I read this guy's post in disbelief. At one point he says he has been contributing for a while because he believes in linux and gentoo, and at another he says that he expected his contributions to be treated as building some sort of long term path that would be financially beneficial to him. How can you write code, contribute it to a major GPL project, then not realize that your contribution is one of thousands, and that there is no major plan to reward individual contributors?
Making demands rarely works out well. Depending on the company, though, they may be willing to do comp time after the project is done - at least for the weekends. So 6 weeks of working, followed by a 12 day paid break.
Let's be realistic here: this isn't a huge death march. It's 4-6 weeks of long hours.
Reverse time to just before the point where someone attains godlike status, and stop them. In current time, they'll suddenly no longer have whatever magic abilities they built up. Solving the whole time travel/continuity issue can be the fun job of some game development studio. That's why i'm paying them $50.
It's actually not snake oil. 2 years ago I built a demo version of the Iomega HipZip firmware that had SRS running on it. They have a hardware and a software version, i'm *guessing* that all handhelds use the software version for price.
The software version shifts all the audio down by 6db or so, then performs its adjustments. The result is that audio players that have SRS end up being always quieter, even when SRS is disabled, so when SRS is enabled it sounds like an improvement. For me this was a major dissapointment, considering the low power of handheld amps in the first place.
As far as the audio quality itself, I could hear that *something* was happening, and they (SRS labs) seemed happy with it, but i would never use it. The shifting was audible, but there was a bit of distortion, so it was impossible to get any sort of clean sound out of it. We had two hardware SRS WOW boxes which did the same thing.
Sure, it's a public place. Like a public park. It's nice to walk down to a public park, play on the swings, etc.
Imagine going down to your neighborhood park and finding 500,000 people jumping up and down. Sure, it's a public place, but it wasn't designed for that. It was designed for 50 or 100 people to hang out for a bit, and move on.
If you think about it a bit more though, this is like someone's back yard. This guy has to pay for bandwidth. He's got a sign saying "sure, come in, sit down for a bit". It's not public, it's private, and he's being generous in letting people use it, but that generocity is abused when slashdot decides to pour people all over his site.
Last example. Think of public marches. The roads are owned by the people, and it's perfectly acceptable for 50,000 people to march through downtown seattle, with streets closed down. BUT, before they can do that, they have to ask permission and obtain a permit from the city. It's simple consideration for others before taking for yourself.
i've posted in the past about how silly it is for slashdot to link to sites that can't handle the load, and to link without permission. but i dont think anybody cares - i think the assumption is that if it is on the web it can handle 100,000 hits per hour, and why would anybody need to ask.
Err, these are competing philosophies. You can't have both types of scheduling going on. Think about it: you have an interactive process which wants to use all the CPU all day long, and you have 6 server processes that want to have balanced scheduling for the clients they are handling. No matter what the scheduler chooses, it is being unfaithful to your bit for each process.
The better answer is to either a) make this option compile time, as someone mentioned, or b) make this option configurable (a la sysctl) at runtime. This would allow distribution maintainers to adjust the setting to match the type of installation they are doing, and users on stock installations to quickly adjust the kind of scheduling they have, just like the little check boxes in windows NT/XP.
ahh yes, i was checking for such a comment before i posted the same thing. it seems like a set top box should do more than just one thing (like play games).
I haven't tried that yet actually. The doors are closed by a push-latch at the top and have those slow-swivel gears on the sides (words...escaping...mind). Usually if i am doing something with the CD drive i just leave the top door open, i'm guessing that the drive wouldn't be able to force it open at all.
I'm planning on getting a slot loading cd drive at some point; i dont know if that would handle things differently (since there is no tray state? who knows)
I have the 610 case. It's an early adopter product. It's good construction, and it's worth $200, but they missed a bunch of details.
Getting a micro-atx mainboard is no big deal. I bought an ASUS A7N266-VM board for $70 at newegg. Athlon XP 1700+ is about $55, and a stick of 512mb mushkin was $130, so all in all, getting new components was reasonable compared to the $205 for the case (at trendetronics.com).
Back to the case itself, the things they missed are:
1) Window for IR receiver
2) Removable drive bracket; it'd be nice to mount all my drives in a bracket then screw that down to the chassis
3) Quiet case fan; the 60mm cooler-master fan it came with was a tad loud. i bought a vantec stealth and that was ok, but nothing magical.
4) Cable routing. Finding a way to run the front usb connectors around the entire case to the internal USB jacks was...interesting. right now i think i have them going under the motherboard
I researched HTPC cases for months before buying this one; there was basically this, a case available in germany only, and another case that was $500 (but looked oh-so-cool). All in all it was worth it; it matches my stereo equipment, it's quiet (now), and gives me a PC that doesn't look anything like a PC.
I think the point is the idle noise. All your examples with VCRs involve actions (hit stop. hit eject. hit rewind). With computers/hard drives, there is noise on a constant basis due to fans and drives spinning. The western digital special editions are probably the worst offenders noise wise - they are fast, but they are loud and run hot.
Other people have mentioned it, but the seagate barracuda IV's and (just over the past month or two) V's are the quietest drives available, come in large capacities, and in the case of the V's perform fairly well in benchmarks (although still slower than the WD drives).
I've recently built a HTPC for my living room using a cooler master 610 case, zalman 5100-cu low noise cpu fan, a baraccuda IV mentioned above, and an athlon xp 1700+. it runs about as loud as my PS2 (which has a dinky low rpm 50mm fan).
I totally agree. I think a lot of this plan was banking on the "look at all the places we will go" aspect of proposed expanded routes, but there's no mention of how that will be funded, when it will actually arrive, etc. The potential for the initial line to go over budget and for seattle residents to feel burned when they pass additional taxes to finish it is so high, it's not even funny.
The worst part about this plan is that it intentionally places the majority of the burden on people who will never use the monorail. People who have nice, $80k BMWs will have to pay something like $800 a year in taxes for the monorail, even though they'd have no reason to use it (they have a BMW). Unemployed carless hippies, who write columns for the stranger, will pay $0 for the monorail construction and will reap the benefits.
I dont think a monorail is a bad idea in general; there may be a plan that actually covers significant portions of downtown, has a tax plan that scales, has a route or connection for easy airport access, etc. But the plan that passed has none of this, wont fix congestion on I5/I90/99 downtown, and by its own flaws will likely prevent the expanded routes from being built.
Uhh if you live in the ID? Yeah you can walk 10 blocks to the monorail station at the stadiums then catch the monorail up to...west seattle or downtown or ballard.
I didn't know it hit queen anne, i thought the route went up 15th, so it went between magnolia and queen anne up to ballard. Also it doesn't make it as far as greenwood or greenlake last i checked.
The real problems in seattle are: traffic on I5, traffic on I90, and traffic on the barely alive alaskan way viaduct (99). The monorail, as proposed, does nothing for people who live in the U district (maybe they could take a bus over to ballard then hop on the monorail? FUN), capitol hill, beacon hill, fremont, most of queen anne, wallingford, blah blah blah. It does very little for people who live in belltown or downtown (what, i'm going to hop on the monorail to go to a mariners game twice a year? well, that was worth $150). It also uses a 30 year tax to fund one single line; how are the other proposed expanded lines going to be funded? additional 30 year taxes of 1.4% annually on cars?
Having "a" monorail might be a good idea, but having _this_ monorail is a bad one.
i dont think it would work for cars. if you can't see the road, then how will anything else see it? and if the image of the road was taken previously, how does that work for things like oncoming traffic (which aren't displayed in the fog image)
You actually can decode ogg on something smaller than a 75mhz arm, it just depends on what else you are doing. The integer version they released runs in
Someone else suggests streaming wav files, but i'm guessing you dont have enough ram to buffer that and on networks with mild congestion you'd get dropouts. Transcoding is really not good for audio quality, for those who consider the above to be a viable option.
As a side note, you mention that it's not possible for manufacturers of inexpensive playback devices to support the format, but that's not correct - iRiver will likely add ogg to their devices (which are mostly cirrus based). Also you dont really have an inexpensive device ($250 is a lot, since you aren't handing out 30% margins to retailers on this).
Yes, i read your comment. I think we're not too far off here: Choosing to support both is more expensive, UNLESS you have silicon that supports both, which will drive the cost of dual a/b base stations down/wifi cards down. This is all i'm getting at - infrastructure providers, like AT&T, aren't going to buy into one technology or the other when there is no clear winner, but if the opportunity to support both comes up, they will.
so if you are expanding the reach of your overall network, you now have double the cost, since you have to drop two access points in. also if you have areas where you have b coverage but not a, your clients with a cards are stranded.
no, really, you need to have pc cards that can handle both a and b before a will see adoption.
there was an article a while back about a company developing single silicon that supports a/b/g. i would expect something like that to make the transition possible. unless you can mix the networks easily (like you can with 10/100 networks) i dont see a transition happening for a while. especially since b is quite a bit cheaper right now.
the range is also a definite issue. for home use it's not a big deal, but the value of wireless isn't home use.:)
i'm fairly certain the riovolt player was written by iriver, which i know has been looking at getting ogg support for a while. no clue as to whether or not sonicblue will actually pay them to make an update with ogg support.
Slashdot Mirror
No, but i would laugh if an NPR reporter quit because he realized, after 2 years, that he wasn't going to get rich off it.
I read this guy's post in disbelief. At one point he says he has been contributing for a while because he believes in linux and gentoo, and at another he says that he expected his contributions to be treated as building some sort of long term path that would be financially beneficial to him. How can you write code, contribute it to a major GPL project, then not realize that your contribution is one of thousands, and that there is no major plan to reward individual contributors?
Making demands rarely works out well. Depending on the company, though, they may be willing to do comp time after the project is done - at least for the weekends. So 6 weeks of working, followed by a 12 day paid break.
Let's be realistic here: this isn't a huge death march. It's 4-6 weeks of long hours.
Three words: Time Machine Hack
Reverse time to just before the point where someone attains godlike status, and stop them. In current time, they'll suddenly no longer have whatever magic abilities they built up. Solving the whole time travel/continuity issue can be the fun job of some game development studio. That's why i'm paying them $50.
That was my thought too. When the Matrix Online comes out, this will give a whole new meaning to 'hacking the matrix'
It's actually not snake oil. 2 years ago I built a demo version of the Iomega HipZip firmware that had SRS running on it. They have a hardware and a software version, i'm *guessing* that all handhelds use the software version for price.
The software version shifts all the audio down by 6db or so, then performs its adjustments. The result is that audio players that have SRS end up being always quieter, even when SRS is disabled, so when SRS is enabled it sounds like an improvement. For me this was a major dissapointment, considering the low power of handheld amps in the first place.
As far as the audio quality itself, I could hear that *something* was happening, and they (SRS labs) seemed happy with it, but i would never use it. The shifting was audible, but there was a bit of distortion, so it was impossible to get any sort of clean sound out of it. We had two hardware SRS WOW boxes which did the same thing.
Sure, it's a public place. Like a public park. It's nice to walk down to a public park, play on the swings, etc.
Imagine going down to your neighborhood park and finding 500,000 people jumping up and down. Sure, it's a public place, but it wasn't designed for that. It was designed for 50 or 100 people to hang out for a bit, and move on.
If you think about it a bit more though, this is like someone's back yard. This guy has to pay for bandwidth. He's got a sign saying "sure, come in, sit down for a bit". It's not public, it's private, and he's being generous in letting people use it, but that generocity is abused when slashdot decides to pour people all over his site.
Last example. Think of public marches. The roads are owned by the people, and it's perfectly acceptable for 50,000 people to march through downtown seattle, with streets closed down. BUT, before they can do that, they have to ask permission and obtain a permit from the city. It's simple consideration for others before taking for yourself.
thank you for this post.
i've posted in the past about how silly it is for slashdot to link to sites that can't handle the load, and to link without permission. but i dont think anybody cares - i think the assumption is that if it is on the web it can handle 100,000 hits per hour, and why would anybody need to ask.
Err, these are competing philosophies. You can't have both types of scheduling going on. Think about it: you have an interactive process which wants to use all the CPU all day long, and you have 6 server processes that want to have balanced scheduling for the clients they are handling. No matter what the scheduler chooses, it is being unfaithful to your bit for each process.
The better answer is to either a) make this option compile time, as someone mentioned, or b) make this option configurable (a la sysctl) at runtime. This would allow distribution maintainers to adjust the setting to match the type of installation they are doing, and users on stock installations to quickly adjust the kind of scheduling they have, just like the little check boxes in windows NT/XP.
I heard SP1 for the 745i will add email. Email viruses in your car, yipee!
ahh yes, i was checking for such a comment before i posted the same thing. it seems like a set top box should do more than just one thing (like play games).
I haven't tried that yet actually. The doors are closed by a push-latch at the top and have those slow-swivel gears on the sides (words...escaping...mind). Usually if i am doing something with the CD drive i just leave the top door open, i'm guessing that the drive wouldn't be able to force it open at all.
I'm planning on getting a slot loading cd drive at some point; i dont know if that would handle things differently (since there is no tray state? who knows)
I have the 610 case. It's an early adopter product. It's good construction, and it's worth $200, but they missed a bunch of details.
Getting a micro-atx mainboard is no big deal. I bought an ASUS A7N266-VM board for $70 at newegg. Athlon XP 1700+ is about $55, and a stick of 512mb mushkin was $130, so all in all, getting new components was reasonable compared to the $205 for the case (at trendetronics.com).
Back to the case itself, the things they missed are:
1) Window for IR receiver
2) Removable drive bracket; it'd be nice to mount all my drives in a bracket then screw that down to the chassis
3) Quiet case fan; the 60mm cooler-master fan it came with was a tad loud. i bought a vantec stealth and that was ok, but nothing magical.
4) Cable routing. Finding a way to run the front usb connectors around the entire case to the internal USB jacks was...interesting. right now i think i have them going under the motherboard
I researched HTPC cases for months before buying this one; there was basically this, a case available in germany only, and another case that was $500 (but looked oh-so-cool). All in all it was worth it; it matches my stereo equipment, it's quiet (now), and gives me a PC that doesn't look anything like a PC.
I think the point is the idle noise. All your examples with VCRs involve actions (hit stop. hit eject. hit rewind). With computers/hard drives, there is noise on a constant basis due to fans and drives spinning. The western digital special editions are probably the worst offenders noise wise - they are fast, but they are loud and run hot.
Other people have mentioned it, but the seagate barracuda IV's and (just over the past month or two) V's are the quietest drives available, come in large capacities, and in the case of the V's perform fairly well in benchmarks (although still slower than the WD drives).
I've recently built a HTPC for my living room using a cooler master 610 case, zalman 5100-cu low noise cpu fan, a baraccuda IV mentioned above, and an athlon xp 1700+. it runs about as loud as my PS2 (which has a dinky low rpm 50mm fan).
That is so fucking amazing i could shit my pants.
I totally agree. I think a lot of this plan was banking on the "look at all the places we will go" aspect of proposed expanded routes, but there's no mention of how that will be funded, when it will actually arrive, etc. The potential for the initial line to go over budget and for seattle residents to feel burned when they pass additional taxes to finish it is so high, it's not even funny.
The worst part about this plan is that it intentionally places the majority of the burden on people who will never use the monorail. People who have nice, $80k BMWs will have to pay something like $800 a year in taxes for the monorail, even though they'd have no reason to use it (they have a BMW). Unemployed carless hippies, who write columns for the stranger, will pay $0 for the monorail construction and will reap the benefits.
I dont think a monorail is a bad idea in general; there may be a plan that actually covers significant portions of downtown, has a tax plan that scales, has a route or connection for easy airport access, etc. But the plan that passed has none of this, wont fix congestion on I5/I90/99 downtown, and by its own flaws will likely prevent the expanded routes from being built.
Uhh if you live in the ID? Yeah you can walk 10 blocks to the monorail station at the stadiums then catch the monorail up to...west seattle or downtown or ballard.
I didn't know it hit queen anne, i thought the route went up 15th, so it went between magnolia and queen anne up to ballard. Also it doesn't make it as far as greenwood or greenlake last i checked.
The real problems in seattle are: traffic on I5, traffic on I90, and traffic on the barely alive alaskan way viaduct (99). The monorail, as proposed, does nothing for people who live in the U district (maybe they could take a bus over to ballard then hop on the monorail? FUN), capitol hill, beacon hill, fremont, most of queen anne, wallingford, blah blah blah. It does very little for people who live in belltown or downtown (what, i'm going to hop on the monorail to go to a mariners game twice a year? well, that was worth $150). It also uses a 30 year tax to fund one single line; how are the other proposed expanded lines going to be funded? additional 30 year taxes of 1.4% annually on cars?
Having "a" monorail might be a good idea, but having _this_ monorail is a bad one.
i can actually see the niven sues bungie case, since halo was exactly what ringworld was described as. but oh well.
for lucas, he shouldn't have to give credit based on similarities
i dont think it would work for cars. if you can't see the road, then how will anything else see it? and if the image of the road was taken previously, how does that work for things like oncoming traffic (which aren't displayed in the fog image)
You actually can decode ogg on something smaller than a 75mhz arm, it just depends on what else you are doing. The integer version they released runs in
Someone else suggests streaming wav files, but i'm guessing you dont have enough ram to buffer that and on networks with mild congestion you'd get dropouts. Transcoding is really not good for audio quality, for those who consider the above to be a viable option.
As a side note, you mention that it's not possible for manufacturers of inexpensive playback devices to support the format, but that's not correct - iRiver will likely add ogg to their devices (which are mostly cirrus based). Also you dont really have an inexpensive device ($250 is a lot, since you aren't handing out 30% margins to retailers on this).
this is after, the reference for this is the cirrus ep7312 which has a ~ 74mhz arm core.
Yes, i read your comment. I think we're not too far off here: Choosing to support both is more expensive, UNLESS you have silicon that supports both, which will drive the cost of dual a/b base stations down/wifi cards down. This is all i'm getting at - infrastructure providers, like AT&T, aren't going to buy into one technology or the other when there is no clear winner, but if the opportunity to support both comes up, they will.
heh insightful.
so if you are expanding the reach of your overall network, you now have double the cost, since you have to drop two access points in. also if you have areas where you have b coverage but not a, your clients with a cards are stranded.
no, really, you need to have pc cards that can handle both a and b before a will see adoption.
there was an article a while back about a company developing single silicon that supports a/b/g. i would expect something like that to make the transition possible. unless you can mix the networks easily (like you can with 10/100 networks) i dont see a transition happening for a while. especially since b is quite a bit cheaper right now.
:)
the range is also a definite issue. for home use it's not a big deal, but the value of wireless isn't home use.
i'm fairly certain the riovolt player was written by iriver, which i know has been looking at getting ogg support for a while. no clue as to whether or not sonicblue will actually pay them to make an update with ogg support.