I think that at any time, in a country as large as China, there are tens, maybe even thousands of potential Mao Zendongs or Stalins.
But they don't get there, because there can be only one supreme ruler, and not all rulers will succeed at all times.
Hitler for instance came in at the right time. Germany was hummiliated, and he was the sort of guy that would make it great again. But his actions created such disgust, that if he was reborn right now, he couldn't get to power.
Rasputin started as a peasant and managed to get into the palace due to recommendation of the Tsaritsa's best friend.
And so on. Those famous people not only had the right genetics, but were born in the right place, often in the right family, knew the right people, and were around in the right circumstances.
Without that, the next potential Stalin might well be some bum living under a bridge. So long he remains there he's got no chance, no matter how close his genetics are.
Ok, a mirror won't work because it's not perfect and will quickly ablate, at which point the laser makes a hole in the missile even if it takes a second or two longer.
But what if you make a missile covered with retroreflectors that reflect that 90% or whatever a mirror can manage, but back at the laser itself?
Could an anti-laser missile be developed, which instead of a payload has several layers of retroreflectors to try to make the laser fry itself?
Again, see the post I referred to. Suing users of a library, users that may not bound by its license (because the GPL prior to version 3 is not a use license) seems like a nasty thing to do.
The GPL is quite clear in its intentions.
I consider exploiting loopholes to achieve a result that's obviously contrary to what was intended a nasty thing to do.
I think of people that do nasty things because they think they have the moral high ground as zealots.
I don't think this is resolvable.
The one using the GPL for their library thinks you're the asshole when you try to work around it. You as the user think the one trying to force you to comply with the GPL is the asshole.
Personally I side with the former, as they make their intention perfectly clear. If you don't like it, go write your own code.
Some authors of GPL code (myself included) actually prefer that the code is not used at all to having the license infringed. To me, having the GPL obeyed is more important than having my software used.
I can't say I feel sad about record labels. I don't want labels. I want to pay to the artist and artist only.
It's sad to see so many people touting stealing.
It's not stealing, it's copyright infringement. Stealing would imply that the one stolen from actually loses something, instead of not gaining.
I wonder how many would feel if people were stealing from them and taking pride in it.
You can't "steal" from me because I charge by the hour. Please take my work and copy it all you want.
And most of the arguments people here are making are false claims that they are making up simply so they can talk themselves out of feeling guilty for stealing from other people.
I'd be buying from the industry, if it was willing to sell me something. Say, I hear Amazon is selling music. Except I can't buy it because I'm not in the US. You could start by getting rid of that stupid state of affairs.
Honestly, I despise the current industry, which keeps trying to push DRM and fees on my hardware, and "3 strikes" laws on my internet connection. I really wish it died already. I'm willing to even accept a complete lack of new music in exchange.
I don't know about other people, but most of what I do with the internet has nothing to do with "content production". I would lose very little if free news became unavailable online. And not like they would, since even if somehow even AP feeds somehow became unavailable, people would still talk about what's going on.
Everybody pays for their internet connection. But that's something that doesn't go to content providers.
Then I can host my blog for free, or if I couldn't, pay $10/month for shared hosting.
Everybody pays for their net access, so it doesn't really count, and many people can pay their small web hosting fee without asking anything from anybody in return. So there are many people out there that are perfectly content with blogging or whatever without getting paid for it.
I sent an email to my ISP, and ESPN as well, saying that if I ever find my ISP is paying for ESPN or anything of that kind, I'll be switching to another one.
I DO NOT WANT this business model. It effectively means I'm paying for something, whether I want it or not, even if I find it a pointless waste of time. If this gets started, you'll end up with fees for 20 other things, whether you want them or not.
I want one thing only: an internet connection. If I need anything else I want to choose which of the available services to go with myself.
But what he/she says is that when you do your tax return online with TurboTax, they have access to your data. That _is_ reasonable.
No, it isn't. I don't see why something besides the IRS should have to know what's on my tax return.
That they have to see the data is simply a limitation of the web model.
The way I see it, things should work in one of two ways:
A. The IRS themselves hosts an online page for tax filings. The data is sent over SSL and goes to the IRS, and IRS only. B. If for some reason a third party is needed (and I don't see why it is), TurboTax should sell me a desktop application which talks directly to the IRS, and nothing else, over a secure connection.
Anything else is unacceptable. I do not like random third parties having access to my data.
"I know beating up people is wrong, but he was a huge jerk and deserved it" "I know stealing is wrong, but Walmart has a huge amount of money and can afford the loss" "People of $nationality/$ethnicity are really intrinsically inferior, so it's ok to treat them like crap" "I know I committed a crime, but the harm done wasn't that great, so I shouldn't have been punished so harshly"
People rarely admit outright that they did something wrong, with no ifs or buts. There's nearly always some reason they feel that justifies it, or at least makes it not so bad.
Most people don't think they're doing something wrong. They just were hanging out with their friends, or having fun, and don't deserve getting dragged through the courts for it. The ones who prosecuted them are just a bunch of jerks, and if they don't respect me why should I respect them?
Another possible factor is that when this happens once, the people involved probably start getting watched more and treated with more suspicion. If people are watching you more, you're more likely to get caught. And if everybody assumes you're going to steal, some people come to the conclusion they might as well go and do that, since they're being assumed to anyway.
I don't care how hard it is, the result is not creative because you're not trying to express yourself, you're just trying to work around the limitation of that the painting can't be taken off the fall and fed into a scanner.
The desired result is an accurate reproduction of the painting, and that's precisely why it's not creative. How hard you have to work to get there is completely irrelevant, the fact is that the end result looks as much as the painting as possible, and any other photographer given the same task would attempt to produce something that looked similar.
Otherwise I could easily obtain copyright for a copy of anybody's work, I'd just have to do it by a laborious method.
And the ozone holes that would cause everyone to get skin cancer
Duh, we're not getting skin cancer because we actually fixed the problem.
1. We discovered a problem: The ozone hole. We found it before it got large enough to start causing really big problems. 2. Predictions were made of what would happen if it continued getting bigger, and the potential consequences were unpleasant. 3. Actions were taken to correct the situation. 4. That made things a lot better. It's not been eliminated, but at least it's on the way to recovery (which will be in ~2060)
Ahh, I see, so you hang out with the housekeeping staff and fully trust them too. You know, the ones who do the shitty job, are thoroughly underpaid but are easily smart enough to realize that somebody "out there" might find confidential information on your system very, very valuable? Same with the building owners your company leases to, right?
Not every company out there is a huge multinational where every worker is a random tiny cog in the machine. I worked at small companies where everybody knows each other and often talks to the boss on a first name basis. People do hang out with the housekeeping staff at those.
Certainly the risk of what you say always exists. But the probability of a ssh user account with a password of "password" getting broken into is about 100%, probably within a day. Your scenario while possible is much less likely.
Sure, the smart card system and password were essentially unbreakable, but they didn't need to be. Smart card resets, password resets, and sticky notes with passwords and pins were so common it was easilly the least secure system I've ever had the privilage of working with. It also severely hampered productivity.
Draconian security practices have significant downsides. At one company I worked at, we had an internal program with user accounts that would automatically log out after a period of inactivity. So we customized the logout delay per computer. The ones that were in offices had a long one. The ones in the warehouse had a short one, because people used the nearest computer. People also had a reason to consistently log out when they'd leave, as program usage was logged, and if you left an active session, you'd be the one blamed for anything that went wrong. IMO it worked very well. Don't annoy people more than really needed, give them a good reason to comply with the security, and it'll work.
Well, for instance, there's lots and lots of VB6 code out there that became obsolete when MS dropped it. The.NET version is different enough that large apps can't be translated and need to be rewritten.
Actually VB6 code is still getting written even today, but it's a dangerous proposition. There's no guarantee it'll run on future Windows versions. Especially there's no guarantee that the OCX you need will work on future Windows version.
COBOL is an exception because it was used in important systems developed entirely in-house with full source available.
But VB6 isn't like that. A vast majority of programs need some OCX or another that performs a crucial task. And the VB code itself is just glue (something every VB book likes to point out). Many VB apps are completely uninteresting and say, use an OCX to interface with some specialized piece of hardware, another OCX to present data (some fancy grid control for instance), and a database. If any of that stops working, you're screwed. And chances are those companies that made that stuff are now gone or uninterested in maintaining it.
Compare for instance, Perl or C. Perl isn't that popular anymore, but it's still actively worked on. Even if development stopped, the source would still be there.
Let me tell you a little secret. Proprietary software developers are just as big assholes.
Sometimes even worse, because sociopathic bosses and the economy make their contribution as well.
In the closed source world you almost never have complete control of your project. What happens if the OS, language, or vital module of your project is dropped by the maker? If you work on.NET for instance, then one day it could be abandoned, to be replaced by something newer and shinier. In comparison, C and Perl are ancient and aren't going anywhere.
Users have to be able to remember their passwords in order for this security to be of any use. Push them beyond that ability, and you're actively making the situation worse.
No, not really.
If people at your office can be trusted, you don't really take a huge risk by having a postit with the password. The complicated password, however, makes it much harder to brute force from the outside, or to brute force a compromised hashed password DB.
A few years back somebody managed to grab the Second Life password database. My password was something like "KKVRJTVRq8KI1eVL", so I could be quite sure that whoever got the DB would have first instantly cracked the several thousands of "password" and "secret" passwords in the DB (there was about a million accounts at the time), and mine would be way down the list, so I could reasonably expect it would resist attempts at cracking while I was getting around to changing it. If I had that password stuck to my monitor it wouldn't have changed any of this in the slightest.
If your password is a unix account password that's accessible through ssh and present on an externally accessible server, you can bet that if your choice was "password" or "secret" your account will be broken into soon enough. There's quite a lot of machines out there trying that sort of thing against each ssh server.
I don't think the sunk costs fallacy applies to the ISS.
If for instance, a software migration is being worked on that's clearly not going to be successful, then throwing more money on it is clearly pointless. Since the old system can be kept, the decision is "waste more money on the non-working new system, and at some later point return to the old one" vs "return to the other one". Keeping spending money on the migration is clearly a complete waste and achieves nothing useful.
But the ISS isn't like that. It's already in orbit, and it's already producing results. If anything useful at all still gets done on it, then deorbiting it will have negative consequences, and if later they want to change their mind they can't just go and put it back in orbit. They'll have to make a new one, or go without whatever research it could have enabled. And making a new one will almost certainly cost a lot more than to keep the current one working.
I'm not against the concept per se, but I thought I'll mention I've never, ever paid for an utility, and probably never will.
In Linux, WinZip-like commerical programs are pretty much inexistent, as all this functionality is available for free.
In Windows, I just got so used to the nagging that I ignored it for years, until one day I thought to visit Wikipedia's Windows compression software page, and found 7zip. Now that's one of the first things I install on a new box, and WinZip and WinRAR are nowhere to be seen.
In fact, one of the reasons I use Linux full time now is that it doesn't have utility software in the Windows style. It has programs that perform the same tasks, but they don't try to appear unique. On Linux most of these programs have their own non-standard appearance and whine to be paid for or registered. In Linux they integrate with the rest of the system and just do what they're supposed to.
That stuff turns off any normal buyers. It might work if you're doing something uber-specialized you sell to large companies, but normal people stay far away from anything like that. Just for a start, how would I sign this contract by hand while being in another country? Do you really expect somebody to print and mail a contract, and wait for a week or two until it gets to the destination?
IMO, for a program destined to the general public abstain from any of the following:
* Required registration * Required email address * Price not listed on the website (since that usually means "an arm, a leg, and a kidney", or "as much as we can get you to pay") * Dongles and other intrusive methods of control * Lack of specific information on what exactly the program does
First of all, there are a lot of people wanting to play mp3, and a very few trying to compose. Especially on Linux.
Second, somebody composing music is unlikely to be using onboard audio anyway. They probably have a fancy soundcard of some sort.
Third, if good latency comes at the cost of no mixing, it's quite possible that this composing software won't be the one that gets to use the soundcard, and it doesn't matter what the latency is when you're not getting sound at all.
There is. ALSA's dmix has been enabled by default for a long time, years. Have you even tried Linux? I can't remember the last time I had to 'configure' sound on Linux. Insert sound card, mixer shows up, play sounds. From the ALSA wiki: "NOTE: For ALSA 1.0.9rc2 and higher you don't need to setup dmix. Dmix is enabled as default for soundcards which don't support hw mixing."
Yeah, it's supposed to work, but for some reason for me it doesn't.
And have you looked at that page? It's full of listings of arcane incantations. Really, I just want the darn audio to always get mixed, without having to get a degree in audio engineering to understand what's going on there.
If you bothered to try, you'd find that it does.
See the dmix page, which says "Normally (without hardware mixing) you cannot use/dev/dsp multiple times directly."
So it seems that if you have onboard audio, and want to have more than one app use/dev/dsp, you're out of luck.
No, there really shouldn't. Mixing can incur some pretty serious latency unless all applications use the same buffer size and data format. Basically, real time/low latency audio apps will need exclusive access.
No, it really should. Because most people right now are using a sound card that doesn't mix in hardware. So when the sound system for some stupid reason refuses to mix, what happens is that users don't get sound because some other program is using the device already.
Again, that it works comes first. Latency is secondary.
And that is that ALSA's way of handling mixing is completely moronic.
As an user, I care about hearing sound first of all. Sound quality (no pops or crackles) comes second, latency comes third.
There should always be sound mixing, with no ifs, buts, exceptions, or configuration required. It should be there by default for anything that tries to play sound, whether through ALSA or the OSS backwards compatibility.
The result of this nonsense is that crap like pulseaudio continues to exist, which is CPU hungry, often skips, fails to work with some programs and crashes frequently (what the hell is up with that?).
Is there any document out there which explains why/dev/dsp doesn't get mixing with ALSA? And why nobody tried to patch that yet?
Slashdot Mirror
I think that at any time, in a country as large as China, there are tens, maybe even thousands of potential Mao Zendongs or Stalins.
But they don't get there, because there can be only one supreme ruler, and not all rulers will succeed at all times.
Hitler for instance came in at the right time. Germany was hummiliated, and he was the sort of guy that would make it great again. But his actions created such disgust, that if he was reborn right now, he couldn't get to power.
Rasputin started as a peasant and managed to get into the palace due to recommendation of the Tsaritsa's best friend.
And so on. Those famous people not only had the right genetics, but were born in the right place, often in the right family, knew the right people, and were around in the right circumstances.
Without that, the next potential Stalin might well be some bum living under a bridge. So long he remains there he's got no chance, no matter how close his genetics are.
Ok, a mirror won't work because it's not perfect and will quickly ablate, at which point the laser makes a hole in the missile even if it takes a second or two longer.
But what if you make a missile covered with retroreflectors that reflect that 90% or whatever a mirror can manage, but back at the laser itself?
Could an anti-laser missile be developed, which instead of a payload has several layers of retroreflectors to try to make the laser fry itself?
The GPL is quite clear in its intentions.
I consider exploiting loopholes to achieve a result that's obviously contrary to what was intended a nasty thing to do.
I don't think this is resolvable.
The one using the GPL for their library thinks you're the asshole when you try to work around it.
You as the user think the one trying to force you to comply with the GPL is the asshole.
Personally I side with the former, as they make their intention perfectly clear. If you don't like it, go write your own code.
Some authors of GPL code (myself included) actually prefer that the code is not used at all to having the license infringed. To me, having the GPL obeyed is more important than having my software used.
Short term. Not long term. Crap accumulates. At some point you'll have to get rid of it.
Would have making the chernobyl reactor with a containment vessel cost more than what ended up being spent on cleanup and abandoning the city?
I can't say I feel sad about record labels. I don't want labels. I want to pay to the artist and artist only.
It's not stealing, it's copyright infringement. Stealing would imply that the one stolen from actually loses something, instead of not gaining.
You can't "steal" from me because I charge by the hour. Please take my work and copy it all you want.
I'd be buying from the industry, if it was willing to sell me something. Say, I hear Amazon is selling music. Except I can't buy it because I'm not in the US. You could start by getting rid of that stupid state of affairs.
Honestly, I despise the current industry, which keeps trying to push DRM and fees on my hardware, and "3 strikes" laws on my internet connection. I really wish it died already. I'm willing to even accept a complete lack of new music in exchange.
That's not enough either, in some languages like Russian the sentence changes depending on the gender.
So for instance a notification like "$avatar logged in" needs a male and a female version.
To make it more fun: "shark" is a feminine noun in russian, so if the nickname was translated as well it'd imply you're female even if you're not.
I'm not paying a cent for that.
I don't know about other people, but most of what I do with the internet has nothing to do with "content production". I would lose very little if free news became unavailable online. And not like they would, since even if somehow even AP feeds somehow became unavailable, people would still talk about what's going on.
It's however, not really expensive.
Everybody pays for their internet connection. But that's something that doesn't go to content providers.
Then I can host my blog for free, or if I couldn't, pay $10/month for shared hosting.
Everybody pays for their net access, so it doesn't really count, and many people can pay their small web hosting fee without asking anything from anybody in return. So there are many people out there that are perfectly content with blogging or whatever without getting paid for it.
I sent an email to my ISP, and ESPN as well, saying that if I ever find my ISP is paying for ESPN or anything of that kind, I'll be switching to another one.
I DO NOT WANT this business model. It effectively means I'm paying for something, whether I want it or not, even if I find it a pointless waste of time. If this gets started, you'll end up with fees for 20 other things, whether you want them or not.
I want one thing only: an internet connection. If I need anything else I want to choose which of the available services to go with myself.
No, it isn't. I don't see why something besides the IRS should have to know what's on my tax return.
That they have to see the data is simply a limitation of the web model.
The way I see it, things should work in one of two ways:
A. The IRS themselves hosts an online page for tax filings. The data is sent over SSL and goes to the IRS, and IRS only.
B. If for some reason a third party is needed (and I don't see why it is), TurboTax should sell me a desktop application which talks directly to the IRS, and nothing else, over a secure connection.
Anything else is unacceptable. I do not like random third parties having access to my data.
They know but they rationalize it.
"I know beating up people is wrong, but he was a huge jerk and deserved it"
"I know stealing is wrong, but Walmart has a huge amount of money and can afford the loss"
"People of $nationality/$ethnicity are really intrinsically inferior, so it's ok to treat them like crap"
"I know I committed a crime, but the harm done wasn't that great, so I shouldn't have been punished so harshly"
People rarely admit outright that they did something wrong, with no ifs or buts. There's nearly always some reason they feel that justifies it, or at least makes it not so bad.
Most people don't think they're doing something wrong. They just were hanging out with their friends, or having fun, and don't deserve getting dragged through the courts for it. The ones who prosecuted them are just a bunch of jerks, and if they don't respect me why should I respect them?
Another possible factor is that when this happens once, the people involved probably start getting watched more and treated with more suspicion. If people are watching you more, you're more likely to get caught. And if everybody assumes you're going to steal, some people come to the conclusion they might as well go and do that, since they're being assumed to anyway.
I disagree.
I don't care how hard it is, the result is not creative because you're not trying to express yourself, you're just trying to work around the limitation of that the painting can't be taken off the fall and fed into a scanner.
The desired result is an accurate reproduction of the painting, and that's precisely why it's not creative. How hard you have to work to get there is completely irrelevant, the fact is that the end result looks as much as the painting as possible, and any other photographer given the same task would attempt to produce something that looked similar.
Otherwise I could easily obtain copyright for a copy of anybody's work, I'd just have to do it by a laborious method.
Duh, we're not getting skin cancer because we actually fixed the problem.
1. We discovered a problem: The ozone hole. We found it before it got large enough to start causing really big problems.
2. Predictions were made of what would happen if it continued getting bigger, and the potential consequences were unpleasant.
3. Actions were taken to correct the situation.
4. That made things a lot better. It's not been eliminated, but at least it's on the way to recovery (which will be in ~2060)
Not every company out there is a huge multinational where every worker is a random tiny cog in the machine. I worked at small companies where everybody knows each other and often talks to the boss on a first name basis. People do hang out with the housekeeping staff at those.
Certainly the risk of what you say always exists. But the probability of a ssh user account with a password of "password" getting broken into is about 100%, probably within a day. Your scenario while possible is much less likely.
Draconian security practices have significant downsides. At one company I worked at, we had an internal program with user accounts that would automatically log out after a period of inactivity. So we customized the logout delay per computer. The ones that were in offices had a long one. The ones in the warehouse had a short one, because people used the nearest computer. People also had a reason to consistently log out when they'd leave, as program usage was logged, and if you left an active session, you'd be the one blamed for anything that went wrong. IMO it worked very well. Don't annoy people more than really needed, give them a good reason to comply with the security, and it'll work.
Well, for instance, there's lots and lots of VB6 code out there that became obsolete when MS dropped it. The .NET version is different enough that large apps can't be translated and need to be rewritten.
Actually VB6 code is still getting written even today, but it's a dangerous proposition. There's no guarantee it'll run on future Windows versions. Especially there's no guarantee that the OCX you need will work on future Windows version.
COBOL is an exception because it was used in important systems developed entirely in-house with full source available.
But VB6 isn't like that. A vast majority of programs need some OCX or another that performs a crucial task. And the VB code itself is just glue (something every VB book likes to point out). Many VB apps are completely uninteresting and say, use an OCX to interface with some specialized piece of hardware, another OCX to present data (some fancy grid control for instance), and a database. If any of that stops working, you're screwed. And chances are those companies that made that stuff are now gone or uninterested in maintaining it.
Compare for instance, Perl or C. Perl isn't that popular anymore, but it's still actively worked on. Even if development stopped, the source would still be there.
Let me tell you a little secret. Proprietary software developers are just as big assholes.
Sometimes even worse, because sociopathic bosses and the economy make their contribution as well.
In the closed source world you almost never have complete control of your project. What happens if the OS, language, or vital module of your project is dropped by the maker? If you work on .NET for instance, then one day it could be abandoned, to be replaced by something newer and shinier. In comparison, C and Perl are ancient and aren't going anywhere.
No, not really.
If people at your office can be trusted, you don't really take a huge risk by having a postit with the password. The complicated password, however, makes it much harder to brute force from the outside, or to brute force a compromised hashed password DB.
A few years back somebody managed to grab the Second Life password database. My password was something like "KKVRJTVRq8KI1eVL", so I could be quite sure that whoever got the DB would have first instantly cracked the several thousands of "password" and "secret" passwords in the DB (there was about a million accounts at the time), and mine would be way down the list, so I could reasonably expect it would resist attempts at cracking while I was getting around to changing it. If I had that password stuck to my monitor it wouldn't have changed any of this in the slightest.
If your password is a unix account password that's accessible through ssh and present on an externally accessible server, you can bet that if your choice was "password" or "secret" your account will be broken into soon enough. There's quite a lot of machines out there trying that sort of thing against each ssh server.
I don't think the sunk costs fallacy applies to the ISS.
If for instance, a software migration is being worked on that's clearly not going to be successful, then throwing more money on it is clearly pointless. Since the old system can be kept, the decision is "waste more money on the non-working new system, and at some later point return to the old one" vs "return to the other one". Keeping spending money on the migration is clearly a complete waste and achieves nothing useful.
But the ISS isn't like that. It's already in orbit, and it's already producing results. If anything useful at all still gets done on it, then deorbiting it will have negative consequences, and if later they want to change their mind they can't just go and put it back in orbit. They'll have to make a new one, or go without whatever research it could have enabled. And making a new one will almost certainly cost a lot more than to keep the current one working.
I'm not against the concept per se, but I thought I'll mention I've never, ever paid for an utility, and probably never will.
In Linux, WinZip-like commerical programs are pretty much inexistent, as all this functionality is available for free.
In Windows, I just got so used to the nagging that I ignored it for years, until one day I thought to visit Wikipedia's Windows compression software page, and found 7zip. Now that's one of the first things I install on a new box, and WinZip and WinRAR are nowhere to be seen.
In fact, one of the reasons I use Linux full time now is that it doesn't have utility software in the Windows style. It has programs that perform the same tasks, but they don't try to appear unique. On Linux most of these programs have their own non-standard appearance and whine to be paid for or registered. In Linux they integrate with the rest of the system and just do what they're supposed to.
Are you crazy?
That stuff turns off any normal buyers. It might work if you're doing something uber-specialized you sell to large companies, but normal people stay far away from anything like that. Just for a start, how would I sign this contract by hand while being in another country? Do you really expect somebody to print and mail a contract, and wait for a week or two until it gets to the destination?
IMO, for a program destined to the general public abstain from any of the following:
* Required registration
* Required email address
* Price not listed on the website (since that usually means "an arm, a leg, and a kidney", or "as much as we can get you to pay")
* Dongles and other intrusive methods of control
* Lack of specific information on what exactly the program does
First of all, there are a lot of people wanting to play mp3, and a very few trying to compose. Especially on Linux.
Second, somebody composing music is unlikely to be using onboard audio anyway. They probably have a fancy soundcard of some sort.
Third, if good latency comes at the cost of no mixing, it's quite possible that this composing software won't be the one that gets to use the soundcard, and it doesn't matter what the latency is when you're not getting sound at all.
Yeah, it's supposed to work, but for some reason for me it doesn't.
And have you looked at that page? It's full of listings of arcane incantations. Really, I just want the darn audio to always get mixed, without having to get a degree in audio engineering to understand what's going on there.
See the dmix page, which says "Normally (without hardware mixing) you cannot use /dev/dsp multiple times directly."
So it seems that if you have onboard audio, and want to have more than one app use /dev/dsp, you're out of luck.
No, it really should. Because most people right now are using a sound card that doesn't mix in hardware. So when the sound system for some stupid reason refuses to mix, what happens is that users don't get sound because some other program is using the device already.
Again, that it works comes first. Latency is secondary.
And that is that ALSA's way of handling mixing is completely moronic.
As an user, I care about hearing sound first of all. Sound quality (no pops or crackles) comes second, latency comes third.
There should always be sound mixing, with no ifs, buts, exceptions, or configuration required. It should be there by default for anything that tries to play sound, whether through ALSA or the OSS backwards compatibility.
The result of this nonsense is that crap like pulseaudio continues to exist, which is CPU hungry, often skips, fails to work with some programs and crashes frequently (what the hell is up with that?).
Is there any document out there which explains why /dev/dsp doesn't get mixing with ALSA? And why nobody tried to patch that yet?