First, I think the spray on cells are a great idea. They could use CFC propellant as a method to improve gain as they are deployed.
But I jest.
More seriously, has anyone looked at the potential for global cooling of converting significant amounts of solar energy to mechanical? Sure some of the input to machines is returned to heat, but any movement is solar heat the we'll never get back.
I'm serious.
And let's not forget that Earth Day was organized to fight global cooling . ..
I don't think it was XML at the time. It's been about a year, but as I remember it it was a shitload of extra Apache config stuff.
Yeah, it can't have been XML because it was full of impossible to follow comments surrounded by hashes. You know the kind:
########## # comment # ##########
I had a better example, but the lamness filter wouldn't let it by. Anyway, it had these huge comment blocks that really didn't shed any light on things. I'm sure they were really meaningful to people who already knew what the stuff did, but that doesn't help me much, does it?
Anyway, I hope it is better now. Maybe a lot of guys like me kept harassing them and they made the install a little nicer.
I'm totally with you on the "one man's pleasure" thing. As I said in another post in this thread, Java stuff always gives me the heebie-jeebies, but I know lots of guys who are smarter than me who love it. Go figure.
I was not fortunate enough to be exposed to *NUX until about four years ago, but I did take Applesoft Basic in the eighth grade. And I happen to have just polished off a little bash script today. That's not the issue.
The issue is that I think it is a fundamentally bad idea to have a server service depend on a special environment variable. IMO it is too fragile a way of doing things. Think "This f-ing thing starts fine from the command line, but fails to start by the init script."
You are, of course, free to disagree with me, but I think your "these kids today" attitude reflects on you far more than it does on me.
Oh, and I was in high school by the time I had any sort of regular access to computer equipped with a mouse.
So, I stand behind my statement that Tomcat (actually, Jakarta as I understand it) was unnecessarily difficult to install at the time I first attempted it (about a year ago, it may be a snap now). I'm not saying that I need a "wizard" to wipe my backside during an install, but Jakarata was too far the other way IMO.
Just as a note for perspective, it took me about four hours and two Mt. Dews to install Apache with SSL and all the trimmings the first time. It took me the better part of three days and a case of the Dew to install Jakarta the first time.
Finally, maybe I am just a dumb kid, but the last paragraph of your reply didn't make any sense to me at all. I'd really like to know what you meant and I hope you will offer a bit of an explanation.
Well, I think it is more that different people see the world differently and that "the Java people" and I are 180 degrees out of phase.
Everything with Java freaks me out for some reason. Every time I delve into something Java related I fell inundated with so much jargon that it is easier to just not use it.
Catalina: Is that the _OLD_ version of Tomcat, or one of the componets, you know I really can't remember.
This is really funny. My point was that the Jakarta stuff has way to many shifting labels for a freaking servelet engine and your response is basically "no it doesn't, it's quite simple . . . now how the hell does it go again?" (BTW, I take it that it is the new new servelet container . . . whatever that means.)
Does lack of knollage mean lack of hillyness?
Why do you feel you need to resort to name calling?
First, don't overlook webapps. I used Squirrelmail for all my email for quite a while. It was great to be able to get all my mail (and archives) securely from ANYWHERE that had an SSL capable browser and an internet connection. PHPGroupWare is another great example. Webapps are BY FAR the most flexable from the client perspective.
Beyond this there are two more strong options.
As about fifty people have said, VNC. VNC has the advantage of working on many platforms and being able to re-direct an entire desktop. VNC "becomes" a webapp via its ability to provide your desktop via any Java capable browser. This is a strong option for your situation.
Xfree86 is a good option for serving up individual apps, and is really handy when paired with ssh (-X option). This option is better suited to a large number of fixed clients (i.e. workstations) using a small number programs (i.e. X clients (geez I hate the X terminology.)) regularly. Not so great for your situation.
Good Luck!
-Peter
Re:Sorry, I don't believe in paying for software.
on
How to Save PGP
·
· Score: 2
How does selling something along with something you get for free drive down your margins?
Let's say that Red Hat and MS each sell an OS for $100. Each expects to spend $50 supporting it. RH has $15/copy (at expected distribution volume) invested in development, and MS has $30, since the write the whole thing from scratch.
Who has the larger margin?
Now, these are all made-up numbers, but I think that they are useful for illustration purposes. Can you make up a set of reasonable numbers to illustrate how bundling support and distribution of software that you largely get for free hurts your margin?
The way I explain that RH isn't making money hand over fist, but MS is is simple. Volume. I think that the reality is that RH spends something on the order of 1/10 what MS does on development, and has something like 1/1000 the (full price paid) distribution. So the numbers are more like 100/50/150 vs. 100/50/30.
OTOH, your $120 billion figure, if I'm not mistaken, is their peak market cap. Which is bullshit. Market cap is literally meaningless. It has nothing to do with actual money. Not money that they have, have spent, people have spent on them. Nothing.
That statement, combined with your statement that adding value by packaging and selling something that you get for free hurts the economy makes me question your grasp of economics.
Now, I know nothing about accounting, but my understanding of the English language leads me to believe that they had a quarterly loss of 17M in 2000 (and a somewhat higher loss in the same quarter of 2001). Which leads me to question your interpretation of any facts.
Finally, who said anything about "open source?" I'm talking about Free Software.
-Peter
Re:Address lines and outdated but orphan BIOS
on
Low-end Laptops?
·
· Score: 2
Isn't 8gb enough for DOS? Surely it is enough to boot.
Any reputable OS only uses the BIOS to get a minimal system loaded to switch to protected mode disk access. (This is all, of course, PC specific.)
As long as the bios can see the first 15 megs, I'm set!
Re:Sorry, I don't believe in paying for software.
on
How to Save PGP
·
· Score: 2
It is abundantly clear that you didn't read the page I linked to.
Most of what you said is based on the exact confusion arising from the phrase "selling software" (and variants you used like "selling App" or "selling gpl software" or "selling code") that is explained in the page I linked to.
So, since you don't care to read that article, let me establish some vocabulary.
If "selling software" is to have any consistent meaning it must be selling the copyrights to a piece of software. Such as when Corel bought WordPerfect. This clearly is not the topic of the discussion.
Now we come to what you are really talking about, which is selling software licenses. When you "buy software" (really "buy a license") you never get anything but the use of the software IAW the license terms. If you actually "bought windows" why may you not sell it? I don't mean en masse, just the CD you bought? Because you didn't buy anything but a license.
Finally we have distributing software. Which is what I was talking about. Wal-Mart makes money by distributing both proprietary and Free Software. It doesn't make a difference to them. Redhat sits on the shelf right next to XP. See my other reply in this thread for more examples of people making money by distributing free software.
Finally, note that if we can agree to the terminology above then you were more correct than you know, since there is there is no license for use of Free Software distributed under the terms of the GPL to sell.
To be totally clear about what I just said; the GPL isn't a "software license" in the sense that many people think it is. The GPL is a software distribution license. It makes no demands on the user (unlike a EULA) except that they may not sue if they don't like the way the program works, or fails to work.
So again, there is no software license to sell. Thus, you are correct that selling licenses for unlicensed software is not a promising business model. That, however, has nothing to do with my original post.
-Peter
Re:Sorry, I don't believe in paying for software.
on
How to Save PGP
·
· Score: 2
I disagree. Redhat charges a premium for priority FTP access to software which can be freely distributed. The FSF itself was formed with money made by selling GNU on tape.
It is true that Free Software does not have the "advantage" of artificial scarcity that proprietary software has. In spite of this, both Cheap Bytes and KRUD both operate in the black AFAIK.
If we expand beyond simple distribution there are additional ways to actually make money by distributing Free Software that have been demonstrated in the real world. Redhat turns a profit, largely by bundling service with distribution. Several of the PHPGroupWare guys support themselves by supporting PHPGroupWare when they aren't hacking on it. Other value-adds exist, such as IBM bundling Free Software with hardware.
But, I suppose it is true that you aren't going to make yourself rich by downloading Free Software on your cablemodem and mailing out burned CDs.
-Peter
Re:Sorry, I don't believe in paying for software.
on
How to Save PGP
·
· Score: 3, Informative
Your reply might be funny if it weren't 180 degrees out of phase with the real universe.
To see what RMS actually thinks about this subject see http://www.gnu.org/philosophy/selling.html.
From that page:
Many people believe that the spirit of the GNU project is that you should not charge money for distributing copies of software, or that you should charge as little as possible -- just enough to cover the cost.
Actually we encourage people who redistribute free software to charge as much as they wish or can. If this seems surprising to you, please read on.
Then again, when has an AC let reality interfere with the contents of his posts?
Many of you may be aware that during the production of Return of the Jedi the title Revenge of the Jedi was used to mislead knock-off merchandise manufactures.
Lucas has pulled this old trick out of the bag. Let me be the first to announce that the actual title of Star Wars: Episode II is to be Attack of the Marketing Drones.
You've almost got it right, except that the device should definitely run over 802.11b wireless. Particularly as you have about as much chance of powering the device over 802.11b as you do powering it over ethernet...
Gee, an AC who doesn't know what he is talking about. I'm so shocked.
First, wireless would double (or more) the cost of the device that I am imagining.
Hyperlink Power-over-Ethernet (PoE) products provide PoE capabilities for access points, access servers, outdoor routers, and other suitable devices.
Perhaps you should spend thirty seconds on google before you shoot your mouth off next time?
But, seriously... It's probably as easy or possibly easier to decode MP3s than it is to decode WAV files. My understanding is, this is exactly what the sliMP3 player is...
So you are saying that it is easier to do decode the MP3 and then do a digital to analog conversion than it is to just do the D/A conversion? You can't possibly defend that statement.
A sound card that uses ethernet instead of pci/isa, what a great way to say this in less than 80 words!
The reason I bring up UNIX is that adding a codec would be as simple as "$ foo_codec_play some_song.foo >/dev/hi-fi" None of this wait for the vendor/flash the device/give up on the vendor and rip out the OS and put in Linux crap.
Seriously, if you skip the remote and just use raw ethernet frames I think this could be done with no brains at all in the "ethernet based soundcard".
I don't know what "infiniband" is, but as I figure it a 1x cdrom (which is where the 44.1kHz PCM thing came from) xfers at 150kB/s and 10meg ethernet is 1280kB/s ((10mb) * (1024k/m) / (8b/B)). So this thing would use 12% of 10meg and 1.2% of 100meg ethernet, not accounting for overhead.
I'd prefer setting the MAC by dips. Then if it does us IP, just set up MAC affinity in DHCP. Anyway, like I said, the driver could handle all this. The more I think about it the more I think IP is overkill. Being limited to being on the same ethernet segment as the device is not a big limit. And you could always add a proxy system (if you want to play on a device on a seperate segment you proxy through a machine on the same segment as the device). Again, make the hardware dumb and the software smart.
Oh, and I updated my email address. Please mail me if you want to persue this. I'll buy all the parts if you build two!
One of the things that makes the PC so powerful is the ability to add "dumb" peripherals that add major functionality by leveraging the generality of the PC.
In this vein, why not have a simple peripheral that:
1. Gets power from ethernet 2. Talks IP or some simpler protocol 3. Has a DAC that is able to convert one uncompressed digital format (say, 44.1kHz PCM) to high quality stereo analog. 4. Optionally has an IR or RF remote that sends simple events back to the PC.
Here is the real beauty, especially on a UNIX system: adding codecs is trivial. Really adding any functionality is trivial. Want a random mode, code it up. Need more space, add it to the PC, want to be able to play all songs that match a regexp, code it up. The real slick thing would be that it could play audio that was encoded in a way never even dreamed of by the hardware OR the software designers through UNIX pipes. There could also be a driver that allows it to appear as a simple/dev (maybe/dev/hi-fi?) device, rather than having to address it by IP.
I suspect that any 2nd year EE student could design and build this thing with a DAC scavenged out of a CD player that suffered mechanical failure.
You could even use this to replace/dev/dsp for "main system sounds." Allowing high quality, hack-free use of the stereo for general PC audio.
Some ideas about obvious problems:
1. Give them all the same MAC, but make the last byte configurable by dip switches (to allow multiple devices on the same ethernet segment). All config could then be handled automagically by the driver.
2. Design it to take power from ethernet OR an A/C adapter.
3. It will need some buffer. A few megs should be overkill.
4. In case this isn't clear, it would be a totally "push" device. It would sit there and just dump whatever came over the wire with its IP/MAC into a buffer, which dumps right into the DAC.
Advantages:
1. Never fight with Linux sound drivers again. 2. All DA conversions prior to the box itself are in software, and therefore don't pick up "chassis noise" like a soundcard does. 3. "infinite" upgradeability through PC software. 4. Total platform independence. If it talks ethernet (and IP?), it can control the device. 5. "Patent-proof" I am pretty sure you can't patent something that is just like something else, only with components/removed/.
The only real question in my mind is wether it is worth it to implement IP on the thing, or if it would be better to just use a very simple, non-routeable protocol on top of ethernet.
Does anyone know of anything like this? By like this I mean something that is simply a DAC with ethernet. There a a bunch of devices out there that decode MP3s, etc. That's not what I am looking for.
Any EEs out there want to take a whack at it?
And, to come back around to being on topic, he would be able to "play" his shoutcast stream locally, and push it out to his stereo with this thing, if it existed.
And I do happen to agree that seniority and time spent with a company should be a factor in determining who gets promoted.
Ugh. It should be a (very small) factor. It does breed discontent when some "wiz kid" goes straight to the top without "paying dues."
OTOH I have had to work sorta-kinda under some real dubmasses who just occupied a chair so long without doing anything to get fired that they got promoted. That wasn't good for the company or my blood pressure.
Okay, so what you meant was "VNC does not ADD multiuser GUI functionality to Windows." which is substantively different from what you said.
Now, will you please explain what the hell you mean by a "samba box?" AFAIK SAMBA is an SMB server/client package for UNIX. Do you mean to say that there is a UNIX port of MSTS? If so, I'd love a link.
Finally, you really need to work on your communication skills. You clearly know what you are trying to say, but it is coming out as half-formed nonsense.
The idea was: never again should a democracy change into a racist totalitarian state, and it's worthwhile to give up that bit of freedom to prevent this from happening again.
Which is to say "we must give up our liberty in order to save it."
"Thanks for throwing out the bath water, hon. Where's the baby?"
My real point is that you don't seem to understand the implication of what you are saying. If you give up the right/power/control to decide what you say it doesn't just dissapear . . . that control goes to the government
History clearly indicates that giving a government power is not a way to ensure that your freedom is protected.
Water is not actually conductive. Water with disolved solids, noteably salt, is quite conductive. It doesn't have to be cloudy or "salty" to be salty enough to be pretty conductive.
mail: if you are prone to outages, reboots or loss of power, you probably don't want to send your mail to your local box. Get a cheap $5 account and send all mail there. My connection went down when I was away on vaca and I lost lots of mail.
Bad call.
Internet mail is VERY tolerant of crappy connections. If you have a decent ISP they'll provide a backup MX for you. If not, find a couple of buddies with equally crappy connections (preferably with diferent providers) and "trade" MX records with them.
A single POP box is enough for most people, but someone who wants to run his own webserver will likely find enough value in running mailing lists, setting up multiple boxes for SPAM control, etc. to make the little bit of extra configuration worth while.
First, I think the spray on cells are a great idea. They could use CFC propellant as a method to improve gain as they are deployed.
.
But I jest.
More seriously, has anyone looked at the potential for global cooling of converting significant amounts of solar energy to mechanical? Sure some of the input to machines is returned to heat, but any movement is solar heat the we'll never get back.
I'm serious.
And let's not forget that Earth Day was organized to fight global cooling . .
-Peter
I don't think it was XML at the time. It's been about a year, but as I remember it it was a shitload of extra Apache config stuff.
Yeah, it can't have been XML because it was full of impossible to follow comments surrounded by hashes. You know the kind:
##########
# comment #
##########
I had a better example, but the lamness filter wouldn't let it by. Anyway, it had these huge comment blocks that really didn't shed any light on things. I'm sure they were really meaningful to people who already knew what the stuff did, but that doesn't help me much, does it?
Anyway, I hope it is better now. Maybe a lot of guys like me kept harassing them and they made the install a little nicer.
I'm totally with you on the "one man's pleasure" thing. As I said in another post in this thread, Java stuff always gives me the heebie-jeebies, but I know lots of guys who are smarter than me who love it. Go figure.
-Peter
You assume too much, dear lady.
I was not fortunate enough to be exposed to *NUX until about four years ago, but I did take Applesoft Basic in the eighth grade. And I happen to have just polished off a little bash script today. That's not the issue.
The issue is that I think it is a fundamentally bad idea to have a server service depend on a special environment variable. IMO it is too fragile a way of doing things. Think "This f-ing thing starts fine from the command line, but fails to start by the init script."
You are, of course, free to disagree with me, but I think your "these kids today" attitude reflects on you far more than it does on me.
Oh, and I was in high school by the time I had any sort of regular access to computer equipped with a mouse.
So, I stand behind my statement that Tomcat (actually, Jakarta as I understand it) was unnecessarily difficult to install at the time I first attempted it (about a year ago, it may be a snap now). I'm not saying that I need a "wizard" to wipe my backside during an install, but Jakarata was too far the other way IMO.
Just as a note for perspective, it took me about four hours and two Mt. Dews to install Apache with SSL and all the trimmings the first time. It took me the better part of three days and a case of the Dew to install Jakarta the first time.
Finally, maybe I am just a dumb kid, but the last paragraph of your reply didn't make any sense to me at all. I'd really like to know what you meant and I hope you will offer a bit of an explanation.
-Peter
Well, I think it is more that different people see the world differently and that "the Java people" and I are 180 degrees out of phase.
Everything with Java freaks me out for some reason. Every time I delve into something Java related I fell inundated with so much jargon that it is easier to just not use it.
Catalina:
Is that the _OLD_ version of Tomcat, or one of the componets, you know I really can't remember.
This is really funny. My point was that the Jakarta stuff has way to many shifting labels for a freaking servelet engine and your response is basically "no it doesn't, it's quite simple . . . now how the hell does it go again?" (BTW, I take it that it is the new new servelet container . . . whatever that means.)
Does lack of knollage mean lack of hillyness?
Why do you feel you need to resort to name calling?
Have I fallen prey to feeding a troll?
-Peter
Apache Tomcat? Easy to install? I swear that thing is packaged by Rain Man or some acid freak or something.
Who even knows what the fucking things correct name is? Is it Tomcat? Jakarta? Catalina?
What kind of server program depends on enviornment variables? I'll tell you, Apache Jakarta Tomcat freaking Catalina with Ant on the side.
Oh, and make sure you put the correct 500 lines of crap about "workers" in the apache config file.
Yeah, it's a breeze.
-Peter
I don't have any inside information here, but it is important to note that you said "half the cost." not "half the profit margin for Intel."
/AMD/'s market, not the pentium.
Clearly celeron yeild was much better at the time, so without Intel's internal numbers we can't know if this was good or bad for them.
This is compounded by the fact that a significant percentage of those celeron sales were cutting into
-Peter
First, don't overlook webapps. I used Squirrelmail for all my email for quite a while. It was great to be able to get all my mail (and archives) securely from ANYWHERE that had an SSL capable browser and an internet connection. PHPGroupWare is another great example. Webapps are BY FAR the most flexable from the client perspective.
Beyond this there are two more strong options.
As about fifty people have said, VNC. VNC has the advantage of working on many platforms and being able to re-direct an entire desktop. VNC "becomes" a webapp via its ability to provide your desktop via any Java capable browser. This is a strong option for your situation.
Xfree86 is a good option for serving up individual apps, and is really handy when paired with ssh (-X option). This option is better suited to a large number of fixed clients (i.e. workstations) using a small number programs (i.e. X clients (geez I hate the X terminology.)) regularly. Not so great for your situation.
Good Luck!
-Peter
How does selling something along with something you get for free drive down your margins?
s _Q12002.html. Maybe "making" was too strong a word. Made a profit in Q1 of '01.
Let's say that Red Hat and MS each sell an OS for $100. Each expects to spend $50 supporting it. RH has $15/copy (at expected distribution volume) invested in development, and MS has $30, since the write the whole thing from scratch.
Who has the larger margin?
Now, these are all made-up numbers, but I think that they are useful for illustration purposes. Can you make up a set of reasonable numbers to illustrate how bundling support and distribution of software that you largely get for free hurts your margin?
The way I explain that RH isn't making money hand over fist, but MS is is simple. Volume. I think that the reality is that RH spends something on the order of 1/10 what MS does on development, and has something like 1/1000 the (full price paid) distribution. So the numbers are more like 100/50/150 vs. 100/50/30.
Perhaps I was mistaken about Red Hat making a profit. I swear I read that somewhere. Ah, wait, here it is http://www.redhat.com/about/presscenter/2001/pres
OTOH, your $120 billion figure, if I'm not mistaken, is their peak market cap. Which is bullshit. Market cap is literally meaningless. It has nothing to do with actual money. Not money that they have, have spent, people have spent on them. Nothing.
That statement, combined with your statement that adding value by packaging and selling something that you get for free hurts the economy makes me question your grasp of economics.
Now, I know nothing about accounting, but my understanding of the English language leads me to believe that they had a quarterly loss of 17M in 2000 (and a somewhat higher loss in the same quarter of 2001). Which leads me to question your interpretation of any facts.
Finally, who said anything about "open source?" I'm talking about Free Software.
-Peter
Isn't 8gb enough for DOS? Surely it is enough to boot.
Any reputable OS only uses the BIOS to get a minimal system loaded to switch to protected mode disk access. (This is all, of course, PC specific.)
As long as the bios can see the first 15 megs, I'm set!
-Peter
If not, why does it need an itinerary?
This is what they want us to pay for.
-Peter
It is abundantly clear that you didn't read the page I linked to.
Most of what you said is based on the exact confusion arising from the phrase "selling software" (and variants you used like "selling App" or "selling gpl software" or "selling code") that is explained in the page I linked to.
So, since you don't care to read that article, let me establish some vocabulary.
If "selling software" is to have any consistent meaning it must be selling the copyrights to a piece of software. Such as when Corel bought WordPerfect. This clearly is not the topic of the discussion.
Now we come to what you are really talking about, which is selling software licenses. When you "buy software" (really "buy a license") you never get anything but the use of the software IAW the license terms. If you actually "bought windows" why may you not sell it? I don't mean en masse, just the CD you bought? Because you didn't buy anything but a license.
Finally we have distributing software. Which is what I was talking about. Wal-Mart makes money by distributing both proprietary and Free Software. It doesn't make a difference to them. Redhat sits on the shelf right next to XP. See my other reply in this thread for more examples of people making money by distributing free software.
Finally, note that if we can agree to the terminology above then you were more correct than you know, since there is there is no license for use of Free Software distributed under the terms of the GPL to sell.
To be totally clear about what I just said; the GPL isn't a "software license" in the sense that many people think it is. The GPL is a software distribution license. It makes no demands on the user (unlike a EULA) except that they may not sue if they don't like the way the program works, or fails to work.
So again, there is no software license to sell. Thus, you are correct that selling licenses for unlicensed software is not a promising business model. That, however, has nothing to do with my original post.
-Peter
I disagree. Redhat charges a premium for priority FTP access to software which can be freely distributed. The FSF itself was formed with money made by selling GNU on tape.
It is true that Free Software does not have the "advantage" of artificial scarcity that proprietary software has. In spite of this, both Cheap Bytes and KRUD both operate in the black AFAIK.
If we expand beyond simple distribution there are additional ways to actually make money by distributing Free Software that have been demonstrated in the real world. Redhat turns a profit, largely by bundling service with distribution. Several of the PHPGroupWare guys support themselves by supporting PHPGroupWare when they aren't hacking on it. Other value-adds exist, such as IBM bundling Free Software with hardware.
But, I suppose it is true that you aren't going to make yourself rich by downloading Free Software on your cablemodem and mailing out burned CDs.
-Peter
To see what RMS actually thinks about this subject see http://www.gnu.org/philosophy/selling.html
From that page:
Then again, when has an AC let reality interfere with the contents of his posts?
-Peter
Many of you may be aware that during the production of Return of the Jedi the title Revenge of the Jedi was used to mislead knock-off merchandise manufactures.
.
Lucas has pulled this old trick out of the bag. Let me be the first to announce that the actual title of Star Wars: Episode II is to be Attack of the Marketing Drones
-Peter
Gee, an AC who doesn't know what he is talking about. I'm so shocked.
First, wireless would double (or more) the cost of the device that I am imagining.
Second, you are flatly wrong about Power over Ethernet (PoE). See http://www.hyperlinktech.com/web/poe.html for just one example. From that page:
Perhaps you should spend thirty seconds on google before you shoot your mouth off next time?
But, seriously... It's probably as easy or possibly easier to decode MP3s than it is to decode WAV files. My understanding is, this is exactly what the sliMP3 player is...
So you are saying that it is easier to do decode the MP3 and then do a digital to analog conversion than it is to just do the D/A conversion? You can't possibly defend that statement.
Get a login. Get a clue. Then get back with us.
-Peter
Weird. Spool is empty. It's an ISP pop account.
Try me at spam@hutnick.com
-Peter
A sound card that uses ethernet instead of pci/isa, what a great way to say this in less than 80 words!
/dev/hi-fi" None of this wait for the vendor/flash the device/give up on the vendor and rip out the OS and put in Linux crap.
The reason I bring up UNIX is that adding a codec would be as simple as "$ foo_codec_play some_song.foo >
Seriously, if you skip the remote and just use raw ethernet frames I think this could be done with no brains at all in the "ethernet based soundcard".
I don't know what "infiniband" is, but as I figure it a 1x cdrom (which is where the 44.1kHz PCM thing came from) xfers at 150kB/s and 10meg ethernet is 1280kB/s ((10mb) * (1024k/m) / (8b/B)). So this thing would use 12% of 10meg and 1.2% of 100meg ethernet, not accounting for overhead.
I'd prefer setting the MAC by dips. Then if it does us IP, just set up MAC affinity in DHCP. Anyway, like I said, the driver could handle all this. The more I think about it the more I think IP is overkill. Being limited to being on the same ethernet segment as the device is not a big limit. And you could always add a proxy system (if you want to play on a device on a seperate segment you proxy through a machine on the same segment as the device). Again, make the hardware dumb and the software smart.
Oh, and I updated my email address. Please mail me if you want to persue this. I'll buy all the parts if you build two!
-Peter
One of the things that makes the PC so powerful is the ability to add "dumb" peripherals that add major functionality by leveraging the generality of the PC.
/dev (maybe /dev/hi-fi?) device, rather than having to address it by IP.
/dev/dsp for "main system sounds." Allowing high quality, hack-free use of the stereo for general PC audio.
/removed/.
In this vein, why not have a simple peripheral that:
1. Gets power from ethernet
2. Talks IP or some simpler protocol
3. Has a DAC that is able to convert one uncompressed digital format (say, 44.1kHz PCM) to high quality stereo analog.
4. Optionally has an IR or RF remote that sends simple events back to the PC.
Here is the real beauty, especially on a UNIX system: adding codecs is trivial. Really adding any functionality is trivial. Want a random mode, code it up. Need more space, add it to the PC, want to be able to play all songs that match a regexp, code it up. The real slick thing would be that it could play audio that was encoded in a way never even dreamed of by the hardware OR the software designers through UNIX pipes. There could also be a driver that allows it to appear as a simple
I suspect that any 2nd year EE student could design and build this thing with a DAC scavenged out of a CD player that suffered mechanical failure.
You could even use this to replace
Some ideas about obvious problems:
1. Give them all the same MAC, but make the last byte configurable by dip switches (to allow multiple devices on the same ethernet segment). All config could then be handled automagically by the driver.
2. Design it to take power from ethernet OR an A/C adapter.
3. It will need some buffer. A few megs should be overkill.
4. In case this isn't clear, it would be a totally "push" device. It would sit there and just dump whatever came over the wire with its IP/MAC into a buffer, which dumps right into the DAC.
Advantages:
1. Never fight with Linux sound drivers again.
2. All DA conversions prior to the box itself are in software, and therefore don't pick up "chassis noise" like a soundcard does.
3. "infinite" upgradeability through PC software.
4. Total platform independence. If it talks ethernet (and IP?), it can control the device.
5. "Patent-proof" I am pretty sure you can't patent something that is just like something else, only with components
The only real question in my mind is wether it is worth it to implement IP on the thing, or if it would be better to just use a very simple, non-routeable protocol on top of ethernet.
Does anyone know of anything like this? By like this I mean something that is simply a DAC with ethernet. There a a bunch of devices out there that decode MP3s, etc. That's not what I am looking for.
Any EEs out there want to take a whack at it?
And, to come back around to being on topic, he would be able to "play" his shoutcast stream locally, and push it out to his stereo with this thing, if it existed.
-Peter
And I do happen to agree that seniority and time spent with a company should be a factor in determining who gets promoted.
Ugh. It should be a (very small) factor. It does breed discontent when some "wiz kid" goes straight to the top without "paying dues."
OTOH I have had to work sorta-kinda under some real dubmasses who just occupied a chair so long without doing anything to get fired that they got promoted. That wasn't good for the company or my blood pressure.
-Peter
Okay, so what you meant was "VNC does not ADD multiuser GUI functionality to Windows." which is substantively different from what you said.
Now, will you please explain what the hell you mean by a "samba box?" AFAIK SAMBA is an SMB server/client package for UNIX. Do you mean to say that there is a UNIX port of MSTS? If so, I'd love a link.
Finally, you really need to work on your communication skills. You clearly know what you are trying to say, but it is coming out as half-formed nonsense.
-Peter
Can you offer any explanation? VNC can provide several remote desktops at the same time.
I'm not saying I advocate this as an ideal solution, but, having used VNC quite a bit I don't see how you can back up your assertion.
Also, what are you talking about in your second sentence? What do you mean by "terminal server?" Surely you don't mean MS Terminal Server . . . ?
-Peter
It's a SouthPark reference.
-Peter
The idea was: never again should a democracy change into a racist totalitarian state, and it's worthwhile to give up that bit of freedom to prevent this from happening again.
Which is to say "we must give up our liberty in order to save it."
"Thanks for throwing out the bath water, hon. Where's the baby?"
My real point is that you don't seem to understand the implication of what you are saying. If you give up the right/power/control to decide what you say it doesn't just dissapear . . . that control goes to the government
History clearly indicates that giving a government power is not a way to ensure that your freedom is protected.
-Peter
AFAIK all the Dells use water.
Water is not actually conductive. Water with disolved solids, noteably salt, is quite conductive. It doesn't have to be cloudy or "salty" to be salty enough to be pretty conductive.
-Peter
mail: if you are prone to outages, reboots or loss of power, you probably don't want to send your mail to your local box. Get a cheap $5 account and send all mail there. My connection went down when I was away on vaca and I lost lots of mail.
Bad call.
Internet mail is VERY tolerant of crappy connections. If you have a decent ISP they'll provide a backup MX for you. If not, find a couple of buddies with equally crappy connections (preferably with diferent providers) and "trade" MX records with them.
A single POP box is enough for most people, but someone who wants to run his own webserver will likely find enough value in running mailing lists, setting up multiple boxes for SPAM control, etc. to make the little bit of extra configuration worth while.
-Peter