember, many of these people are just trying to live a decent life), perhaps people working on BSD based projects. Such developers feel grateful to the BSD community and often want to return the favour. We don't really care to impose our will upon others though. We give without expecting anything in return. A true gift, as everyones parents hopefully taught them.
BSD: free as in 'Look someone else did our work for us, and we don't have to give anything back'
Umm. Look, this is EXACTLY what BSD developers expect. This why it's released BSD. Shall I break out the dictionary definition of free? I'll give you a hint, it means having no restrictions. The GPL is very much a restriction. It is a very different philosophy than BSD and public domain code. It is in no way free as in the definition of the word.
Rather, it caters to a specific ideology, namely that of the FSF. Granted, you may consider GPL to be free, and it is certainly freer than proprietary software, but it is not quite as free as BSD and public domain.
You can be selfish and require people to share with you if you like. But, statements like yours are really irritating to those of us who want to give our code away for free.
The guys comment was very informative given the news story. Linksys doesn't want to 'share' with you. They shouldn't have used GPL code then, but rather BSD code. Right?
And please don't go on about how they *should* share. They are under no obligation to do so, if they did the country they operate from would certainly not be considered a free country anymore.
I agree with you completely, I'm currently a grad student.
Interestingly, in Canada the trend is the opposite. More and more Canadian universities are giving students almost complete IP rights. For instance, my university does this, they only require that they can have eternal free use of whatever I come up with while I'm there. Not a bad deal at all.
Of coures maybe things are different in America? Are grad students in funded exclusivly by the government as they are here? If not I suppose this law may not have as much of an impact. If it is the same, then this could perhaps hurt US academic research community.
I've been using Debian GNU/Linux (unstable) for the last 3-4 years and it runs perfectly stable.
Good for you. I've been using for a year now, at work. It's not perfectly stable. It breaks *all* the time actually. Not to mention, there are still some things that are just not avaliable even for unstable. Openoffice 1.1 for example. And better still, 1.0 coredumps on start for me. Yay. Go superior distribution!
And I suppose gcc is not inferior on the G5? Yes you should use the fastest compilers on each system. But we have yet to see any results, or any good compiler for Apples system.
If they are understating performance, they are doing it to themselves as well. They are in no way lying. It's not like they are hiding the raw results or the configuration of the machines. Really, get a grip. If you really think that a heavily optimised G5 will be super slow compared to a heavily optimised P4, then great. We'll all find out once the tools are avaliable.
Until then, these results are certainly more informative to me than comparing gcc on G5 to icc on P4.
True, but if the OS can't use a 64-bit address space, can it use the G5 SIMD? What if you are working with large floating point values, or long integers? Even 32-bit procesors have 64 or 128 bit extensions for special math operations.
What other features of the G5 can't it use? We really don't have enough information to know much of anything. It's interesting that the G5 peformed as well as it did in the test environment.
It depends very much on the chip design. It may be that the G5 is optimised for running in 64-bit mode. We really have to wait for the hardware to be avaliable before we know real world performance.
If benchmarking a 64-bit processor against a 32-bit is obviously ridiculous, then why do you have a problem with Apple claming that this processor is more powerful than the P4?
I guess since the G5 doesn't have SSE and hyperthreading than it's also ridiculous to use those right? Really, get a damn grip.
Your car analogy is incorrect. If I create a program that allocates a huge array once, then just munches the data in it, the allocation library will have zero performance impact. If I allocate and deallocate constantly, then I'm benchmarking the memory manager and not the CPU. Spec designed this test to test the CPU. Good design for high performance code is to reuse data structures so you don't get the penalty of construction and deconstruction.
I said it's unclear what performance impact the allocation library will have. It is certainly not likely that it will make drastic changes to the results and if it does, them perhaps the CPU benchmark is not doing the greatest job of really benchmarking the CPU.
Do you really need to use insults to get your point across? It really doesn't help your argument.
The scores on spec.org were achived with a different compiler. To be fair to Apple, you would have to compare those agains a compiler and operating system that excells at supporting the G5. The OS 10.3 is not ready yet. I assume that sometime a compiler that is good with the G5 will also be released.
Right now all you can do is guess. The current comparison is demostrating the performance for a typical usage. This is not understating anything. More likely it's just clever marketing.
I just read the veritest document myself. This guy did not show the complete picture at all. First, in favour of Apple, these tests were run on Mac OS X 10.2.7. I don't think this is the 64-bit kernel that can really take advantage of the G5. I'm not a mac head, but I'm guessing this is Smeagol?
So right from the start the G5 is seriously crippled in these tests. Especially if they don't even take advantage of 64-bit as seems to be the case. Now, on the the other points that the spl dude makes.
Sure the special malloc library seems a bit unfair. But then again, do these tests really focus on memory allocation? I'd think they are limited by CPU power more than memory allocation. In any case, it'd be nice if we saw results without this library.
Now for the comfusing part. The 8300 only has a single cpu. For the base tests, they use hyperthreading and an SMP kernel. They do the exact same test for the 650's base test, hyperthreading with a single processor and an SMP kernel. The G5 system is run with a single processor as well.
So it seems this test is well balanced and fair. The confusing part is the rate tests.
For the 8300 they have no hyperthreading and a uniprocessor kernel. For the 650 they have no hyperthreading and an SMP kernel with two processors. The G5 system is run with two processors. It's unclear why they chose not to use hyperthreading on the rate test. It could be that hyperthreading actually reduced the scores of these tests. I'm no expert on the SPEC tests and hyperthreading, but what I do know is that hyperthreading is an intelligent technology. It can't always increase speed, it depends on what kind of code it's running. In the rate test it's possible that hyperthreading is unable to yeild any improvements, in which case the overhead of enabling hyperthreading may make the scores worse than without hyperthreading.
At anyrate, the tests were a LOT more fair than the dpl guy makes them out. And considering that the G5 could be seriously crippled by not running 64-bit and who knows what other optimisatoins, I'd say that the numbers are still impressive.
Yep. That's true. I don't imagine we'll see SMP in a production OpenBSD anytime soon. In fact, the spinlocks project, which was working on SMP for OpenBSD, seems dead. Oh well.
So when someone says you can't make money with the GPL and therefore the GPL is bad, you get a 1000 posts saying how you CAN make money.
Now when somone really doesn't make money, everyone yells at him saying that's not what the GPL is for. Of course, it's probably not the same people posting the differing views.
The take home message? Don't blindly trust anything you read on slashdot. The GPL really doesn't make you any money. No matter what you say bad about the GPL, you will get flamed.
I know OpenBSD has a group in a totally different project working on SMP for OpenBSD. The different BSDs have fairly different kernel internals as far as I know. It certainly wouldn't be a cut and paste job.
I don't think Free Software (GPL) is free as in freedom nor do I believe it is good for the world. It is akin to a religous cult. Please take your preaching to the church (FSF).
My original comment was a response to a comment that read, to me at least, as a very condescending "you just don't get it" type of dismissal. I haven't been slumming it in/. for very long (just check out the user# in the high 6 figures), but I'm aware the BSD gets a rap as the elitist's OS. To go ahead and make the connection was obvious...
Yes, I kind of realized this after I first posted. The 'make fun of anything but linux' attitude of the slashdot community as a whole just really gets to me sometimes. It's simply very dissapointing to read comments on a very important release of an interesting operating system and have the first five to ten highlighted comments simply making jokes about it.
I agree the response you referred to was a little insulting. Mind you, the original comment on java was also very likely only made to make fun of BSD as well.
On a side note, you may be interested to know that BSD is *very* similar to Solaris in many ways. For instance, the disk slice editor, bad or good, is almost identical to Solaris'.
Anyways, forgive my overly negative initial response.:) It was only made out of frustration with the the very vocal portion of Linux community.
p.s. VMS? It's interesting to find someone still using it. I hear it has/had some *very* intersting architectural aspects. I've only breifly played with a VMS system. Back when the 90% of people didn't yet know what the internet was.
You have the exasperating and exasperated tone of a proselytising atheist who's just been told that atheisim is a religion, just one that worships the absence of God. For "God and atheisim" read "Linux and BSD".
Ok, now this really SHOULD be moderated as funny. You totally crack me up.:)
Oh, hey, big words don't make up for swearing.;)
Oh hell. I'll bite. Guess what? I use Linux,and BSD, and Windows XP Yes, I actually have machines with each installed.
It's clear you haven't spent much time in the BSD community. I've yet to see a RTFM, or a 'why don't you code it' comment. There may be some anti-GPL comments, but that's another issue entirely.
On the other hand, whenever a story about BSD is posted here on slashdot, there are a dozen or so jokers like you who like to bad mouth the OS and/or it's users. Tell me, how many such wise cracks get moderated up to funny if they are about Linux? It's very obvious that this is mostly a Linux centric site.
If you happen to be one of the probably less than 20% of the readers who is *not* a linux fan, then I apologize for calling you one. Regardless, it's not funny to make fun of another community, especially one you haven't spent much time in.
And don't mistake my sarcasm for making jokes about the linux community. I am critisizing it in response to an uncalled for outburst.
Here we have a company that could easy just release closed source software. They may even make more money that way. Instead they release the current version free for non-commercial use and the previous version GPL free.
Where the open source community should be thankful for this and applaud a potentially useful mix of semi-open source commerical software with GPL software, all that anyone can do is complain that they promote themselves in the GPL version.
Well. It's GPL. You can always remove the recommendation from the readme. The high and mighty ideology of the FSF is really really tiring.
Well, if I use a debit card the fees are less for the store maybe (actually zero for the store), but lots more for me. At least here in Canada, I get charged something absurd like $1.25 a purchase. That adds up quickly. Visa gives me no charges, but does charge the store, which is better for me.
I don't think there is a way around the bank tax, it's just a matter of who pays I guess. It's an unavoidable fact of internet commerce.
I would have thought this too, but now I'm not so sure. My grocery store takes visa, and has lower price limit. I asked about this because I thought that they'd be losing a lot of money if people only bought one or two things per day (like me). Apparently, they have a deal that they get unlimited trasactions for a set fee.
Perhaps Apple has a similar deal? Or maybe the lady didn't know what she was talking about?
Ah, I've double check edmy sources. Apparently it is the companies I'm dealing with that refuse to ship with Canada Post. And UPS and FedEX both charge brokerage fees that is often close the price of the item. My mistake. Sorry. Should check facts before let lips flap.
Slashdot Mirror
I guess when the FSF talks about "free software", they don't mean "free to you", but that the software it self is somehow "free" -- i.e., not locked up in a proprietary product. The limitations in the GPL as to what you can do to the software isn't to keep you free (you're already in a free country) but to keep the software free -- i.e., to protect it from you.
Yes, I totally agree with you here. I fear that most people don't understand this though. The 'Free Software' slogan sounds good. I figure it makes people think they are gaining some sort of freedom by using it, silimar to living in a free country. But it is the *code* that is ensured this freedom. If the term 'Protected Software' or some other more descriptive term was used I don't the GPL would have nearly as many die hard supporters.
Once you assign the freedom to an non living object, a lot of warm fuzzy feeling goes away. The moral high ground isn't so obvious anymore.
Anyways, it's unlikely to change given how effective the 'Free Software' slogan is. That slogan is simply good marketing.
On a side note, another reason people release BSD code is to promote quick adoption of a proposed standard. Probably the best example of this is the TCP/IP stack. The success of the BSD based TCP/IP stack is evidenced by the fact that almost all operating system's TCP/IP stacks are based off of the BSD TCP/IP stack. Even Windows uses the BSD TCP/IP stack.
Of course, I'm not trying to suggest the GPL doesn't have it's uses. I've even released a project as GPL. But I did not do it out of kindness. Rather I did it out of greed.
I understand the reasons behind the GPL, but the analogy with a free country is completely wrong. This is my main gripe with the FSF dogma. Unlike in a country where anarchy is in conflict with the freedoms garunteed, public domain code is not in conflict with any freedoms. For instance, people who write software are free to keep it closed, open, or put it under a license somewhere in between.
Public domain code can be considered free code since all of the above freedoms are preserved when the code is extended. The author of the extensions is free to do whatever they please with the code.
The GPL on the other hand takes away my freedom of choosing a license. It is therefore not free as in a free country. It is freer than proprietary software and freeware, but it is not as free as BSD, which is not as free as public domain.
I don't really believe that GPL code has the rights to use the "Free Software" dogma. I'd have no problem if it was advertised as "Share Software" or something similar. But the analogy with a free country is not correct.
When I write code it's either under the BSD or a propietary license. GPL sofware is just as closed to me as free to use proprietary code. I can't use the source since I don't use the GPL.
Most people who release BSD code have benefitted from BSD code in the past. In these cases GPL software was likely too restrictive to use. There are many exaples, grad students working on code with a different license, inexperienced developers making corporate code (remember, many of these people are just trying to live a decent life), perhaps people working on BSD based projects. Such developers feel grateful to the BSD community and often want to return the favour. We don't really care to impose our will upon others though. We give without expecting anything in return. A true gift, as everyones parents hopefully taught them.
BSD: free as in 'Look someone else did our work for us, and we don't have to give anything back'
Umm. Look, this is EXACTLY what BSD developers expect. This why it's released BSD. Shall I break out the dictionary definition of free? I'll give you a hint, it means having no restrictions. The GPL is very much a restriction. It is a very different philosophy than BSD and public domain code. It is in no way free as in the definition of the word.
Rather, it caters to a specific ideology, namely that of the FSF. Granted, you may consider GPL to be free, and it is certainly freer than proprietary software, but it is not quite as free as BSD and public domain.
You can be selfish and require people to share with you if you like. But, statements like yours are really irritating to those of us who want to give our code away for free.
The guys comment was very informative given the news story. Linksys doesn't want to 'share' with you. They shouldn't have used GPL code then, but rather BSD code. Right?
And please don't go on about how they *should* share. They are under no obligation to do so, if they did the country they operate from would certainly not be considered a free country anymore.
I agree with you completely, I'm currently a grad student.
Interestingly, in Canada the trend is the opposite. More and more Canadian universities are giving students almost complete IP rights. For instance, my university does this, they only require that they can have eternal free use of whatever I come up with while I'm there. Not a bad deal at all.
Of coures maybe things are different in America? Are grad students in funded exclusivly by the government as they are here? If not I suppose this law may not have as much of an impact. If it is the same, then this could perhaps hurt US academic research community.
I've been using Debian GNU/Linux (unstable) for the last 3-4 years and it runs perfectly stable.
Good for you. I've been using for a year now, at work. It's not perfectly stable. It breaks *all* the time actually. Not to mention, there are still some things that are just not avaliable even for unstable. Openoffice 1.1 for example. And better still, 1.0 coredumps on start for me. Yay. Go superior distribution!
Nice link there. Good to know I don't have to eat my speculation on the hyperthreading. ;)
If I only had mod points...
And I suppose gcc is not inferior on the G5? Yes you should use the fastest compilers on each system. But we have yet to see any results, or any good compiler for Apples system.
If they are understating performance, they are doing it to themselves as well. They are in no way lying. It's not like they are hiding the raw results or the configuration of the machines. Really, get a grip. If you really think that a heavily optimised G5 will be super slow compared to a heavily optimised P4, then great. We'll all find out once the tools are avaliable.
Until then, these results are certainly more informative to me than comparing gcc on G5 to icc on P4.
True, but if the OS can't use a 64-bit address space, can it use the G5 SIMD? What if you are working with large floating point values, or long integers? Even 32-bit procesors have 64 or 128 bit extensions for special math operations.
What other features of the G5 can't it use? We really don't have enough information to know much of anything. It's interesting that the G5 peformed as well as it did in the test environment.
It depends very much on the chip design. It may be that the G5 is optimised for running in 64-bit mode. We really have to wait for the hardware to be avaliable before we know real world performance.
If benchmarking a 64-bit processor against a 32-bit is obviously ridiculous, then why do you have a problem with Apple claming that this processor is more powerful than the P4?
I guess since the G5 doesn't have SSE and hyperthreading than it's also ridiculous to use those right? Really, get a damn grip.
Your car analogy is incorrect. If I create a program that allocates a huge array once, then just munches the data in it, the allocation library will have zero performance impact. If I allocate and deallocate constantly, then I'm benchmarking the memory manager and not the CPU. Spec designed this test to test the CPU. Good design for high performance code is to reuse data structures so you don't get the penalty of construction and deconstruction.
I said it's unclear what performance impact the allocation library will have. It is certainly not likely that it will make drastic changes to the results and if it does, them perhaps the CPU benchmark is not doing the greatest job of really benchmarking the CPU.
Do you really need to use insults to get your point across? It really doesn't help your argument.
The scores on spec.org were achived with a different compiler. To be fair to Apple, you would have to compare those agains a compiler and operating system that excells at supporting the G5. The OS 10.3 is not ready yet. I assume that sometime a compiler that is good with the G5 will also be released.
Right now all you can do is guess. The current comparison is demostrating the performance for a typical usage. This is not understating anything. More likely it's just clever marketing.
I just read the veritest document myself. This guy did not show the complete picture at all. First, in favour of Apple, these tests were run on Mac OS X 10.2.7. I don't think this is the 64-bit kernel that can really take advantage of the G5. I'm not a mac head, but I'm guessing this is Smeagol?
So right from the start the G5 is seriously crippled in these tests. Especially if they don't even take advantage of 64-bit as seems to be the case. Now, on the the other points that the spl dude makes.
Sure the special malloc library seems a bit unfair. But then again, do these tests really focus on memory allocation? I'd think they are limited by CPU power more than memory allocation. In any case, it'd be nice if we saw results without this library.
Now for the comfusing part. The 8300 only has a single cpu. For the base tests, they use hyperthreading and an SMP kernel. They do the exact same test for the 650's base test, hyperthreading with a single processor and an SMP kernel. The G5 system is run with a single processor as well.
So it seems this test is well balanced and fair. The confusing part is the rate tests.
For the 8300 they have no hyperthreading and a uniprocessor kernel. For the 650 they have no hyperthreading and an SMP kernel with two processors. The G5 system is run with two processors. It's unclear why they chose not to use hyperthreading on the rate test. It could be that hyperthreading actually reduced the scores of these tests. I'm no expert on the SPEC tests and hyperthreading, but what I do know is that hyperthreading is an intelligent technology. It can't always increase speed, it depends on what kind of code it's running. In the rate test it's possible that hyperthreading is unable to yeild any improvements, in which case the overhead of enabling hyperthreading may make the scores worse than without hyperthreading.
At anyrate, the tests were a LOT more fair than the dpl guy makes them out. And considering that the G5 could be seriously crippled by not running 64-bit and who knows what other optimisatoins, I'd say that the numbers are still impressive.
Yep. That's true. I don't imagine we'll see SMP in a production OpenBSD anytime soon. In fact, the spinlocks project, which was working on SMP for OpenBSD, seems dead. Oh well.
So when someone says you can't make money with the GPL and therefore the GPL is bad, you get a 1000 posts saying how you CAN make money.
Now when somone really doesn't make money, everyone yells at him saying that's not what the GPL is for. Of course, it's probably not the same people posting the differing views.
The take home message? Don't blindly trust anything you read on slashdot. The GPL really doesn't make you any money. No matter what you say bad about the GPL, you will get flamed.
Maybe.
I know OpenBSD has a group in a totally different project working on SMP for OpenBSD. The different BSDs have fairly different kernel internals as far as I know. It certainly wouldn't be a cut and paste job.
Umm, not to rain on the parade, but is there any possability that this article is a hoax?
Does someone have a second source?
It's always good to make sure the evidence is real before your crucify someone.
I don't think Free Software (GPL) is free as in freedom nor do I believe it is good for the world. It is akin to a religous cult. Please take your preaching to the church (FSF).
Oh bother, forgot to close the italic tag. Can't I edit slashdot comments yet?
Yes, I kind of realized this after I first posted. The 'make fun of anything but linux' attitude of the slashdot community as a whole just really gets to me sometimes. It's simply very dissapointing to read comments on a very important release of an interesting operating system and have the first five to ten highlighted comments simply making jokes about it.
I agree the response you referred to was a little insulting. Mind you, the original comment on java was also very likely only made to make fun of BSD as well.
On a side note, you may be interested to know that BSD is *very* similar to Solaris in many ways. For instance, the disk slice editor, bad or good, is almost identical to Solaris'.
Anyways, forgive my overly negative initial response. :) It was only made out of frustration with the the very vocal portion of Linux community.
p.s. VMS? It's interesting to find someone still using it. I hear it has/had some *very* intersting architectural aspects. I've only breifly played with a VMS system. Back when the 90% of people didn't yet know what the internet was.
You have the exasperating and exasperated tone of a proselytising atheist who's just been told that atheisim is a religion, just one that worships the absence of God. For "God and atheisim" read "Linux and BSD".
Ok, now this really SHOULD be moderated as funny. You totally crack me up. :)
Oh, hey, big words don't make up for swearing. ;)
Oh hell. I'll bite. Guess what? I use Linux,and BSD, and Windows XP Yes, I actually have machines with each installed.
It's clear you haven't spent much time in the BSD community. I've yet to see a RTFM, or a 'why don't you code it' comment. There may be some anti-GPL comments, but that's another issue entirely.
On the other hand, whenever a story about BSD is posted here on slashdot, there are a dozen or so jokers like you who like to bad mouth the OS and/or it's users. Tell me, how many such wise cracks get moderated up to funny if they are about Linux? It's very obvious that this is mostly a Linux centric site.
If you happen to be one of the probably less than 20% of the readers who is *not* a linux fan, then I apologize for calling you one. Regardless, it's not funny to make fun of another community, especially one you haven't spent much time in.
And don't mistake my sarcasm for making jokes about the linux community. I am critisizing it in response to an uncalled for outburst.
And yet /.ers still portray *BSD'ers as elitist assholes. How the hell can this be so?
And yet another linux fan gets moderated to funny making fun of another community! Wow! I am so surprised!
Oops, I forgot to swear in my post. Now I'll never be moderated as funny. Darn.
Right on.
Here we have a company that could easy just release closed source software. They may even make more money that way. Instead they release the current version free for non-commercial use and the previous version GPL free.
Where the open source community should be thankful for this and applaud a potentially useful mix of semi-open source commerical software with GPL software, all that anyone can do is complain that they promote themselves in the GPL version.
Well. It's GPL. You can always remove the recommendation from the readme. The high and mighty ideology of the FSF is really really tiring.
Well, if I use a debit card the fees are less for the store maybe (actually zero for the store), but lots more for me. At least here in Canada, I get charged something absurd like $1.25 a purchase. That adds up quickly. Visa gives me no charges, but does charge the store, which is better for me. I don't think there is a way around the bank tax, it's just a matter of who pays I guess. It's an unavoidable fact of internet commerce.
I would have thought this too, but now I'm not so sure. My grocery store takes visa, and has lower price limit. I asked about this because I thought that they'd be losing a lot of money if people only bought one or two things per day (like me). Apparently, they have a deal that they get unlimited trasactions for a set fee.
Perhaps Apple has a similar deal? Or maybe the lady didn't know what she was talking about?
Ah, I've double check edmy sources. Apparently it is the companies I'm dealing with that refuse to ship with Canada Post. And UPS and FedEX both charge brokerage fees that is often close the price of the item. My mistake. Sorry. Should check facts before let lips flap.