Arguably he's done it to boost the public's impression of him, but at least he's done something real good to do it unlike latching on to divisive issues. He's punished so many crooked corps: Wall Street, insurance, payola, and now the music companies again.
Here's a profile: Wikipedia Profile
The consumers will lose a great public advocate when he goes on to become Governor (although one hopes he'll use his clout there to do even more reform.)
There's already a revert war in progress between some user Jim Hall, who wants to title a section Downing Street Memo: The Left Questions Bush's Intentions on Buildup to War in Iraq) and another user Alainbloch who wants to title it Downing Street Memo: An Intent to Deceive?.
The changelog contains comments such as "removing falsehoods" and "adding fair commentary".
I predict this is going to devolve into an all-out edit war from between both "liberals" and "conservatives" with too much time on their hands -- like what happened to the George Bush and John Kerry entries on Wikipedia during last year's election, but magnified by a hundred times because of the LA Times' higher visibility.
Wikis are useless for contentious topics meant to be reader-editable; each reader will inevitably inject his own bias into the article -- some will even try to eliminate what they perceive as existing bias. On Wikipedia, say in an article about the history of Macs, a fanboy may end up putting in too much of Steve Jobs' bio...relatively harmless. Here, in the Colosseum of the LA Times Wikitorial, gladiators from the left and right will battle it out over "falsehoods" and "fair and balanced"...to no end whatsoever.
Is this high-bitrate, CD-quality audio that will be streaming? If not, then this scheme will have limited effect, particularly among the non-Britney listener crowd. Besides, unless the university has Napster servers onsite or something, or maybe uses bandwidth shaping to give the Napster streams the highest priority, with people downloading other stuff all the time, the stream will probably be interrupted from time to time. To me, and to others too I'm sure, there is nothing more annoying than a stream that breaks up...even if it's only once every 10 songs.
Also, what about those who'd prefer to use their own "system" to listen to their music? This covers the gamut from those using alternative OS's to those who simply prefer a particular player (Winamp, Foobar2000, etc.). If this is a Windows-only, WMP/Proprietary Player-only scheme, it definitely isn't going to be all that popular.
Lastly, what about portables? Can you put one copy of a song on a portable of your choice?
There's too many imponderables with this scheme, and if it's typically restricted streaming (which I think it'll be, with Napster the source), then the best this thing can hope to be is a very fast preview for songs that people will want to buy/download.
As someone who has been using catchalls for more than 2 years, I feel it's worth it, couple with a good client-side spam filter (Thunderbird works for me). I use a slightly different scheme that works against spammers who typically target only the first-level domain (i.e., mydomain.*):
There's my main email for friends and family: me@mydomain.net
There's my secondary email for less important personal uses: me@mercury.mydomain.net
There's the catchall, *@mercury.mydomain.net, which I use to hand out customized addresses to commercial sites, both so I can easily sort the important ones (Airline discount emails, etc.), and so I can track any lying bastards that sell me out.
Thunderbird has rules corresponding to the above 3, and the rest of the catchall email goes into the catchall folder. If I'm expecting something, from a signup for example, I'll quickly check the catchall folder. Otherwise, I check it about once a day.
In general, I haven't had wildcard spam creep into mercury. I guess that's because spammers don't generally bother with wildcarding subdomains. Also, mercury doesn't do http, meaning it's not generally visible on the web.
http://forums.speedlabs.org/ (membership required) has a series of mods/steps to let your 401S/411S write DVD+Rs (and -Rs when they arrive) at 8X, effectively converting them to a 8xxS drive. Drives thus converted can later be flashed with original 8xxS firmware. (401s must first be "overclocked" to 411s and from then to 8xx).
Right, so this ancient study says that when overloaded, tube distortion *sounds* more pleasing than transistor distortion. Well, the true test of a device is under normal operating conditions. Next we'll have something that says tapes sound better than CD's, because tapes have a softer clipping threshold than digital devices (which, for CDs, clip at +32767 or -32768).
Surely no audiophile is going to buy a tube amp because he wants it to sound better when overloaded...for the same amount, or generally quite less, he could get a digital amp with enough power to NOT overload at the given level. But then, some of these (no offense to the reasonable ones) are the people who believe gold connectors "reduce distortion" and "improve sound"...for SPDIF connectors!!
This has the potential to become another non-conventional music outlet like iTMS, but only if they do it right.
The "NPR-not-MTV listener" they are catering to will have widely varying music tastes, not just the Top 40. How much of a selection will each Starbucks provide? Do they plan to have T1 linkups to a central server? If they work with local storage, then the source tracks will probably be already compressed tracks, affecting quality. I don't see each Starbucks having a half-terabyte RAID array to hold losslessly compressed originals.
Secondly, price. This can be a one-stop-music-shop, catering not just to those who see it and burn/buy a CD on a whim. Since it doesn't offer any of the advantages of iTMS-style music downloads (instant transfer to computers, portables, etc.), they better price it at less than $0.99 a track. A fixed-price option, e.g. 1 80-minute CD for $12-$15 might be very popular.
It's upto Starbucks to use its enormous geographical clout to negotiate a sweetheart deal with the recording industry, and make it as attractive to the customer as possible. Otherwise, with audio-CD only Discmans going the way of the dodo, and the growing popularity of iTMS-like solutions, this scheme will turn out at best to be a novelty.
The problem is that "automated comparisons" don't mimic human system responses (the ear, or the eye for video). Take video: the eye would finds grainy VHS tape more pleasing than a digital video that displayed some blocking. The blocked digital video, mathematically, is much closer to the original than the the VHS with its added noise...
These types of psychovisual (or psychoacoustic) responses are what make automated tools almost useless for judging the perceived quality of any lossy encoder. Perceived, that's the key word....it may not be mathematically up to scratch with the original, but if you PERCEIVE it to be as good as the original, thats what matters (this is of course for CD-quality high bitrate tests).
Theoretically, yes. A lossy encoder (like MP3) works by throwing away parts of the original audio which a normal human wouldn't hear, using something called a psychoacoustic model. Re-encoding such lossy material, with ANOTHER lossy encoder, may result in the new encoder throwing away useful stuff that you could hear with the original.
Three scenarios where the perceptible quality probably wouldn't change:
1. Low bitrate MP3 -> High bitrate (256 kbps ATRAC3 was "CD-quality" according to the Japanese review in an earlier post). Bloating 128 kbps MP3s to 256 kbps ATRAC will probably sound transparent.
2. Very high bitrate MP3 -> ATRAC3. If your source is a 256/320 kbps MP3, ATRAC3 should make a reasonably transparent conversion. Of course, if you use some shitty bitrate for ATRAC3, it'll show up in the results.
3. Garbage In, Garbage Out: If your source content is crap (Britney Spears, et al), you probably won't notice that the ATRAC3 version is somewhat crappier.
Of course, 256 kbps is VERY high for "CD-quality" lossy encoding. LAME (MP3 encoder) achieves this very nicely for most music using VBR (variable bit rate) at around 192 kbps. Newer encoders such as Ogg Vorbis and AAC (MP4) do this at even lower bitrates, around 140-160 kbps.
Sony could easily have designed an ATRAC3 container for MP3s, instead of this whole re-encoding rigmarole. The container could add whatever DRM and other crap they wanted to force on the user, maybe even encypting the original MP3 or something. Any portable that does NOT natively support MP3s (and soon, AAC/MP4s) is still-born. Sony are shooting themselves in the foot, as has become usual with their increasingly proprietary electronic devices.
Yeah sure. As if the MPAA makes RAID arrays of uncompressed movie footage available on demand. Face it, most material IS going to be encoded from pre-encoded material -- the two most common today would be DVDs and digital camcorders. The only exception for home or SOHO users might be recording stuff from a TV tuner card, but if your source is analog TV, there are worse links in the chain than your choice of MPEG-4 encoder.
Who would a raw film test with MPEG-4 codecs help out? I'm not aware of any entity that uses MPEG-4 as a "first" encoder for raw data -- MPEG-2 is king, and with the advent of HDTV and Blu-ray (super sized) DVDs, is expected to stay so for the near future.
WHOIS authenticity is a moot point; if law enforcement really wants to know who's behind a site, they can just subpoena the hosting provider (which can obviously be found from reverse-DNSing the site IP or just looking at the DNS records).
This is just another shill to give pseudo-law-enforcement's (read: **AA) teeth more bite. If some site is really peddling material they claim is copyrighted, they should just DMCA the hosting provider and then go through the courts to subpoena the provider and get the identity of the site operator. After all, isn't that the purpose of the DMCA?
With *TeX, it's what I call a true WYCIWYG (what you code is what you get) program; unlike Word which isn't a true WYSIWYG program --- formatting pictures/charts/etc. and positioning them is a pain in any Word document beyond 50 pages. Apart from the inevitable memory-hog slowdown that people mentioned; I've "edited" a 300+ page "magazine" that was full of charts/pictures and found the whole frame idea an increasing pain. I too had to do it in 20-page sections.
LaTex is much more structured; and to be honest, if you've ever done any sort of programming, it's a dead ringer for use in making any large, multipage document. And it's free, open-source,... all that goodness.
If you will check out http://www.oxenhielm.com/ , Elin produces screenshots of three emails from Zhou, where Zhou talks to her about the paper, congratulates her on publication, and a third. Elin also states that it was Zhou who suggested she try the method she used to try to solve the relevant part of Hilbert's Problem.
I think it's just Zhou trying to cover her ass after she didn't pay enough attention to what Elin was doing, whether it be out of neglect or ignorance.
The Forebearers Club Thanks You...
on
Snail Mail Tech
·
· Score: 0, Offtopic
We believe that humans evolved from four bears, our forebearers, a couple of generations ago. Thank you for your kind mention. Now if only you would do it more often...
For the uninitiated, Cooper Union in the parent list is a full-scholarship school (AFAIK, the only one in the nation, apart from the military academies, but then again, Cooper doesn't bind you to anything after you graduate). As long as you have your own housing and commute (in NYC), all you pay is about ~$1k/year in student/lab fees. Very small (~450 engineering students). I'll be a sophomore there, and have had fun so far (in the academic sense; social life sucks). Take a look if you want to go to a reputed school but don't want to spend big bucks (or are not in CA:-). Cooper offered explicit EE, ChemE, CivE, MechE, GeneralE majors until this year, now they offer an integrated BE program with concentrations in different areas.
People from the "lower" castes (Dalits is the PC term, with Scheduled/Backward Castes the official term) are a sizeable chunk of the electorate and are wooed actively and encouraged to vote for and by political parties that claim to have their best interests at heart, and that often have party officials of similar castes. For example, see this brief profile of the The Bahujan Samaj Party, a major player in the field in Central/Northern India.
Guess my univ isn't "upright" then. ALL incoming traffic is blocked, hell, even pings (ICMP) is. Outbound on all well-known P2P ports is also blocked. Guess they just can't afford packet-shaping.
But yes, IMHO the best solution is via port/packet-based bandwidth limiting.
That's the official line from the CUCC. Considering I've got http dl's @ 400 kbytes/s on the dorm network (which feeds of the main pipe), I'd say it's atleast two T1s, most probably three.
Slashdot Mirror
Arguably he's done it to boost the public's impression of him, but at least he's done something real good to do it unlike latching on to divisive issues. He's punished so many crooked corps: Wall Street, insurance, payola, and now the music companies again. Here's a profile: Wikipedia Profile The consumers will lose a great public advocate when he goes on to become Governor (although one hopes he'll use his clout there to do even more reform.)
Don't forget, Wall Street comes under his jurisdiction. That's why he's gone after so many crooked corporations.
There's already a revert war in progress between some user Jim Hall, who wants to title a section Downing Street Memo: The Left Questions Bush's Intentions on Buildup to War in Iraq) and another user Alainbloch who wants to title it Downing Street Memo: An Intent to Deceive?. The changelog contains comments such as "removing falsehoods" and "adding fair commentary". I predict this is going to devolve into an all-out edit war from between both "liberals" and "conservatives" with too much time on their hands -- like what happened to the George Bush and John Kerry entries on Wikipedia during last year's election, but magnified by a hundred times because of the LA Times' higher visibility. Wikis are useless for contentious topics meant to be reader-editable; each reader will inevitably inject his own bias into the article -- some will even try to eliminate what they perceive as existing bias. On Wikipedia, say in an article about the history of Macs, a fanboy may end up putting in too much of Steve Jobs' bio...relatively harmless. Here, in the Colosseum of the LA Times Wikitorial, gladiators from the left and right will battle it out over "falsehoods" and "fair and balanced"...to no end whatsoever.
...does it run Linux?
Is this high-bitrate, CD-quality audio that will be streaming? If not, then this scheme will have limited effect, particularly among the non-Britney listener crowd. Besides, unless the university has Napster servers onsite or something, or maybe uses bandwidth shaping to give the Napster streams the highest priority, with people downloading other stuff all the time, the stream will probably be interrupted from time to time. To me, and to others too I'm sure, there is nothing more annoying than a stream that breaks up...even if it's only once every 10 songs.
Also, what about those who'd prefer to use their own "system" to listen to their music? This covers the gamut from those using alternative OS's to those who simply prefer a particular player (Winamp, Foobar2000, etc.). If this is a Windows-only, WMP/Proprietary Player-only scheme, it definitely isn't going to be all that popular.
Lastly, what about portables? Can you put one copy of a song on a portable of your choice?
There's too many imponderables with this scheme, and if it's typically restricted streaming (which I think it'll be, with Napster the source), then the best this thing can hope to be is a very fast preview for songs that people will want to buy/download.
As someone who has been using catchalls for more than 2 years, I feel it's worth it, couple with a good client-side spam filter (Thunderbird works for me). I use a slightly different scheme that works against spammers who typically target only the first-level domain (i.e., mydomain.*):
There's my main email for friends and family: me@mydomain.net
There's my secondary email for less important personal uses: me@mercury.mydomain.net
There's the catchall, *@mercury.mydomain.net, which I use to hand out customized addresses to commercial sites, both so I can easily sort the important ones (Airline discount emails, etc.), and so I can track any lying bastards that sell me out.
Thunderbird has rules corresponding to the above 3, and the rest of the catchall email goes into the catchall folder. If I'm expecting something, from a signup for example, I'll quickly check the catchall folder. Otherwise, I check it about once a day.
In general, I haven't had wildcard spam creep into mercury. I guess that's because spammers don't generally bother with wildcarding subdomains. Also, mercury doesn't do http, meaning it's not generally visible on the web.
http://forums.speedlabs.org/ (membership required) has a series of mods/steps to let your 401S/411S write DVD+Rs (and -Rs when they arrive) at 8X, effectively converting them to a 8xxS drive. Drives thus converted can later be flashed with original 8xxS firmware. (401s must first be "overclocked" to 411s and from then to 8xx).
Right, so this ancient study says that when overloaded, tube distortion *sounds* more pleasing than transistor distortion. Well, the true test of a device is under normal operating conditions. Next we'll have something that says tapes sound better than CD's, because tapes have a softer clipping threshold than digital devices (which, for CDs, clip at +32767 or -32768).
Surely no audiophile is going to buy a tube amp because he wants it to sound better when overloaded...for the same amount, or generally quite less, he could get a digital amp with enough power to NOT overload at the given level. But then, some of these (no offense to the reasonable ones) are the people who believe gold connectors "reduce distortion" and "improve sound"...for SPDIF connectors!!
This has the potential to become another non-conventional music outlet like iTMS, but only if they do it right.
The "NPR-not-MTV listener" they are catering to will have widely varying music tastes, not just the Top 40. How much of a selection will each Starbucks provide? Do they plan to have T1 linkups to a central server? If they work with local storage, then the source tracks will probably be already compressed tracks, affecting quality. I don't see each Starbucks having a half-terabyte RAID array to hold losslessly compressed originals.
Secondly, price. This can be a one-stop-music-shop, catering not just to those who see it and burn/buy a CD on a whim. Since it doesn't offer any of the advantages of iTMS-style music downloads (instant transfer to computers, portables, etc.), they better price it at less than $0.99 a track. A fixed-price option, e.g. 1 80-minute CD for $12-$15 might be very popular.
It's upto Starbucks to use its enormous geographical clout to negotiate a sweetheart deal with the recording industry, and make it as attractive to the customer as possible. Otherwise, with audio-CD only Discmans going the way of the dodo, and the growing popularity of iTMS-like solutions, this scheme will turn out at best to be a novelty.
The problem is that "automated comparisons" don't mimic human system responses (the ear, or the eye for video). Take video: the eye would finds grainy VHS tape more pleasing than a digital video that displayed some blocking. The blocked digital video, mathematically, is much closer to the original than the the VHS with its added noise...
These types of psychovisual (or psychoacoustic) responses are what make automated tools almost useless for judging the perceived quality of any lossy encoder. Perceived, that's the key word....it may not be mathematically up to scratch with the original, but if you PERCEIVE it to be as good as the original, thats what matters (this is of course for CD-quality high bitrate tests).
Theoretically, yes. A lossy encoder (like MP3) works by throwing away parts of the original audio which a normal human wouldn't hear, using something called a psychoacoustic model. Re-encoding such lossy material, with ANOTHER lossy encoder, may result in the new encoder throwing away useful stuff that you could hear with the original. Three scenarios where the perceptible quality probably wouldn't change: 1. Low bitrate MP3 -> High bitrate (256 kbps ATRAC3 was "CD-quality" according to the Japanese review in an earlier post). Bloating 128 kbps MP3s to 256 kbps ATRAC will probably sound transparent. 2. Very high bitrate MP3 -> ATRAC3. If your source is a 256/320 kbps MP3, ATRAC3 should make a reasonably transparent conversion. Of course, if you use some shitty bitrate for ATRAC3, it'll show up in the results. 3. Garbage In, Garbage Out: If your source content is crap (Britney Spears, et al), you probably won't notice that the ATRAC3 version is somewhat crappier. Of course, 256 kbps is VERY high for "CD-quality" lossy encoding. LAME (MP3 encoder) achieves this very nicely for most music using VBR (variable bit rate) at around 192 kbps. Newer encoders such as Ogg Vorbis and AAC (MP4) do this at even lower bitrates, around 140-160 kbps. Sony could easily have designed an ATRAC3 container for MP3s, instead of this whole re-encoding rigmarole. The container could add whatever DRM and other crap they wanted to force on the user, maybe even encypting the original MP3 or something. Any portable that does NOT natively support MP3s (and soon, AAC/MP4s) is still-born. Sony are shooting themselves in the foot, as has become usual with their increasingly proprietary electronic devices.
Yeah sure. As if the MPAA makes RAID arrays of uncompressed movie footage available on demand. Face it, most material IS going to be encoded from pre-encoded material -- the two most common today would be DVDs and digital camcorders. The only exception for home or SOHO users might be recording stuff from a TV tuner card, but if your source is analog TV, there are worse links in the chain than your choice of MPEG-4 encoder.
Who would a raw film test with MPEG-4 codecs help out? I'm not aware of any entity that uses MPEG-4 as a "first" encoder for raw data -- MPEG-2 is king, and with the advent of HDTV and Blu-ray (super sized) DVDs, is expected to stay so for the near future.
WHOIS authenticity is a moot point; if law enforcement really wants to know who's behind a site, they can just subpoena the hosting provider (which can obviously be found from reverse-DNSing the site IP or just looking at the DNS records).
This is just another shill to give pseudo-law-enforcement's (read: **AA) teeth more bite. If some site is really peddling material they claim is copyrighted, they should just DMCA the hosting provider and then go through the courts to subpoena the provider and get the identity of the site operator. After all, isn't that the purpose of the DMCA?
With *TeX, it's what I call a true WYCIWYG (what you code is what you get) program; unlike Word which isn't a true WYSIWYG program --- formatting pictures/charts/etc. and positioning them is a pain in any Word document beyond 50 pages. Apart from the inevitable memory-hog slowdown that people mentioned; I've "edited" a 300+ page "magazine" that was full of charts/pictures and found the whole frame idea an increasing pain. I too had to do it in 20-page sections.
LaTex is much more structured; and to be honest, if you've ever done any sort of programming, it's a dead ringer for use in making any large, multipage document. And it's free, open-source,... all that goodness.
I'm from Rhode Island, you insensitive clod!
If you will check out http://www.oxenhielm.com/ , Elin produces screenshots of three emails from Zhou, where Zhou talks to her about the paper, congratulates her on publication, and a third. Elin also states that it was Zhou who suggested she try the method she used to try to solve the relevant part of Hilbert's Problem.
I think it's just Zhou trying to cover her ass after she didn't pay enough attention to what Elin was doing, whether it be out of neglect or ignorance.
We believe that humans evolved from four bears, our forebearers, a couple of generations ago. Thank you for your kind mention. Now if only you would do it more often...
Or , this!
0 31 115/ids_photos_en/r3048724578.jpg
http://story.news.yahoo.com/news?tmpl=story&u=/
No, this is Qusay.
Note that Dad gave Uday some WMD brand "batteries" last Christmas, but he had to hide them...
Bruce Schneier founded/owns Counterpane....
For the uninitiated, Cooper Union in the parent list is a full-scholarship :-). Cooper offered explicit EE, ChemE, CivE, MechE, GeneralE majors until this year, now they offer an integrated BE program with concentrations in different areas.
school (AFAIK, the only one in the nation, apart from the military academies, but then again, Cooper doesn't bind you to anything after you graduate). As long as you have your own housing and commute (in NYC), all you pay is about ~$1k/year in student/lab fees. Very small (~450 engineering students). I'll be a sophomore there, and have had fun so far (in the academic sense; social life sucks). Take a look if you want to go to a reputed school but don't want to spend big bucks (or are not in CA
People from the "lower" castes (Dalits is the PC term, with Scheduled/Backward Castes the official term) are a sizeable chunk of the electorate and are wooed actively and encouraged to vote for and by political parties that claim to have their best interests at heart, and that often have party officials of similar castes. For example, see this brief profile of the The Bahujan Samaj Party, a major player in the field in Central/Northern India.
Guess my univ isn't "upright" then. ALL incoming traffic is blocked, hell, even pings (ICMP) is. Outbound on all well-known P2P ports is also blocked. Guess they just can't afford packet-shaping.
But yes, IMHO the best solution is via port/packet-based bandwidth limiting.
That's the official line from the CUCC. Considering I've got http dl's @ 400 kbytes/s on the dorm network (which feeds of the main pipe), I'd say it's atleast two T1s, most probably three.
True. My bad. Freshman. :)
Seriously, I got the ECHO (which is not UDP dependent) and the Tracert (which is UDP dependent) mixed up.