I don't think anyone commenting truly understands the possibilities that will be available when 3 dimensional screens
To say nothing of the physiological impact. In the human eye, rods outnumber cones by a huge margin; but we rely more on cones when focusing on bright 2D surfaces, like a computer screen. Therefore, some physicians believe that prolonged and repeated computer work disadvantages the eye over time because of the underutilization of rod cells. I can't confirm that specific theory, but I certainly believe that our eyes will find some relief in holographic displays.
Because children's books have enjoyed most of these book-related innovations, it's easy to overlook the evolution of "book technology" and the ways in which we're bridging the gap between digital media and the printed page.
Nevertheless, BlackMagic still looks like a View-Master, which will prevent some people from seeing it as a serious innovation. I wonder what it will take for this or (more likely) a different technology to be accepted eventually as a hardware standard by textbook publishers, fine art books, etc.
To put this into a broader context, we've already seen numerous proprietary technologies for making children's books interactive; we also have companion CD-ROMs, online rich media supplements, audiobook alternatives for an increasing number of titles, books bundled with audio recordings, and telephone book reading services offered by libraries. Most of these technologies "liberate" the text by adding sound, while only the multimedia supplements liberate illustrations. Therefore I appreciate BlackMagic's achievement, which, like LeapFrog's LeapPad, localizes the enhancements--as opposed to the CD-ROM (et al) that are inherently detached from the book itself.
While this bill is mad and deserves serious criticism, read it before debating it. A lot of the concerns voiced about this piece of legislation are nullified by the bill's actual stipulations. By the same token, if Schwarzenegger passed this thing without further revision, then it's actually even more restrictive in some ways than the articles indicate. Unfortuantely, the press releases just aren't detailed enough; and, personally, I don't know where to find more substantive information.
For years now we've been hearing that increased production volumes and market competition would drive down prices (oh, like the compact disc market?). Needless to say that the price drops have been less than phenomenal.
As a funny aside, a recent episode of the hilarious Scottish TV comedy Still Game had Winston shopping for LCD televisions. After finding that they cost about £2000 more than the £80 he expected, he mail-orders one of those DIY projection TV kits, which summarily sets his living room on fire.
That's because Novell has withstood the onslaught from Microsoft and still managed to eke out a survival. The folks at Novell know how to fight back against Microsoft.
While I agree that Novell has proven their tenacity in withstanding Microsoft in the past, I think Microsoft is considering Novell's market position as a whole. Novell has a level of experience, an infrastructure, and a market position with which no other distribution vendor can compete--and now they have an excellent, well-established product.
Novell has a customer base and a positive reputation developed over the course of more than 20 years. Can Mandrakesoft or any other distribution vendor claim that? Even more importantly, the other vendors have already competed and, apparently, failed to impress Microsoft--from Mandrake's quasi-Chapter 11 to Red Hat's disavowal of the desktop market. (If IBM acquired Turbolinux or a robust Debian-like distro, they might get Microsoft's attention, too.)
Beer has a lot more to offer than just antioxidants...
So does wine, but the respective benefits of each aren't all shared; so it's rather misleading for the author to proclaim "Beer Found to be as Healthy as Wine". A popular example of wine's healthfulness is the French paradox, well known among oenophiles.
I use AIM, Yahoo Messenger, and Jabber (all via Trillian) and Jabber is by far the most featureless of the three. Where's video or audio IM, buddy icons, etc. ?
What's being touted here is that Jabber can function as groupware for intranet messaging, and that it is a viable option for IM in the enterprise. But even for Internet messaging, it has a very attractive feature: encryption.
Trillian Pro offers 128-bit Blowfish encryption for ICQ, but that requires both parties to use Trillian Pro (a non-free Windows client).
On the other hand, Jabber is more readily extensible, and already offers both SSL and the somewhat SSH-like JEP-0116 encryption scheme.
I know some international business people that have already adopted Jabber for its privacy features. Jabber may never take hold among teens and "tweens", but it has a chance among other groups, such as Internet professionals who (believe it or not) conduct business via instant messaging.
I wanna know what they're smoking if they think that media drives don't need error correcting capabilities.
I think it's safe to assume that they're only talking about read errors, not write errors. Also, the extent is a deemphasis on error correction, not an elimination of it. The point is that playing audio or video streams is a comparatively fault-tolerant, non-mission critical, task.
Contrary to the baseless assumptions by the previous reply from 'marcus', I'm sure that the SATA-IO engineers have enough sense to determine a fault tolerance threshold that won't result in perceptible or substantial degradation of playback quality, possibly taking into account the limits of software-level interpolation.
Those bus types would have to be bridged to ATA anyway, so their presence would be superfluous.
What's more, the CE-SATA standard isn't just about power reduction and connector size, but customizing SATA drives for the unique needs of personal multimedia players. From the article: "The proposed specification could reduce drives' emphasis on correcting errors, which matters much more for banking applications than for serving up video pixels...".
Even MORE devices with unreliable hard drives? Oh my God....
At the risk of feeding a troll, is this supposed to be a slam against SATA, miniature hard drives, or the 2.5" drives present in a very few early personal audio/multimedia players? What's the problem?
I assume that by "use smaller connectors", they meant that SATA is smaller than the connectors currently being used in mini hard drives.
I interpreted it differently. The article is about supporting miniature hard drives in consumer electronics devices. For that purpose, even SATA's connectors, small as they are, are rather large when you're trying to fit everything into a palm-sized device.
If "someone" recommended Red Hat eLearning, I assume your company is adopting Red Hat? At the outset eLearning might be a reasonable choice, but if you really want to understand Linux, you'll probably want to laern more than just the Red Hat way of doing things. Experiment with Fedora or Red Hat 9 at home; then, after a few months, test a distribution that doesn't rely centrally on RPM and you'll gain a new, edifying perspective.
The sooner we get an education system which does not teach religion or political or patriotic based material the better.
Norman Dodd (former Congressional Investigator of Tax-Exempt Foundations) has delivered a considerable body of testimony on how such agenda came to be introduced into the American public education system.
An excerpt from an interview that touches on this subject follows:
"... their interest shifts over to preventing what they call a reversion of life in the United States to what it was prior to 1914, when World War I broke out.... they come to the conclusion that, to prevent a reversion, we must control education in the United States....
"They then decide that the key to the success... lay in the alteration of the teaching of American History. So, they approach four of the then most prominent teachers of American History in the country -- people like Charles and Mary Byrd. Their suggestion to them is this: Will they alter the manner in which they present their subject? And they get turned down, flatly....
"So, they then decide that it is necessary for them to... "build [their] own stable of historians." Then, they approach the Guggenheim Foundation, which specializes in fellowships, and say, "When we find young men in the process of studying for doctorates in the field of American History, and we feel that they are the right caliber, will you grant them fellowships on our say-so? And the answer is, "Yes."
"So, under that condition, eventually they assemble twenty... potential teachers of American History to London. There, they are briefed in what is expected of them -- when, as, and if they secure appointments in keeping with the doctorates they will have earned.
"That group of twenty historians ultimately becomes the nucleus of the American Historical Association. And then, toward the end of the 1920s, the Endowment grants to the American Historical Association $400,000 for a study of our history in a manner which points to what this country look forward to in the future.
"That culminates in a seven-volume study, the last volume of which is, of course, in essence, a summary of the contents of the other six. The essence of the last volume is this: the future of this country belongs to collectivism, administered with characteristic American efficiency."
Maybe you noticed that Dodd mentions "a study of our history in a manner which points to what this country look forward to in the future." This in itself is very interesting. Not only was American historiography reinvented, but the desire to situation students' lives on a concordant timeline that includes both rewritten history and a future context indicates the "apocalyptic" paradigm described by Frank Kermode.
The other option is Openoffice. It has a database form and report mode, but you will need MySQL or PostgreSQL at the backend. Also, there is almost no documentation for it.
Now why would you need MySQL or PostgreSQL when both OpenOffice and FileMaker Pro supports ODBC/JDBC?
Continuing with FM7 and bridging it to applications that staff know and use (such as Excel) is, to my mind, the most appropriate solution. The original poster complained about the growing problem of data lock-in, so I assume s/he isn't relying on ODBC or JDBC at present, and would benefit from this easy-to-implement approach.
The trouble here, really, is how do we handle legitimate email from banks, ISPs, etc?
As someone asserted earlier, Web Caller-ID is probably not the solution for email validation. Don't email header analysis, WHOIS (where applicable), and traceroute provide enough leads? Perhaps a software tool that automates and abstracts this process would be more appropriate. (Or perhaps Web Caller-ID provides this functionality?)
Web Caller-ID would seem to be particularly useful when used to deconstruct sites that don't attempt to emulate a brand-name website, since the benchmark in such cases isn't as obvious. Although the article emphasizes spoofs of established sites such as Citibank, eBay, etc., anyone adept enough to use Web Caller-ID effectively should be able to spot spoofs of such sites without the aid of ad hoc software.
It's just an example to show by comparison the severity of the problem. In other words, even if the average number of prompts is less than 25% of the maximum number of possible prompts (a relatively small proportion), it's still way too much--regardless of the nature of the required input. I'd like to say more about it, but I'll reserve comment until after I've tried the final build for myself.
If you scan through some of those screens, you'll discover that the reason there are so many is that you can take branches in the installer... A fair number of the screenshots are also screenshots of progress bars, which are noninteractive
I realize this, and all of these points were implicit in my post. However, this is still a catastrophe. The fact remains that users with certain configurations must navigate a huge number of screens in order to complete the installation. Extremely few users preconfigure unattended installations. But let's consider that the average number of interactive prompts is 50. That is still far too many.
Mods and flamers get ready: I'm about to criticize Debian (even though it's my favorite distro).
The fact that there are 231 screenshots of the new installer should raise some flags. 231!! Excluding a handful of error screens and progress bars, that suggests that in some circumstances the user would have to field more than 200 interactive prompts during the installation process. I should hope that many of these can either be consolidated or eliminated.
I had high hopes (too high) about the new hardware detection; I would be happy if these kinds of prompts disappear from the final build. You know the kind... the ones that require either clairvoyance, a second computer for hardware research, or the degree of advance preparation that only the IRS would demand.
That's what A+, CNE, etc are for. Pay money, pass a test that actually represents your level of expertise in the field, and get a time-limited certification showing competance. What we have here though is just an obvious money grab.
Not quite (for the case in point). However, for quite some time now CompTIA has offered the HTI+ Certification program, which seems to be the most relevant IT cert currently available. Whether HTI+ ("Home Technology Integrator") is a good fit for the Louisiana Radio and Television Technicians Board's rather vague requirements is difficult to say.
I worked for an audiovisual services company for about three years. Although a certificate from a sound engineering school carries some clout at such companies, they tend to rely heavily on their internal training programs, which are seriously lacking.
I think consumers would be well-served by a third-party A/V certification program that acknowledges IT skills. As A/V technology continues to merge with computer technology, basic accreditation will become even more of a necessity (hence HTI+). For example, the corporate training program at my former company didn't involve any real IT instruction whatsoever, and I was the only staff member with any IT skills--which were called upon frequently.
"Our goal is to make the Linux desktop a first class citizen and we think today's releases are a good first step in that direction."
While few would contest that RealNetworks' GPL'd contribution is a boon, it's too bad that their announcement was so self-congratulatory that it tacitly insulted the Linux community, and belittled the work of everyone who has ever positively contributed to it; especially since those collective efforts dwarf RealNetworks' late contribution.
The earlier versions of their FinePix line used Smartmedia and now the newer ones use xD cards, so they must know what they're doing. Fuji makes awesome cameras for not too much $$$ also!
xD-Picture Card was developed by Fujifilm and Olympus, and for years there's been evidence of SmartMedia phasing out of the market.
After reading a lot of reviews, I bought a FinePix A105 at a low price. While shopping around, however, I counted the xD format as a strike against the FinePix because xD is expensive and isn't widely supported--not even by most multi-card readers. It's good to know that xD performs well, but it still makes me feel somewhat like a Betamax user. Ergonomically, I think Memory Stick (PRO) is what appeals to people most, both for its design and the way in which it's installed/removed; but both xD and Memory Stick demonstrate a conflict of interests: the card size that human hands like best is greater than the card size optimal for installation in digital cameras.
Worsening the ergonomics issue, some cameras place the xD slot inconveniently behind a door on the bottom of the camera; so if the camera is mounted on a tripod or other stand, you have to detach it before you can access the xD door.
Although the author of the article declares that "the system can automatically recover wide-angle views of what people are looking at" (emphasis mine), to me one of the most exciting potential applications is to further human understanding of what animals choose to look at.
With our current knowledge of ocular biology we can make some assertions about what color ranges different species can see, but being able to study more precisely what they choose to focus on and what conditions attract their attention would advance our understanding of other species tremendously.
Photoshop doesn't tell you that when you save your file - you have to check the colour depth and change yourself. These same people also don't seem to use the post-creation compression tools to get the real compression benefit that can be had with PNG.
The Save for Web feature in the Creative Suite versions of Photoshop, ImageReady, and Illustrator makes these controls readily available, including manual palettization, etc. It's true that older versions of these products had either little or no support for PNG-8, but CS users shouldn't need external compression tools.
Are you compressing your PNGs with pngout or pngcrush? (pngout usually works much better) What kind of colour depth are we talking here?
ImageMagick and Photoshop CS, but I've only compared results in Photoshop CS. The results are consistent up to 32 colors. It's inconclusive, but for me LZW seems to require less overhead for files 104px and 32 colors.
At this point, external optimizers might be superfluous after Photoshop's web optimization. Adobe seems to like to improve on standard algorithms whenever possible (compare its noise-based color reduction to similar standard algorithms). I don't know whether or not Adobe's PNG support is directly derived from zlib/libpng, but zlib/libpng isn't credited in the PNG module's "About" box.
Slashdot Mirror
I don't think anyone commenting truly understands the possibilities that will be available when 3 dimensional screens
To say nothing of the physiological impact. In the human eye, rods outnumber cones by a huge margin; but we rely more on cones when focusing on bright 2D surfaces, like a computer screen. Therefore, some physicians believe that prolonged and repeated computer work disadvantages the eye over time because of the underutilization of rod cells. I can't confirm that specific theory, but I certainly believe that our eyes will find some relief in holographic displays.
Because children's books have enjoyed most of these book-related innovations, it's easy to overlook the evolution of "book technology" and the ways in which we're bridging the gap between digital media and the printed page.
Nevertheless, BlackMagic still looks like a View-Master, which will prevent some people from seeing it as a serious innovation. I wonder what it will take for this or (more likely) a different technology to be accepted eventually as a hardware standard by textbook publishers, fine art books, etc.
To put this into a broader context, we've already seen numerous proprietary technologies for making children's books interactive; we also have companion CD-ROMs, online rich media supplements, audiobook alternatives for an increasing number of titles, books bundled with audio recordings, and telephone book reading services offered by libraries. Most of these technologies "liberate" the text by adding sound, while only the multimedia supplements liberate illustrations. Therefore I appreciate BlackMagic's achievement, which, like LeapFrog's LeapPad, localizes the enhancements--as opposed to the CD-ROM (et al) that are inherently detached from the book itself.
SB 1506
While this bill is mad and deserves serious criticism, read it before debating it. A lot of the concerns voiced about this piece of legislation are nullified by the bill's actual stipulations. By the same token, if Schwarzenegger passed this thing without further revision, then it's actually even more restrictive in some ways than the articles indicate. Unfortuantely, the press releases just aren't detailed enough; and, personally, I don't know where to find more substantive information.
For years now we've been hearing that increased production volumes and market competition would drive down prices (oh, like the compact disc market?). Needless to say that the price drops have been less than phenomenal.
As a funny aside, a recent episode of the hilarious Scottish TV comedy Still Game had Winston shopping for LCD televisions. After finding that they cost about £2000 more than the £80 he expected, he mail-orders one of those DIY projection TV kits, which summarily sets his living room on fire.
That's because Novell has withstood the onslaught from Microsoft and still managed to eke out a survival. The folks at Novell know how to fight back against Microsoft.
While I agree that Novell has proven their tenacity in withstanding Microsoft in the past, I think Microsoft is considering Novell's market position as a whole. Novell has a level of experience, an infrastructure, and a market position with which no other distribution vendor can compete--and now they have an excellent, well-established product.
Novell has a customer base and a positive reputation developed over the course of more than 20 years. Can Mandrakesoft or any other distribution vendor claim that? Even more importantly, the other vendors have already competed and, apparently, failed to impress Microsoft--from Mandrake's quasi-Chapter 11 to Red Hat's disavowal of the desktop market. (If IBM acquired Turbolinux or a robust Debian-like distro, they might get Microsoft's attention, too.)
Beer has a lot more to offer than just antioxidants...
So does wine, but the respective benefits of each aren't all shared; so it's rather misleading for the author to proclaim "Beer Found to be as Healthy as Wine". A popular example of wine's healthfulness is the French paradox, well known among oenophiles.
I use AIM, Yahoo Messenger, and Jabber (all via Trillian) and Jabber is by far the most featureless of the three. Where's video or audio IM, buddy icons, etc. ?
What's being touted here is that Jabber can function as groupware for intranet messaging, and that it is a viable option for IM in the enterprise. But even for Internet messaging, it has a very attractive feature: encryption.
Trillian Pro offers 128-bit Blowfish encryption for ICQ, but that requires both parties to use Trillian Pro (a non-free Windows client).
On the other hand, Jabber is more readily extensible, and already offers both SSL and the somewhat SSH-like JEP-0116 encryption scheme.
I know some international business people that have already adopted Jabber for its privacy features. Jabber may never take hold among teens and "tweens", but it has a chance among other groups, such as Internet professionals who (believe it or not) conduct business via instant messaging.
I wanna know what they're smoking if they think that media drives don't need error correcting capabilities.
I think it's safe to assume that they're only talking about read errors, not write errors. Also, the extent is a deemphasis on error correction, not an elimination of it. The point is that playing audio or video streams is a comparatively fault-tolerant, non-mission critical, task.
Contrary to the baseless assumptions by the previous reply from 'marcus', I'm sure that the SATA-IO engineers have enough sense to determine a fault tolerance threshold that won't result in perceptible or substantial degradation of playback quality, possibly taking into account the limits of software-level interpolation.
Who has heard of USB 2?
Firewire?
Both are plenty fast.
Both have small connectors.
Those bus types would have to be bridged to ATA anyway, so their presence would be superfluous.
What's more, the CE-SATA standard isn't just about power reduction and connector size, but customizing SATA drives for the unique needs of personal multimedia players. From the article: "The proposed specification could reduce drives' emphasis on correcting errors, which matters much more for banking applications than for serving up video pixels...".
Even MORE devices with unreliable hard drives? Oh my God....
At the risk of feeding a troll, is this supposed to be a slam against SATA, miniature hard drives, or the 2.5" drives present in a very few early personal audio/multimedia players? What's the problem?
I assume that by "use smaller connectors", they meant that SATA is smaller than the connectors currently being used in mini hard drives.
I interpreted it differently. The article is about supporting miniature hard drives in consumer electronics devices. For that purpose, even SATA's connectors, small as they are, are rather large when you're trying to fit everything into a palm-sized device.
If "someone" recommended Red Hat eLearning, I assume your company is adopting Red Hat? At the outset eLearning might be a reasonable choice, but if you really want to understand Linux, you'll probably want to laern more than just the Red Hat way of doing things. Experiment with Fedora or Red Hat 9 at home; then, after a few months, test a distribution that doesn't rely centrally on RPM and you'll gain a new, edifying perspective.
And this is different from any of his other star trek appearances how?
20% fewer spasms
Norman Dodd (former Congressional Investigator of Tax-Exempt Foundations) has delivered a considerable body of testimony on how such agenda came to be introduced into the American public education system.
An excerpt from an interview that touches on this subject follows:
Maybe you noticed that Dodd mentions "a study of our history in a manner which points to what this country look forward to in the future." This in itself is very interesting. Not only was American historiography reinvented, but the desire to situation students' lives on a concordant timeline that includes both rewritten history and a future context indicates the "apocalyptic" paradigm described by Frank Kermode.
The other option is Openoffice. It has a database form and report mode, but you will need MySQL or PostgreSQL at the backend. Also, there is almost no documentation for it.
Now why would you need MySQL or PostgreSQL when both OpenOffice and FileMaker Pro supports ODBC/JDBC?
Continuing with FM7 and bridging it to applications that staff know and use (such as Excel) is, to my mind, the most appropriate solution. The original poster complained about the growing problem of data lock-in, so I assume s/he isn't relying on ODBC or JDBC at present, and would benefit from this easy-to-implement approach.
The trouble here, really, is how do we handle legitimate email from banks, ISPs, etc?
As someone asserted earlier, Web Caller-ID is probably not the solution for email validation. Don't email header analysis, WHOIS (where applicable), and traceroute provide enough leads? Perhaps a software tool that automates and abstracts this process would be more appropriate. (Or perhaps Web Caller-ID provides this functionality?)
Web Caller-ID would seem to be particularly useful when used to deconstruct sites that don't attempt to emulate a brand-name website, since the benchmark in such cases isn't as obvious. Although the article emphasizes spoofs of established sites such as Citibank, eBay, etc., anyone adept enough to use Web Caller-ID effectively should be able to spot spoofs of such sites without the aid of ad hoc software.
It's just an example to show by comparison the severity of the problem. In other words, even if the average number of prompts is less than 25% of the maximum number of possible prompts (a relatively small proportion), it's still way too much--regardless of the nature of the required input. I'd like to say more about it, but I'll reserve comment until after I've tried the final build for myself.
If you scan through some of those screens, you'll discover that the reason there are so many is that you can take branches in the installer ...
A fair number of the screenshots are also screenshots of progress bars, which are noninteractive
I realize this, and all of these points were implicit in my post. However, this is still a catastrophe. The fact remains that users with certain configurations must navigate a huge number of screens in order to complete the installation. Extremely few users preconfigure unattended installations. But let's consider that the average number of interactive prompts is 50. That is still far too many.
Mods and flamers get ready: I'm about to criticize Debian (even though it's my favorite distro).
The fact that there are 231 screenshots of the new installer should raise some flags. 231!! Excluding a handful of error screens and progress bars, that suggests that in some circumstances the user would have to field more than 200 interactive prompts during the installation process. I should hope that many of these can either be consolidated or eliminated.
I had high hopes (too high) about the new hardware detection; I would be happy if these kinds of prompts disappear from the final build. You know the kind... the ones that require either clairvoyance, a second computer for hardware research, or the degree of advance preparation that only the IRS would demand.
That's what A+, CNE, etc are for. Pay money, pass a test that actually represents your level of expertise in the field, and get a time-limited certification showing competance. What we have here though is just an obvious money grab.
Not quite (for the case in point). However, for quite some time now CompTIA has offered the HTI+ Certification program, which seems to be the most relevant IT cert currently available. Whether HTI+ ("Home Technology Integrator") is a good fit for the Louisiana Radio and Television Technicians Board's rather vague requirements is difficult to say.
I worked for an audiovisual services company for about three years. Although a certificate from a sound engineering school carries some clout at such companies, they tend to rely heavily on their internal training programs, which are seriously lacking.
I think consumers would be well-served by a third-party A/V certification program that acknowledges IT skills. As A/V technology continues to merge with computer technology, basic accreditation will become even more of a necessity (hence HTI+). For example, the corporate training program at my former company didn't involve any real IT instruction whatsoever, and I was the only staff member with any IT skills--which were called upon frequently.
"Our goal is to make the Linux desktop a first class citizen and we think today's releases are a good first step in that direction."
While few would contest that RealNetworks' GPL'd contribution is a boon, it's too bad that their announcement was so self-congratulatory that it tacitly insulted the Linux community, and belittled the work of everyone who has ever positively contributed to it; especially since those collective efforts dwarf RealNetworks' late contribution.
The earlier versions of their FinePix line used Smartmedia and now the newer ones use xD cards, so they must know what they're doing. Fuji makes awesome cameras for not too much $$$ also!
xD-Picture Card was developed by Fujifilm and Olympus, and for years there's been evidence of SmartMedia phasing out of the market.
After reading a lot of reviews, I bought a FinePix A105 at a low price. While shopping around, however, I counted the xD format as a strike against the FinePix because xD is expensive and isn't widely supported--not even by most multi-card readers. It's good to know that xD performs well, but it still makes me feel somewhat like a Betamax user. Ergonomically, I think Memory Stick (PRO) is what appeals to people most, both for its design and the way in which it's installed/removed; but both xD and Memory Stick demonstrate a conflict of interests: the card size that human hands like best is greater than the card size optimal for installation in digital cameras.
Worsening the ergonomics issue, some cameras place the xD slot inconveniently behind a door on the bottom of the camera; so if the camera is mounted on a tripod or other stand, you have to detach it before you can access the xD door.
Although the author of the article declares that "the system can automatically recover wide-angle views of what people are looking at" (emphasis mine), to me one of the most exciting potential applications is to further human understanding of what animals choose to look at.
With our current knowledge of ocular biology we can make some assertions about what color ranges different species can see, but being able to study more precisely what they choose to focus on and what conditions attract their attention would advance our understanding of other species tremendously.
Photoshop doesn't tell you that when you save your file - you have to check the colour depth and change yourself. These same people also don't seem to use the post-creation compression tools to get the real compression benefit that can be had with PNG.
The Save for Web feature in the Creative Suite versions of Photoshop, ImageReady, and Illustrator makes these controls readily available, including manual palettization, etc. It's true that older versions of these products had either little or no support for PNG-8, but CS users shouldn't need external compression tools.
Are you compressing your PNGs with pngout or pngcrush? (pngout usually works much better) What kind of colour depth are we talking here?
ImageMagick and Photoshop CS, but I've only compared results in Photoshop CS. The results are consistent up to 32 colors. It's inconclusive, but for me LZW seems to require less overhead for files 104px and 32 colors.
At this point, external optimizers might be superfluous after Photoshop's web optimization. Adobe seems to like to improve on standard algorithms whenever possible (compare its noise-based color reduction to similar standard algorithms). I don't know whether or not Adobe's PNG support is directly derived from zlib/libpng, but zlib/libpng isn't credited in the PNG module's "About" box.
Yeah, may I be one of the first to say "YEEEEEHAWWW!"
Quantum computing in general is one of the most exciting technological developments ever. I love reading about progress in this field.