I researched business class offerings for Charter Communication's broadband offerings about a year ago and found the prices to be absolutely excessive. For business services, they were offering 256K d/l and 128K u/l for almost $150/month!
I bet if you check into the offerings by Comcast and Cox under business you'll find similar types of pricing schemes. It is absolutely outrageous. The local representative I talked with tried to convince me that I get a higher quality of service since problems with business customers are resolved with higher priority than residential customers. Unfortunately, in practice, this doesn't hold true.
In the two years I have been using cable broadband, the only outages I have experienced were outages that took place with equipment in NOCs, not problems with equipment on-site or in the 'last mile'. So priority would have done no good whatsoever - they'll fix the problem that is affecting thousands of subscribers just as quickly as they would if some of those subscribers were business customers (which I'm sure they are).
I presented this to the local rep and started getting attitude - why did this guy get defensive? Maybe he's used to dealing with types that can't call his company on their policies. I believe I certainly did.
Why pay between $150 and $250 per month for 256k-512k/128k service via cable when the local telco offers non-shared business class DSL at 768k symmetrical for less than $100 per month?!? I presented the local rep with those hard numbers and he gave me the lines about long waits for installing DSL... which is why I would pay $50 to $150 a month more for the length of the service, so that I could be up and running in one week vs. two to three weeks? I think not.
Incidentally, the local telco hooks up business DSL far faster than is average in the DSL industry - I experienced two DSL hookups in Chicago (both residential) via Covad (local 'last-mile' was Ameritech, go figure) and found the waits and lack of service (status reports) unforgivable. Locally, however (Madison, WI) a business DSL line was installed at the same time that phone service was installed (new office) and the DSL was active by our move-in date (2-3 days later)- no problems there!
If a company has a patent for creating a DRM OS, then the SSSCA can't possibly pass, right? That would create an instant monopoly, if I understand broadly what's going on here.
Either that, or Microsoft would have to license the patented technology on a royalty-free basis, which for Microsoft's uses, makes it rather useless, right?
...and to charter pipeline (aka charter "sipping straw") at (drum roll please) a max of 12Kbytes a second...
You must have an overburdened branch - my switch from @Home to Charter Pipeline went relatively smoothly (for such a major network shift) and my bandwidth is about what it was - 1.5-3Mbps.
Those who are electrical engineers correct me if I'm wrong, but this device appears to be centric to Europe in its video output. Its listed as having SCART, S-Video, and Composite (PAL) output. From what I know, the Composite signal isn't the only thing that is PAL/NTSC specific - S-Video is as well! If that is in fact the case, I don't believe any North American television will be of much use with this device. Too bad, in my opinion.
Patents can generally only be enforced over implementations sold or distributed in the country where the patent is issued. So, if a company has a US patent that is essential to a W3C Recommendation, then it could only seek royalties for products/services distributed in the US.
makes me shiver. Will developers that seek to write code implementing RAND standards need to become expatriots? If I recall, the EC still maintains that patents on software are a no-go (link: 11/2000).(more) Seems to me that developers in EC countries could then freely create software implementing RAND standards. Subsequently, OS distributions including open-source software based in the US (i.e. RedHat, Caldera, etc.) would be at a disadvantage to distributions such as Mandrake.
From what I was able to figure out from previous posts, I can copy termcap files from linux->SCO so that SCO recognizes my linux console - at that point, I'd be using a 'linux' or 'linux-console' terminal type, right?
A quick search on thinkgeek revealed that currently all MP3 players are USB driven (including the nomad). Me thinks that having a firewire driven MP3 player is anything but lame. Too bad for all those geeks out there still using PCs - guess you'll need to invest in a Firewire extension card:). And as far as the disk-space goes, expect that to increase in the next release of the iPod. I'm sure Apple wants to sell a few of these babies before it pours a bunch more money into it. Expect this one to come down in price and be superseded by one with more disk-space.
Also, note that this player is just slightly bigger than the Rio500/600 players w/ far more space... something that the nomad can't stake a claim on - the nomad is far to horsey to fit in any of MY pockets!
Finally - design matters. If you like 'em or not, Apple is definitely putting out products that look ultra-cool and dovetail nicely with their other product lines. Most of the players on thinkgeek look ultra-geeky, even the nomad (though its design and the design of the Rios aren't bad). I'm no designer, but I've come to appreciate good design and I'm willing to pay a bit extra for it.
Mac-on-Linux is akin to VMware - it runs Mac code natively on PPC processors by virtualizing the underlying hardware. It is amazingly powerful and the last I was hearing on the MOL lists is that OS X support will be back soon (yes, it was there when OS X was in its infancy, since then, a few changes in OS X have broken things in MOL). With Mac OS 9.x (and earlier), however, MOL is solid as a rock. I can run days, even weeks without any instability - eventually, though, Photoshop or something else will cause me to reboot MacOS.
If anyone's been scared to try out Linux on a PPC machine, for fear of losing MacOS, check out Mac-on-Linux... you'll have it to fall back on in case you get stuck in Linux. Most PPC Linux distros that I know of ship with Mac-on-Linux, so getting it running should be a snap. If you have any problems, the MOL mailing lists are amazingly helpful, often times Sam Rydh, the creator/maintainer of MOL will post responses himself.
FYI - sound works great, but video acceleration is lacking (much like VMware). USB support is also not there (yet) - but, if your device is seen in Linux, you can use it in MOL.
What will help, quite simply, is donating money. I support the EFF through my donations and most recently, I've donated money directly to Dmitry's fund (dmitry@shmoo.com). I encourage you to do the same, as this is quite simply the only thing that will have an effect. Money talks - if its important to you (I don't rightly see how it couldn't be) - put your money where your mouth is.
This guy is amazingly responsible. Too often, I have witnessed that the 'ease' of communication on the Internet has born exactly the type of irresponsible communicaes that Taco refers to. Ever notice that the searing email you sent off to someone that pissed you off is quite different than your tone of voice when you speak on the phone to a rep at a company (unless its your phone company!)
I can't speak for you, but for me - this is the case. After all, the person I'm talking to didn't personally ruin my day - they're probably getting paid close to minimum wage to listen to me moan and groan about something. If I want help, I certainly won't be yelling at them over the phone and basically using every expletive that comes to mind!
Take heed! Being 'Open' is an opportunity for greatness - the community can use it to breed a spirit of commonality and shared responsibility! Don't expect Win2K users to feel they have something in common with the next Win2K user. But we can be different - we can show the corporations that are willing to support Linux that we do appreciate their efforts, even when they make mistakes or missteps.
We've already builty the community, guys (and girls). Now we should wield the power of it in a positive way.
From what I've read (which isn't exhaustive, mind you) - the way these types of things are supposed to work is that they prevent a CD-ROM device from reading the CD on a hardware level. Apparently, regardless what commands you give to your CD-ripping application, the CD-ROM drive will be more picky about the way it reads the CD than an Audio CD player. And this pickiness is what gets it in trouble - the 'wildly erroneous' data that they've introduced confuses the firmware of the CD-ROM such that it believes that the data recorded on the CD is absolute garbage - and basically reports errors back to the OS, much as it would if the CD were scratched to hell.
I read about this in a previous article on Slashdot. From what I understand, all these schemes work from the same concept.
But you did anyway... the school she went to is not a UCLA type school - its one that only a select handful of students can attend at any given time (say less than 20). And besides having that degree, she also has a degree in business from Columbia in Chicago - which is really her main focus.
My apologies for not remembering the name of the school - but it wasn't exactly relevant to the topic (unless someone wanted to start a flame war on good audio engineering schools)
From what I recall, the school is called the Los Angeles Recording Workshop though I don't know if that was just a portion of the degree or not.
Actually, you may not be entirely correct. My fiancee and I have had quite a few discussions about high-end stereo equipment (like discrete amplified channels, speakers such as Definitive, Sonus Faber, and whoever it is that makes the filament speakers) as well as the quality of CD recordings vs. vinyl or tape or digital (DAT), etc. Now, mind you - she is an audio engineer, with a degree from one of the most reputed schools in the country (I think its in LA - I forget what its called).
One of the things she explains to me most often is that what you hear (on your CD) is NOT what was recorded in the studio, on stage at the symphony, etc. It is what has been cleaned and processed to sound best on a variety of common setups - like an average stereo, a boombox, and a common (stock) sound system in a common type of car.
So, while many folks believe they can hear more details than others (and admittedly, I'm sure recordings exist that still have these details in them), by and large - these details were lost on the studio post-processing floor. The goal of post-processing is apparently not to deliver the most hi-fi sound - its to deliver the music in a way that it sounds good in the most diverse environments.
To me, this sounds like what we get on CD is truly the lowest common denominator. At that point - encoding into MP3 or Vorbis doesn't seem to make more of a difference. I personally prefer Vorbis because (a) it sounds good (b) the file sizes are small and (c) it is patent free - that means a lot to me. Notice I didn't make any comparisons - I don't presume to say that it is better or smaller than anything else - I don't really care. Its the only format out there that allows me to enjoy my music in a format that is smaller than.wav files and more transportable than CDs whilst knowing that the algorithms are patent free to boot. Sweet deal if you ask me.
I entirely agree - it is absolutely unnecessary to name a vector drawing application a derivative of 'Illustrator'. Notice that Macromedia's application names (Freehand, Fireworks, Flash) bare no resemblance to Adobe's product names that perform similar functions (Photoshop, Imageready, Illustrator, Livemotion) - if they can do it, certainly the KIllustrator author can as well.
Now, if they were trying to cash in on the legitimacy of the 'Adobe Illustrator' product when their project was named, well busted.
Just because its free and you're making no money off it doesn't mean you can infringe on a company's branding. Let's look at the various image manipulation/viewing programs out there for Linux now (that I can think of in 5 seconds)- GIMP, ImageMagick, xv, Paint, etc. - none of these infringe on a corporation's branding (unless M$ want's to argue about Paint, I hardly think that it significantly adds or detracts from its corporate brand!!).
The comment about AIM and such is perfectly valid as well - come up with new names, people will use your application because it came up on a search, they were told to use it by a friend, its preinstalled on their Linux distro or whatever else - be more creative! If Open Source projects continue to rip off names, they'll continue to get in legal trouble and frankly, I have no sympathy for them.
GIMP has become quite popular on its merits, not because people were under the impression it was a free version of Photoshop. I've read often that GIMP is considered comparable to Photoshop and in some limited aspects it is - however, in many, many important aspects, it doesn't come anywhere close. Especially when tied in with Imageready, Photoshop is an immensely powerful tool that (unfortunately) so far GIMP cannot compare with. However - there are SO many things that I don't need Photoshop for and GIMP does those things miraculously well - and good for it!
Let your applications persevere on their merits, not on a borrowed name.
Just an FYI for some folks that may be unfamiliar with what's going on beneath the GUI for Mac OS X. As a ADC member, I've done a bit of looking into the nuts & bolts of OS X and its actually pretty interesting.
Here's a good URL to check for more info:
Darwin FAQ.
For those of you too lazy to read the entire thing, it mentions here that Darwin is based on NextStep and uses NetBSD applications and FreeBSD libraries.
How does Darwin relate to OS X? Well, the way that I've come to look at it is simple: Darwin is the *NIX and OS X (the proprietary part) is the GUI that users will learn to love.
In the FAQ, you'll also see mention of how much of the Darwin source is entirely processor/platform independent and that they actually have Darwin running on a particular configuration of an x86 machine. All that is needed to bring Darwin to x86 fully is more developers.
The FAQ also mentions that any updates to the Mac OS X source (the none-proprietary parts) are automatically reflected in the Darwin tree and that updates to the Darwin tree are eventually folded into the releases of OS X. That's powerful - think about it.
All in all, I think OS X is an excellent step for Apple and for the *NIX/Open Source community in general.
Not to offend anyone, but this information just isn't anything special. Of the information that is provided by the sandbox:
monitor depth
browser font size
connection speed
connection type
monitor horz
monitor vert
browser view horz
browser view vert
The only portion that can't be gleaned from the browser via JavaScript is the connection speed/type. While you probably won't be able to determine the connection type, it certainly isn't difficult to determine the connection speed. And when it comes to anything related to web design, the connection type itself is unimportant. Hence, when you determine the connection speed via a simple server-side script, you have all the relevant information that the sandbox provides.
Now, I agree with many of the posters - you shouldn't use connection speed info to determine what you serve up, maybe to make an informed decision of the default, but if you have multiple types of content created, you should always give the user the option to choose.
Note that gleaning the sandbox info and using it to determine what content to serve up may backfire on you anyway: two cases come to mind: a shared network at home is one, the second is downloading the kernel or maybe the latest Mandrake iso while you're surfing the web... in either case, there may not be nearly as much bandwidth available as the sandbox reports.
Germany may not have a Microsoft, but I believe RedHat has representation in Germany, SuSE is definitely huge in Germany (maybe even a German company?) so the options there are pretty open.
As best I can tell, we already do. One ATC operator could send a plane onto a collision course with another, and then there would be
nothing to tell the planes not to.
Not exactly. The Traffic Collision Avoidance System (TCAS) is an aircraft-to-aircraft system that works from a multitude of input data (as practically every on-board system does) to determine with the largest precision & accuracy the location of the airplane which is then broadcast out to other TCAS receivers in an attempt to avoid collisions.
Each TCAS receiver classifies incoming data in various levels: No Threat, Proximity, Traffic Alert, Response (?) Alert - which in aviation lingo translates to TAs (Traffic Alert) and RAs (Response Alert). The TAs and RAs are the only things of significance to the pilot since they are potential collisions (especially the RAs) - hence their color on the navigational and primary flight displays: yellow for TAs and red for RAs.
Based on his current flight (ascending, descending, level) the pilot keys in the TCAS system to show him only tracks of interest - if he's ascending, it makes no difference what's below him, etc.
This system is based on ground location data (from a multitude of sources, all based on radio towers located on the ground, with known positions) GPS data and differential GPS (where available - the only flight system I'm aware of that uses this in the US is the Boeing 777 with Honeywell's flight management computer (FMC) and Honeywell's large format display systems (LFDS) - similar to what was posted about the new shuttle display systems here on/. not too long ago.
Differential GPS is primarily only used for approaches, I believe - since that's the only time more precision than GPS is necessary (and even after the military scrambling of GPS ended, more precision is necessary to land the plane than military GPS provides) - 10 meter accuracy (or whatever it is exactly) puts you on the runway or off of it - not nearly enough for the autoland systems to feel comfortable with!
My reference: ex-avionics engineer for Rockwell Collins. That said, none of the statements made (express or implied) reflect the position, past or present, or presume to be 'expert advice' in any matters avionic, of Rockwell International Corp.
Just a real quickie comment:(meaning I don't have time to read anything but the/. abstract right now)
HTTP/NG has been in development for at least 2 yrs (probably slightly more) and it was designed to handle the deficiencies of HTTP while remaining based on many similar principles. Multiple channels was a HUGE part of the differences between HTTP and HTTP/NG.
I'll take some time to read about BXXP and compare it to my knowledge of HTTP/NG - but just wanted to throw this out...
From the original post, this company will accomplish the lunar mission at a cost far lower than NASA would accomplish.
Here's my warning: a large portion of the cost of designing/manufacturing the hardware and designing/writing the software for avionics & spacecraft is tucked away in the testing portion of each called verification.
As an avionics software engineer, I've witnessed this first hand. Conformance to standards such as D0178B (commercial avionics) and the mil standard (exact reference eludes me at the moment) is expensive! These standards are designed with one thing in mind: safety. Other considerations may also be folded in, but bet your bottom dollar, safety is the first priority. A comparison between the commercial avionics certification standard and the mil spec will show you just how much safety matters: the FAA cares a lot more about the safety of 300 travelers than a pilot & navigator in an F-18 (does that even have a pilot & navigator?).
This post in no way is meant to imply that this company is NOT following safety standards, rather (as the subject reads) providing a warning that low cost at the cost of safety is not low at all.
No it isn't. If you're programming for portability - maybe. Knowing your target platform is absolutely imperative. The goal here wasn't to program for x different platforms - it was to show that you have what it takes to code up algorithms to solve some of the toughest theoretical CS problems, under the constraint of time.
What happens when you work on a small embedded 8- or 16-bit processor?
Especially on this type of work, you want to know exactly what the compiler is going to do - you don't want to throw in a bunch of junk code to make your code portable. Its called optimization. Anywayz, they weren't compiling on a small embedded processor, they were compiling on 32-bit x86 processors running NT where its assumed that the compiler being used by the contestants creates target code exactly the same as that used by the judges. If not - this should be mentioned.
What happens when you work on a 64-bit system?
Well, since you asked, in this case, the algorithm would still work. The fact remains that they weren't compiling on a 64-bit platform, tho - see above.
The C language spec makes no guarantee for the exact range of an int.
This isn't what's at issue here. Maybe in some other thread, your comment has relevance, however, nobody in this thread lacks knowledge of what is and isn't guaranteed by the C language spec. Nobody really cares what you know about the C language spec, either.
I won't make the uninformed assumption that you've never been to a programming competition or that you don't know what its like to sit for hours trying to fix something that doesn't end up being a problem with your code... but it would certainly seem that way. You're pounding issues that are highly irrelevant.
Slashdot Mirror
I researched business class offerings for Charter Communication's broadband offerings about a year ago and found the prices to be absolutely excessive. For business services, they were offering 256K d/l and 128K u/l for almost $150/month!
I bet if you check into the offerings by Comcast and Cox under business you'll find similar types of pricing schemes. It is absolutely outrageous. The local representative I talked with tried to convince me that I get a higher quality of service since problems with business customers are resolved with higher priority than residential customers. Unfortunately, in practice, this doesn't hold true.
In the two years I have been using cable broadband, the only outages I have experienced were outages that took place with equipment in NOCs, not problems with equipment on-site or in the 'last mile'. So priority would have done no good whatsoever - they'll fix the problem that is affecting thousands of subscribers just as quickly as they would if some of those subscribers were business customers (which I'm sure they are).
I presented this to the local rep and started getting attitude - why did this guy get defensive? Maybe he's used to dealing with types that can't call his company on their policies. I believe I certainly did.
Why pay between $150 and $250 per month for 256k-512k/128k service via cable when the local telco offers non-shared business class DSL at 768k symmetrical for less than $100 per month?!? I presented the local rep with those hard numbers and he gave me the lines about long waits for installing DSL ... which is why I would pay $50 to $150 a month more for the length of the service, so that I could be up and running in one week vs. two to three weeks? I think not.
Incidentally, the local telco hooks up business DSL far faster than is average in the DSL industry - I experienced two DSL hookups in Chicago (both residential) via Covad (local 'last-mile' was Ameritech, go figure) and found the waits and lack of service (status reports) unforgivable. Locally, however (Madison, WI) a business DSL line was installed at the same time that phone service was installed (new office) and the DSL was active by our move-in date (2-3 days later)- no problems there!
Here's the link in case you want to try this out:
http://freshmeat.net/projects/ettercap/"
If a company has a patent for creating a DRM OS, then the SSSCA can't possibly pass, right? That would create an instant monopoly, if I understand broadly what's going on here.
Either that, or Microsoft would have to license the patented technology on a royalty-free basis, which for Microsoft's uses, makes it rather useless, right?
Those who are electrical engineers correct me if I'm wrong, but this device appears to be centric to Europe in its video output. Its listed as having SCART, S-Video, and Composite (PAL) output. From what I know, the Composite signal isn't the only thing that is PAL/NTSC specific - S-Video is as well! If that is in fact the case, I don't believe any North American television will be of much use with this device. Too bad, in my opinion.
The following:
makes me shiver. Will developers that seek to write code implementing RAND standards need to become expatriots? If I recall, the EC still maintains that patents on software are a no-go (link: 11/2000).(more) Seems to me that developers in EC countries could then freely create software implementing RAND standards. Subsequently, OS distributions including open-source software based in the US (i.e. RedHat, Caldera, etc.) would be at a disadvantage to distributions such as Mandrake.
I'm all for Ireland ... mmmm Guiness :)
I don't quite understand.
How do I 'use the scoansi terminal type'?
From what I was able to figure out from previous posts, I can copy termcap files from linux->SCO so that SCO recognizes my linux console - at that point, I'd be using a 'linux' or 'linux-console' terminal type, right?
Any insights, please send to my email.
Kermit doesn't do terminal emulation :( - at least this is what I've found in my quest so far.
A quick search on thinkgeek revealed that currently all MP3 players are USB driven (including the nomad). Me thinks that having a firewire driven MP3 player is anything but lame. Too bad for all those geeks out there still using PCs - guess you'll need to invest in a Firewire extension card :). And as far as the disk-space goes, expect that to increase in the next release of the iPod. I'm sure Apple wants to sell a few of these babies before it pours a bunch more money into it. Expect this one to come down in price and be superseded by one with more disk-space.
Also, note that this player is just slightly bigger than the Rio500/600 players w/ far more space ... something that the nomad can't stake a claim on - the nomad is far to horsey to fit in any of MY pockets!
Finally - design matters. If you like 'em or not, Apple is definitely putting out products that look ultra-cool and dovetail nicely with their other product lines. Most of the players on thinkgeek look ultra-geeky, even the nomad (though its design and the design of the Rios aren't bad). I'm no designer, but I've come to appreciate good design and I'm willing to pay a bit extra for it.
Mac-on-Linux is akin to VMware - it runs Mac code natively on PPC processors by virtualizing the underlying hardware. It is amazingly powerful and the last I was hearing on the MOL lists is that OS X support will be back soon (yes, it was there when OS X was in its infancy, since then, a few changes in OS X have broken things in MOL). With Mac OS 9.x (and earlier), however, MOL is solid as a rock. I can run days, even weeks without any instability - eventually, though, Photoshop or something else will cause me to reboot MacOS.
If anyone's been scared to try out Linux on a PPC machine, for fear of losing MacOS, check out Mac-on-Linux ... you'll have it to fall back on in case you get stuck in Linux. Most PPC Linux distros that I know of ship with Mac-on-Linux, so getting it running should be a snap. If you have any problems, the MOL mailing lists are amazingly helpful, often times Sam Rydh, the creator/maintainer of MOL will post responses himself.
FYI - sound works great, but video acceleration is lacking (much like VMware). USB support is also not there (yet) - but, if your device is seen in Linux, you can use it in MOL.
This site is mirrored at:
John Haller's server.
-Brice Ruth
What will help, quite simply, is donating money. I support the EFF through my donations and most recently, I've donated money directly to Dmitry's fund (dmitry@shmoo.com). I encourage you to do the same, as this is quite simply the only thing that will have an effect. Money talks - if its important to you (I don't rightly see how it couldn't be) - put your money where your mouth is.
For Rod's sake, mod this up!
This guy is amazingly responsible. Too often, I have witnessed that the 'ease' of communication on the Internet has born exactly the type of irresponsible communicaes that Taco refers to. Ever notice that the searing email you sent off to someone that pissed you off is quite different than your tone of voice when you speak on the phone to a rep at a company (unless its your phone company!)
I can't speak for you, but for me - this is the case. After all, the person I'm talking to didn't personally ruin my day - they're probably getting paid close to minimum wage to listen to me moan and groan about something. If I want help, I certainly won't be yelling at them over the phone and basically using every expletive that comes to mind!
Take heed! Being 'Open' is an opportunity for greatness - the community can use it to breed a spirit of commonality and shared responsibility! Don't expect Win2K users to feel they have something in common with the next Win2K user. But we can be different - we can show the corporations that are willing to support Linux that we do appreciate their efforts, even when they make mistakes or missteps.
We've already builty the community, guys (and girls). Now we should wield the power of it in a positive way.
From what I've read (which isn't exhaustive, mind you) - the way these types of things are supposed to work is that they prevent a CD-ROM device from reading the CD on a hardware level. Apparently, regardless what commands you give to your CD-ripping application, the CD-ROM drive will be more picky about the way it reads the CD than an Audio CD player. And this pickiness is what gets it in trouble - the 'wildly erroneous' data that they've introduced confuses the firmware of the CD-ROM such that it believes that the data recorded on the CD is absolute garbage - and basically reports errors back to the OS, much as it would if the CD were scratched to hell.
I read about this in a previous article on Slashdot. From what I understand, all these schemes work from the same concept.
Not to sound like a dick,...
But you did anyway ... the school she went to is not a UCLA type school - its one that only a select handful of students can attend at any given time (say less than 20). And besides having that degree, she also has a degree in business from Columbia in Chicago - which is really her main focus.
My apologies for not remembering the name of the school - but it wasn't exactly relevant to the topic (unless someone wanted to start a flame war on good audio engineering schools)
From what I recall, the school is called the Los Angeles Recording Workshop though I don't know if that was just a portion of the degree or not.
Actually, you may not be entirely correct. My fiancee and I have had quite a few discussions about high-end stereo equipment (like discrete amplified channels, speakers such as Definitive, Sonus Faber, and whoever it is that makes the filament speakers) as well as the quality of CD recordings vs. vinyl or tape or digital (DAT), etc. Now, mind you - she is an audio engineer, with a degree from one of the most reputed schools in the country (I think its in LA - I forget what its called).
One of the things she explains to me most often is that what you hear (on your CD) is NOT what was recorded in the studio, on stage at the symphony, etc. It is what has been cleaned and processed to sound best on a variety of common setups - like an average stereo, a boombox, and a common (stock) sound system in a common type of car.
So, while many folks believe they can hear more details than others (and admittedly, I'm sure recordings exist that still have these details in them), by and large - these details were lost on the studio post-processing floor. The goal of post-processing is apparently not to deliver the most hi-fi sound - its to deliver the music in a way that it sounds good in the most diverse environments.
To me, this sounds like what we get on CD is truly the lowest common denominator. At that point - encoding into MP3 or Vorbis doesn't seem to make more of a difference. I personally prefer Vorbis because (a) it sounds good (b) the file sizes are small and (c) it is patent free - that means a lot to me. Notice I didn't make any comparisons - I don't presume to say that it is better or smaller than anything else - I don't really care. Its the only format out there that allows me to enjoy my music in a format that is smaller than .wav files and more transportable than CDs whilst knowing that the algorithms are patent free to boot. Sweet deal if you ask me.
Now, if they were trying to cash in on the legitimacy of the 'Adobe Illustrator' product when their project was named, well busted .
Just because its free and you're making no money off it doesn't mean you can infringe on a company's branding. Let's look at the various image manipulation/viewing programs out there for Linux now (that I can think of in 5 seconds)- GIMP, ImageMagick, xv, Paint, etc. - none of these infringe on a corporation's branding (unless M$ want's to argue about Paint, I hardly think that it significantly adds or detracts from its corporate brand!!).
The comment about AIM and such is perfectly valid as well - come up with new names, people will use your application because it came up on a search, they were told to use it by a friend, its preinstalled on their Linux distro or whatever else - be more creative! If Open Source projects continue to rip off names, they'll continue to get in legal trouble and frankly, I have no sympathy for them.
GIMP has become quite popular on its merits, not because people were under the impression it was a free version of Photoshop. I've read often that GIMP is considered comparable to Photoshop and in some limited aspects it is - however, in many, many important aspects, it doesn't come anywhere close. Especially when tied in with Imageready, Photoshop is an immensely powerful tool that (unfortunately) so far GIMP cannot compare with. However - there are SO many things that I don't need Photoshop for and GIMP does those things miraculously well - and good for it!
Let your applications persevere on their merits, not on a borrowed name.
Cable:
3 Megabits/s down, 128k up (consistent)
DSL (TDS Metrocom):
768k/768k, at my location, actual is 707k symmetric.
Cost for Cable: $40/mo.
Cost for DSL: $150/mo.
Cable is a residential line, DSL is a business line.
Just an FYI for some folks that may be unfamiliar with what's going on beneath the GUI for Mac OS X. As a ADC member, I've done a bit of looking into the nuts & bolts of OS X and its actually pretty interesting.
Here's a good URL to check for more info: Darwin FAQ. For those of you too lazy to read the entire thing, it mentions here that Darwin is based on NextStep and uses NetBSD applications and FreeBSD libraries.
How does Darwin relate to OS X? Well, the way that I've come to look at it is simple: Darwin is the *NIX and OS X (the proprietary part) is the GUI that users will learn to love.
In the FAQ, you'll also see mention of how much of the Darwin source is entirely processor/platform independent and that they actually have Darwin running on a particular configuration of an x86 machine. All that is needed to bring Darwin to x86 fully is more developers.
The FAQ also mentions that any updates to the Mac OS X source (the none-proprietary parts) are automatically reflected in the Darwin tree and that updates to the Darwin tree are eventually folded into the releases of OS X. That's powerful - think about it.
All in all, I think OS X is an excellent step for Apple and for the *NIX/Open Source community in general.
Not to offend anyone, but this information just isn't anything special. Of the information that is provided by the sandbox:
The only portion that can't be gleaned from the browser via JavaScript is the connection speed/type. While you probably won't be able to determine the connection type, it certainly isn't difficult to determine the connection speed. And when it comes to anything related to web design, the connection type itself is unimportant. Hence, when you determine the connection speed via a simple server-side script, you have all the relevant information that the sandbox provides.
Now, I agree with many of the posters - you shouldn't use connection speed info to determine what you serve up, maybe to make an informed decision of the default, but if you have multiple types of content created, you should always give the user the option to choose.
Note that gleaning the sandbox info and using it to determine what content to serve up may backfire on you anyway: two cases come to mind: a shared network at home is one, the second is downloading the kernel or maybe the latest Mandrake iso while you're surfing the web ... in either case, there may not be nearly as much bandwidth available as the sandbox reports.
Germany may not have a Microsoft, but I believe RedHat has representation in Germany, SuSE is definitely huge in Germany (maybe even a German company?) so the options there are pretty open.
Not exactly. The Traffic Collision Avoidance System (TCAS) is an aircraft-to-aircraft system that works from a multitude of input data (as practically every on-board system does) to determine with the largest precision & accuracy the location of the airplane which is then broadcast out to other TCAS receivers in an attempt to avoid collisions.
Each TCAS receiver classifies incoming data in various levels: No Threat, Proximity, Traffic Alert, Response (?) Alert - which in aviation lingo translates to TAs (Traffic Alert) and RAs (Response Alert). The TAs and RAs are the only things of significance to the pilot since they are potential collisions (especially the RAs) - hence their color on the navigational and primary flight displays: yellow for TAs and red for RAs.
Based on his current flight (ascending, descending, level) the pilot keys in the TCAS system to show him only tracks of interest - if he's ascending, it makes no difference what's below him, etc.
This system is based on ground location data (from a multitude of sources, all based on radio towers located on the ground, with known positions) GPS data and differential GPS (where available - the only flight system I'm aware of that uses this in the US is the Boeing 777 with Honeywell's flight management computer (FMC) and Honeywell's large format display systems (LFDS) - similar to what was posted about the new shuttle display systems here on /. not too long ago.
Differential GPS is primarily only used for approaches, I believe - since that's the only time more precision than GPS is necessary (and even after the military scrambling of GPS ended, more precision is necessary to land the plane than military GPS provides) - 10 meter accuracy (or whatever it is exactly) puts you on the runway or off of it - not nearly enough for the autoland systems to feel comfortable with!
My reference: ex-avionics engineer for Rockwell Collins. That said, none of the statements made (express or implied) reflect the position, past or present, or presume to be 'expert advice' in any matters avionic, of Rockwell International Corp.
Just a real quickie comment:(meaning I don't have time to read anything but the /. abstract right now)
HTTP/NG has been in development for at least 2 yrs (probably slightly more) and it was designed to handle the deficiencies of HTTP while remaining based on many similar principles. Multiple channels was a HUGE part of the differences between HTTP and HTTP/NG.
I'll take some time to read about BXXP and compare it to my knowledge of HTTP/NG - but just wanted to throw this out ...
Regards,
Brice.
From the original post, this company will accomplish the lunar mission at a cost far lower than NASA would accomplish.
Here's my warning: a large portion of the cost of designing/manufacturing the hardware and designing/writing the software for avionics & spacecraft is tucked away in the testing portion of each called verification.
As an avionics software engineer, I've witnessed this first hand. Conformance to standards such as D0178B (commercial avionics) and the mil standard (exact reference eludes me at the moment) is expensive! These standards are designed with one thing in mind: safety. Other considerations may also be folded in, but bet your bottom dollar, safety is the first priority. A comparison between the commercial avionics certification standard and the mil spec will show you just how much safety matters: the FAA cares a lot more about the safety of 300 travelers than a pilot & navigator in an F-18 (does that even have a pilot & navigator?).
This post in no way is meant to imply that this company is NOT following safety standards, rather (as the subject reads) providing a warning that low cost at the cost of safety is not low at all.
As usual, my $0.02
Regards,
Jobe
That is a foolish assumption to make.
No it isn't. If you're programming for portability - maybe. Knowing your target platform is absolutely imperative. The goal here wasn't to program for x different platforms - it was to show that you have what it takes to code up algorithms to solve some of the toughest theoretical CS problems, under the constraint of time.
What happens when you work on a small embedded 8- or 16-bit processor?
Especially on this type of work, you want to know exactly what the compiler is going to do - you don't want to throw in a bunch of junk code to make your code portable. Its called optimization . Anywayz, they weren't compiling on a small embedded processor, they were compiling on 32-bit x86 processors running NT where its assumed that the compiler being used by the contestants creates target code exactly the same as that used by the judges. If not - this should be mentioned.
What happens when you work on a 64-bit system?
Well, since you asked, in this case, the algorithm would still work. The fact remains that they weren't compiling on a 64-bit platform, tho - see above.
The C language spec makes no guarantee for the exact range of an int.
This isn't what's at issue here. Maybe in some other thread, your comment has relevance, however, nobody in this thread lacks knowledge of what is and isn't guaranteed by the C language spec. Nobody really cares what you know about the C language spec, either.
I won't make the uninformed assumption that you've never been to a programming competition or that you don't know what its like to sit for hours trying to fix something that doesn't end up being a problem with your code ... but it would certainly seem that way. You're pounding issues that are highly irrelevant.
as always - my $0.02
Jobe