No, but I think some albumns exist as whole peices for various reasons.
Beatles albums are good examples since they often have very distinct styles. Sgt Pepper vs, say Magical Mystery Tour.
One concern I have is that if people *never* buy or download an entire albumn, they'll never get the musician's message.
Does that mean I say to force them to listen to all of it. No, but it does mean it's something to consider when downloading only one song off an album.
Also, if you're only hearing (and therefore interested) in ONE song, then it's probably due to that song being on the radio. That's pretty limited and you might want to expand your horizons past what some large corporation tells you to like.
There are some music peices which are really part of a whole.
Examples that come to mind are Pink Floyd's Dark Side of the Moon, Kraftert's Man Machine or others.
An artist sometimes puts an albumn together not JUST to fill up a record or CDs worth of music but as a single thing. When you say you want to rip one part of it out, I think you're doing a disservice to the artist.
The issue for me is about control and morals. The moral issue is that of cultural exchange- and that's what music, books and movies are. The issue of control is the issue of 'to what length will these companies to go keep control over consumers'.
This is a question that mainly applies to the US (the original poster was discussing a US company).
We in the US have a belief (which is built into our law) that a person is presumed innocent of wrongdoing until proven otherwise.
It applies to courts but, like many laws should be- the law is an extension of our collective views. We, as a society believe that someone is innocent until proven guilty.
I do not know Mexico well. I've never been there. People I do know who have lived there have told me that the justice system in Mexico is certainly different than in the US in a variety of ways.
If you have a different way and it works for you, then that's fine, but I believe that innocent until proven guilty is a good idea, and I think that, if it's not, it should be placed into an international agreement on how laws work (Human Rights perhaps).
I also believe that in some countries with Napoleonic law, those who accuse, if they're found to be wrong, are liable. That's not so in the US.
The point is the parent of my original post was saying that there was no legitimate reason to use a certain amount of bandwidth, so those who use more must therefore be doing something illegitimate (implying illegal).
I was arguing that this is not a good position to take- that innocence until proven guilty is a good thing, and I provided answers to "legitimate use" while also pointing out that "legitimate" vs "illegitimate" was never spelled out.
Bandwidth bills on a static page are also trivial.
A well behaved program won't go GETs on every RSS page, but will do HEADS, compare them to what it already has, and decide from there to get or not get the new page.
A HEAD request is very small, and unless you're doing millions of them, this shouldn't be an issue.
I don't think the original poster is anti-Unix, he just feels that the "GUI" aspect of X should encompas sound, just like X servers now handle keyboard and mouse events.
Sound apps not needing to know the hardware is no different than X applications not needing to know the hardware when they make xlib calls.
Right now, every system that wants to provide network transparent sound has to reinvent the wheel since no one can agree on the "right" way to do it. Having one chosen as the "blessed" one by Xouert may end the argument.
As to requiring esd outside X, I don't see much of the point. I suppose you *could* do sound in consoles with something like emacs or screen, but that's highly unusual. Most people want sound events locally, so outside X you can send events right to/dev/dsp.
As to the issues of large monolithic code, I tend to agree, but think that the original poster's comment didn't necessarily imply code bloat, only a "blessed' X sound event system that comes with a new X server and makes it easier for application developers to write code that will run sound over the network (which will make projects like the Linux Terminal Server project easier).
The reason that no interoperabiluty exists is there's no incentive. Why not? Users aren't demanding it. Why aren't users damanding it? They're not demanding it becuase they don't percieve a need for it? Why not? Becuase in thier minds, they have a solution, which is the various programs such as Gaim.
Now:
1) If you'd read the first comment, I discuss the need for interoperability. The comment I made about Gaim was explaining why this level of interoperability hasn't been demanded.
2) You didn't bother following the link I provided which explains, in detail, my view on IM protocol interoperability. Had you bothered to follow the link and read, instead of just shooting off two posts, you would have seen that I share your view.
If you did follow the link, then you didn't read it, or didn't understand it. I can't imagine how, since I made my views very clear.
3) You seem to have difficulty understanding how I can explain a phenomenon in terms of someone's perception other than my own. I'm saying that the reason there hasn't been interoperability is that there hasn't been user demand. The way to change that is to either:
a) Change user demands
b) Provide some kind of incentive for change
c) Create a requirement for change (regulation, etc.)
But I wasn't addressing the mechanism for making change, rather adressing why it hasn't happened yet.
In fact, I was complaining that Gaim, while useful (I use it) do a disservice to people trying to provide interoperability between the IM networks.
That's exactly what I said- that programs that let users use multiple programs reduce the need for interoperability in user's minds meaning there isn't the outcry there might otherwise be.
I'm a bit unclear about the differences between SIP and XMPP and where Jabber, which could have been used as an interoperability standard, all fit together.
At the high end, these all seem like simple namespace issues and would map onto Jabber nicely. An AIM user, for example, could be user@aim but the end user doesn't need to know that, they could just be presented with an icon representing AOL or something.
The real issue is that there doesn't seem to be much in the way of motivation into making the IM systems compatible. Due to programs like Gaim and Trillian, less users are demanding it.
So the question for me is where's the money in it for AOL, Yahoo and Microsoft? AOL is clearly selling IM to companies like T-Mobile for phone connectivity, but I can't imagine that "pays the bills" for the millions of messages they route.
How is IMing profitable and what incentive do these companies really have in making interoperability a reality?
BTW- I see the real danger here in that IM is replacing other technology, such as TTY. I even wrote about this in my blog about a month ago:
(yes shameless plug)
Actually, my girlfriend is deaf and carries a cell phone. It works pretty well.
She has one of those T-Mobile Sidekick ones where you connect the Internet (web browser, AIM, and even an available SSH client).
It's actually quite useful. She can get her email, AIM and SMS messages in one place. There are even AIM -> TTY services so she can make "voice" calls on the road.
It's still a phone too. So if there's some sort of emergency and a hearing person is there- they could use it.
You're making a number of assumptions which I don't think always hold true.
You say that you could do these measurements without the GA. That's true, but the GA serves two purposes:
1) To do the measruements of the combinations of
optimizations (which you mention)
2) To do the thing without human intervention
The second one is the important one. The fact is that few people are going to spend the time and energy to do the optimizations "right". A few hours or days at some point figuring out "pretty good" compiler options seems very reasonable though and I can see a lot of my own users doing it.
The issue of "is compiler optimization worth it?" depends on your problem domain. But if you're doing huge numbers of computations that may take hours to compete a single run, the answer is a resounding "yes" with a caviot.
The caviot is that code optimization must come first. We know the compile can't fix bad code, and good code optimization can improve things in ranges we'll never see from simple compiler optimizations.
Nonetheless, if a compiler optimization can shave, say 10 minutes from one run of a program that takes 3 hours to run- it's worth it. Why? Becuase that program is going to be run at least 1000 times. That ten minutes now becomes about six days of wall time. I can do a lot of processing in six days.
The idea isn't to attack at all, rather to reply as an interested customer.
The scenario is that you recieve a mail about getting, say pills that make your nostrils bigger. All spammers will need a way to ensure that you can make a purchase, and it's through that mechanism that you inquire for more information about nostril enhancement through magic pills.
If everyone who recieved an email did this, they would get thousands of requests.
If they only reply to a few of them then the company selling the pills looses sales.
So instead, they hire more staffers. When they do that, they are potentially eating into thier own profits.
Given sufficient numbers of respondants, this would make it suddenly unprofitable to mail everyone in the world, leading to an incentive to stop mass spamming.
That's the idea at least. There's no "attack" involved.
While people usually talk about X compatibility- that's not usually what they want. They want application compatibility. They want Mozilla, GNOME, Emacs to work on their computer *usually*.
Look at Opie and you realize that, given sufficient numbers of apps, you don't care as much about the libraries.
So what's important would be to port things like GTK and QT to the new target (however you choose to do that, be it at the X toolkit layer or the GDK layer, or whatever).
An example of the success of this method are the number of Free applications for Microsoft Windows and OS X.
While some apps talk to X directly, the number of users who will be effected is smaller.
If something else can provide the ability to use the same apps, allow same/similar features and provide some benefits, then I don't see why we can't give it a chance, even if it breaks one compatibility.
It's the infinite compatibility problem that has forced x86 down people's throats for so many years.
Sometimes you need to make a new protocol and provide a migration path.
And will it have the same features as the original, including:
* Power supply for computer located inside printer * Being able to boot off tapes, BUT if you boot up
with tapes inside, the magnetic field will
erase them * Chip degredation temperature lower than unit
operating temperature * Being unable to save word processor documents
You say that you use the two terms Free Software and Open Source interchangibly. At the same time, depending on the year, month, phase of the moon, etc. you seem to declare your loyalty to one or another of these two camps almost to the exclusion of the other.
Can you please elaborate your views on this disconnect in people's mind on where exactly you stand on the issue?
Slashdot Mirror
No, but I think some albumns exist as whole peices for various reasons.
Beatles albums are good examples since they often have very distinct styles. Sgt Pepper vs, say Magical Mystery Tour.
One concern I have is that if people *never* buy or download an entire albumn, they'll never get the musician's message.
Does that mean I say to force them to listen to all of it. No, but it does mean it's something to consider when downloading only one song off an album.
Also, if you're only hearing (and therefore interested) in ONE song, then it's probably due to that song being on the radio. That's pretty limited and you might want to expand your horizons past what some large corporation tells you to like.
- Serge
I'd disagee in one area- and that's the album.
There are some music peices which are really part of a whole.
Examples that come to mind are Pink Floyd's Dark Side of the Moon, Kraftert's Man Machine or others.
An artist sometimes puts an albumn together not JUST to fill up a record or CDs worth of music but as a single thing. When you say you want to rip one part of it out, I think you're doing a disservice to the artist.
The issue for me is about control and morals. The moral issue is that of cultural exchange- and that's what music, books and movies are. The issue of control is the issue of 'to what length will these companies to go keep control over consumers'.
- Serge
This is a question that mainly applies to the US (the original poster was discussing a US company).
We in the US have a belief (which is built into our law) that a person is presumed innocent of wrongdoing until proven otherwise.
It applies to courts but, like many laws should be- the law is an extension of our collective views. We, as a society believe that someone is innocent until proven guilty.
I do not know Mexico well. I've never been there. People I do know who have lived there have told me that the justice system in Mexico is certainly different than in the US in a variety of ways.
If you have a different way and it works for you, then that's fine, but I believe that innocent until proven guilty is a good idea, and I think that, if it's not, it should be placed into an international agreement on how laws work (Human Rights perhaps).
I also believe that in some countries with Napoleonic law, those who accuse, if they're found to be wrong, are liable. That's not so in the US.
The point is the parent of my original post was saying that there was no legitimate reason to use a certain amount of bandwidth, so those who use more must therefore be doing something illegitimate (implying illegal).
I was arguing that this is not a good position to take- that innocence until proven guilty is a good thing, and I provided answers to "legitimate use" while also pointing out that "legitimate" vs "illegitimate" was never spelled out.
- Serge
This is a completely bass akward way of looking at it, which was the original poster's point.
Your line of arguing might extend to the pleding the fifth. "If he has nothing to hide, why doesn't he say anything?"
No, innocent until proven guilty means exactly that.
I don't see anyone arguing against bandwidth limits, rather that they need to be spelled out.
Examples of legitimate use might be playing online games, streaming online video, doing X over the network, etc.
Until you know *exactly* what is being done, you can't argue whether or not its legitimate (especially since you never define legitimate).
- Serge
Bandwidth bills on a static page are also trivial.
A well behaved program won't go GETs on every RSS page, but will do HEADS, compare them to what it already has, and decide from there to get or not get the new page.
A HEAD request is very small, and unless you're doing millions of them, this shouldn't be an issue.
- Serge
I don't think the original poster is anti-Unix, he just feels that the "GUI" aspect of X should encompas sound, just like X servers now handle keyboard and mouse events.
/dev/dsp.
Sound apps not needing to know the hardware is no different than X applications not needing to know the hardware when they make xlib calls.
Right now, every system that wants to provide network transparent sound has to reinvent the wheel since no one can agree on the "right" way to do it. Having one chosen as the "blessed" one by Xouert may end the argument.
As to requiring esd outside X, I don't see much of the point. I suppose you *could* do sound in consoles with something like emacs or screen, but that's highly unusual. Most people want sound events locally, so outside X you can send events right to
As to the issues of large monolithic code, I tend to agree, but think that the original poster's comment didn't necessarily imply code bloat, only a "blessed' X sound event system that comes with a new X server and makes it easier for application developers to write code that will run sound over the network (which will make projects like the Linux Terminal Server project easier).
- Serge
> Or they can ask me for help - not that I have tons > of time lately.
:)
Ah- that's why you haven't responded to the Slashdot interview from a few months back.
- Serge
The reason that no interoperabiluty exists is there's no incentive. Why not? Users aren't demanding it. Why aren't users damanding it? They're not demanding it becuase they don't percieve a need for it? Why not? Becuase in thier minds, they have a solution, which is the various programs such as Gaim.
Now:
1) If you'd read the first comment, I discuss the need for interoperability. The comment I made about Gaim was explaining why this level of interoperability hasn't been demanded.
2) You didn't bother following the link I provided which explains, in detail, my view on IM protocol interoperability. Had you bothered to follow the link and read, instead of just shooting off two posts, you would have seen that I share your view.
If you did follow the link, then you didn't read it, or didn't understand it. I can't imagine how, since I made my views very clear.
3) You seem to have difficulty understanding how I can explain a phenomenon in terms of someone's perception other than my own. I'm saying that the reason there hasn't been interoperability is that there hasn't been user demand. The way to change that is to either:
a) Change user demands
b) Provide some kind of incentive for change
c) Create a requirement for change (regulation, etc.)
But I wasn't addressing the mechanism for making change, rather adressing why it hasn't happened yet.
In fact, I was complaining that Gaim, while useful (I use it) do a disservice to people trying to provide interoperability between the IM networks.
- Serge
That's exactly what I said- that programs that let users use multiple programs reduce the need for interoperability in user's minds meaning there isn't the outcry there might otherwise be.
- Serge
I'm a bit unclear about the differences between SIP and XMPP and where Jabber, which could have been used as an interoperability standard, all fit together.
At the high end, these all seem like simple namespace issues and would map onto Jabber nicely. An AIM user, for example, could be user@aim but the end user doesn't need to know that, they could just be presented with an icon representing AOL or something.
The real issue is that there doesn't seem to be much in the way of motivation into making the IM systems compatible. Due to programs like Gaim and Trillian, less users are demanding it.
So the question for me is where's the money in it for AOL, Yahoo and Microsoft? AOL is clearly selling IM to companies like T-Mobile for phone connectivity, but I can't imagine that "pays the bills" for the millions of messages they route.
How is IMing profitable and what incentive do these companies really have in making interoperability a reality?
BTW- I see the real danger here in that IM is replacing other technology, such as TTY. I even wrote about this in my blog about a month ago: (yes shameless plug)
TTYs and the Internet
- Serge
Actually, my girlfriend is deaf and carries a cell phone. It works pretty well.
She has one of those T-Mobile Sidekick ones where you connect the Internet (web browser, AIM, and even an available SSH client).
It's actually quite useful. She can get her email, AIM and SMS messages in one place. There are even AIM -> TTY services so she can make "voice" calls on the road.
It's still a phone too. So if there's some sort of emergency and a hearing person is there- they could use it.
- SergeYou're making a number of assumptions which I don't think always hold true.
You say that you could do these measurements without the GA. That's true, but the GA serves two purposes:
1) To do the measruements of the combinations of
optimizations (which you mention)
2) To do the thing without human intervention
The second one is the important one. The fact is that few people are going to spend the time and energy to do the optimizations "right". A few hours or days at some point figuring out "pretty good" compiler options seems very reasonable though and I can see a lot of my own users doing it.
The issue of "is compiler optimization worth it?" depends on your problem domain. But if you're doing huge numbers of computations that may take hours to compete a single run, the answer is a resounding "yes" with a caviot.
The caviot is that code optimization must come first. We know the compile can't fix bad code, and good code optimization can improve things in ranges we'll never see from simple compiler optimizations.
Nonetheless, if a compiler optimization can shave, say 10 minutes from one run of a program that takes 3 hours to run- it's worth it. Why? Becuase that program is going to be run at least 1000 times. That ten minutes now becomes about six days of wall time. I can do a lot of processing in six days.
- Serge
I believe you're missing the point.
The idea isn't to attack at all, rather to reply as an interested customer.
The scenario is that you recieve a mail about getting, say pills that make your nostrils bigger. All spammers will need a way to ensure that you can make a purchase, and it's through that mechanism that you inquire for more information about nostril enhancement through magic pills.
If everyone who recieved an email did this, they would get thousands of requests.
If they only reply to a few of them then the company selling the pills looses sales.
So instead, they hire more staffers. When they do that, they are potentially eating into thier own profits.
Given sufficient numbers of respondants, this would make it suddenly unprofitable to mail everyone in the world, leading to an incentive to stop mass spamming.
That's the idea at least. There's no "attack" involved.
- Serge Wroclawski
You can watch it at the URLs above or maybe even buy a DVD from somewhere.
- Serge
Using the SHA1 mechanism, you can avoid using email addresses.
Yes it stinks but it's the only reasonable anti-spam measure that works right now.
- Serge
Furthermore, using PGP, trust values could be assigned to the information.
- Serge Wroclawski
Do you realize that your post is in response to a post by Ian or are you trolling?
No I don't agree, and here's why:
While people usually talk about X compatibility- that's not usually what they want. They want application compatibility. They want Mozilla, GNOME, Emacs to work on their computer *usually*.
Look at Opie and you realize that, given sufficient numbers of apps, you don't care as much about the libraries.
So what's important would be to port things like GTK and QT to the new target (however you choose to do that, be it at the X toolkit layer or the GDK layer, or whatever).
An example of the success of this method are the number of Free applications for Microsoft Windows and OS X.
While some apps talk to X directly, the number of users who will be effected is smaller.
If something else can provide the ability to use the same apps, allow same/similar features and provide some benefits, then I don't see why we can't give it a chance, even if it breaks one compatibility.
It's the infinite compatibility problem that has forced x86 down people's throats for so many years.
Sometimes you need to make a new protocol and provide a migration path.
- Serge Wroclawski
Then you use your revocation key (you did remember to store your recovation key didn't you?).
Now where's my ADAM replica?
And will it have the same features as the original, including:
* Power supply for computer located inside printer
* Being able to boot off tapes, BUT if you boot up
with tapes inside, the magnetic field will
erase them
* Chip degredation temperature lower than unit
operating temperature
* Being unable to save word processor documents
Linux runs on smart vibrators?
That's too much information for me...
This is an issue with the Zaurus 5500 as well. It lasts a little more than 2 hours.
Of course I power up an SD and CF card with it, but I consider that normal use.
It's made me remember/realize the one feature new PDAs need and that's wireless networking.
The Zaurus can use it's CF slot for that (and I do). It's too bad the new Agenda won't have one built-in
- Serge Wroclawski
I had an Agenda and I have a Zarus.
The key to the Zaurus are the two expansion slots, the keyboard, and the fact it runs OpenZaurus.
It looks like the new Softfield PDA will have the SD slot (less useful than CompactFlash) and MAY in time be able to run OpenZaurs.
If it does, it will be a useful device, but you can already pick up a Sharp Zaurus 5500 for less than $300 (I paid about $280 for mine)
- Serge Wroclawski
You can read the paper I believe they are making reference to in the article at http://www.avirubin.com/vote/
Bruce,
You say that you use the two terms Free Software and Open Source interchangibly. At the same time, depending on the year, month, phase of the moon, etc. you seem to declare your loyalty to one or another of these two camps almost to the exclusion of the other.
Can you please elaborate your views on this disconnect in people's mind on where exactly you stand on the issue?
- Serge