What AOL has to consider is its 34million users turning round and saying "the latest version of AOL is broke", if it's not rendering IE specific content correctly.
They did this once before, though. If I remember correctly, when AOL made the switch to IE, Netscape had a strong majority of the browser market and IE was still of the very poor quality that you can expect from early Microsoft releases. Websites were targetted to Netscape at the time, there were pages that didn't render right in IE, and yet AOL made the switch.
The other thing to consider is that Mozilla's rendering is downright excellent these days. I haven't had any problems rendering sites with it for a long time now. Are users really going to be getting a lot of error messages after the switch? And even if they do, why would they blame AOL? Years of Windows use has conditioned people to expect errors all the time which they can't do anything about so they shrug their shoulders and move on.
AOL has been testing Mozilla with their Compuserve users for awhile now and the tests have reportedly gone well. I don't think this is a bluff.
This would have been appropriate for the Slashback section as well, but it was probably published too recently to have made it in, so I will bring it up...
"Bother," said the Borg. "We've assimilated Pooh."
Ah, now there's a way to defeat the Borg that was never tried on Star Trek... offer up Pooh as bait and then after his consciousness has been assimilated, point the Borg at the HoneyNet project.
Why? You may have traveled faster than the speed of sound, but it still took you a lot longer to reach Boston than your "I'm leaving" took to travel within San Francisco.
I don't get what you're saying. In the example, I didn't "reach" Boston - Boston was the starting point (San Francisco was the end point). If I were travelling faster than the speed of sound, how is it that my sound "I'm leaving" would get there first when I am, by definition, travelling faster than it? I wasn't suggesting that the sound is sent across a phone or sent via radio waves, I'm talking about somebody observing to the actual, original sound.
It can be shown by relatively basic relativity that, if for one observer, event B occurs after event A but separated by less than the time it would take light to travel from A to B, then there is some observer for whom the time-ordering of A & B is reversed. That is, for some observer moving at constant velocity relative to the first, B occurs first.
Couldn't this same logic be used to prove that nothing can move faster than the speed of sound? Say I hop in my supersonic jet, shout "I'm leaving", fly from Boston to San Francisco, and then say "I'm here". Somebody standing in San Francisco will hear me say "I'm here" before they hear "I'm leaving". Following the same argument you used, this should make faster than sound travel impossible because the person standing in San Francisco will observe B before A even though A happened before B. Of course, we all know that supersonic travel is possible, so this shows that observations of occurrences do not need to follow chronological order.
I think the puny processor power would make that impossible. A Tivo runs on a 50MHz PowerPC, and I don't think it ever is involved in any audio or video processing.
"Impossible" is a pretty strong word. I'd bet that a few months of assembly code optimization could produce software fast enough. However, that is an awful lot of development time to spend on a single feature, which is why waiting for hardware support may be a better option.
The other option is that this doesn't need to be done in real time. What if you could instruct your TiVo to work on time-compressing certain shows whenever its CPU is idle? It then becomes irrelevant if the software can't time-compress the MPEG stream in real time. It will throw out whatever frames it can before you start watching so that you can potentially get the full advantages of the time-compression we're talking about, but if you start watching before it's done with the compression it can just let you watch the uncompressed version.
I'm curious now - what does the TiVo use its CPU for when it would otherwise be idle? It seems like it could be put to good use on some sort of space/time optimization feature like this.
I've suggested this too, and been told that it's a hardware problem. The MPEG chips Tivo uses can't deliver sound in any other speed than normal.
What's to stop them from using software to rewrite the MPEG stream before it hits the MPEG decoder chip? Yes, that would probably take a substantial development effort on TiVo's part, but it should be possible. The trick is to get enough people to ask for it so that TiVo sees it as worth the hassle.
This is only vaguely on topic, but what I don't understand is why no PVR maker offers this feature
Yes! This would be an excellent feature. Please request this from TiVo - they are asking for feature suggestions. I requested this very feature a few months ago, and if enough people chime in with the same request it might just catch their attention.
To answer your question, my guess is that no PVRs offer this feature simply because PVRs have only been around for a relatively short amount of time and they just haven't had enough time to add all the features that somebody would want yet.
I just follow my policy as with all spam: Do nothing. Delete it. Don't click on links, don't ask to be removed, and don't waste time complaining to someone. I just delete it.
That's what I do too, but the one thing I do above and beyond this is to set up a filter to automatically delete mail from the sender's domain or address if I receive more than one piece of spam from the same sender. If everybody started doing this, then spammers would have a huge disincentive to send spam because they would no longer be able to send the recipient any mail at all (which they may eventually have a legitimate need to do). Yes, they could change their "From" address, but if they are attempting to masquerade as a legitimate business (you know, the types that preface their messages with "This is not spam" and have a presumably stable web presence), they will have a much harder time defending themselves in court if there is a record of their addresses jumping around. This obviously doesn't apply to all spammers, but it applies to enough for my "Trash" folder to have been automatically filled by 830 pieces of unread mail over the last three months or so.
So we've seen a steady a drop of programmers per computer from 10 to.0001 in the past 40 years, a factor of 100,000 or a bit slower than moores law.
That works out to a decline of programmers per computer of around 25% (compounded) per year, or a halving every 2.4 years - not quite as dramatic as Moore's Law, which would be every 1.5 years. To use the original author's date of 2015, projecting out this rate of decline would mean that the number of programmers per computer should be roughly 2.37% of what it is today. Of course, this doesn't necessarily mean that the absolute number of programmers will fall that far or even at all. If we have 42 computers in 2015 for every one that we have today, then that would require the same number of programmers as today (if the number of programmers per computer continues to decline at the same rate). Who knows how many things will have embedded computers in 2015 - in any case, I seriously doubt that there will be less computers in 2015, so even holding the number of computers constant and thereby using 2.37% as a lower bound, the number of programmers left in 2015 would still not be as dire as the original poster predicted.
HOWEVER. What I want to see is something that will store this information, possibly in a db or somesuch, and then spit it out easily into multiple formats.
I did this with my own resume awhile back (http://www.twmacinta.com/resume/resume_gen.php). Initially, my thinking was close to yours and I thought about making a generalized resume system that reads from a db. What I did, instead, was to store all of my resume information as string literals (inside of arrays) within a PHP program. This is obviously not something you would want to do if you were writing a generic resume generation program for others to use, but since I am the only user of the code itself I opted for this route because it took very little time to develop. I would recommend this route to other developers because updating your resume data is pretty easy (just add another string to the array) and supporting new formats is just a matter of writing some PHP to transform the arrays into that format. All of the formats on my page are automatically generated except for Word and PDF, for which I have written intermediate HTML formats that can be very easily converted into Word and PDF by hand.
What weird timing... I just finished watching a Nova documentary on the gamma ray bursts just an hour or two ago. They entitled the episode Death Star and PBS has set up a website for it at http://www.pbs.org/wgbh/nova/gamma/. The short of it is that scientists believe that these bursts are the result of stars about 30 times the size of our sun collapsing (and forming black holes in the process). They referred to this as a "hyper-nova" as distinguished from the smaller "super-nova". The process gives off a tremendous amount of energy - the most of any process that we know of since the big bang, according to the Nova narrator.
Anyway, PBS tends to re-run Nova episodes quite a bit where I live, so check your local listings - you might be able to catch it again real soon if you missed it the first time.
I'm sure helped to tarnish the use of MP3s as nothing but piracy
How exactly did he do that? I'd argue just the opposite - MP3.com helped legitimize the use of MP3's for indie artists wishing to publish their own music (that's not piracy). Their other service, Beam-It, was what they were sued over and that wasn't piracy either - all parties involved had paid for the CDs of the music they listened to, so who was the pirate?
You may very well be right that Robertson likes to play up the David vs. Goliath thing in media stunts, but I have to disagree with you when you imply that MP3.com was a bad thing. It was (and still is) a great boon for indie artists and music lovers in general, and I'm optimistic that Lindows can be equally benefical to Linux (though I admit their odds of success seem rather far out).
and now he's on to Linux's trying to make a quick buck from some VC suckers.
For as long as I can remember, MP3.com has supported Linux as a desktop client for their service. My point is that I don't think that Robertson's interest in Linux has recently materialized just to make a quick buck. Of course, I could be wrong, but even if I am that's not necessarily a bad thing. He reportedly registered the domain name mp3.com before knowing what mp3s were (after noticing that 'mp3' was a frequently searched for term), and like I said before, I think MP3.com has had a net positive effect on music.
Oh man... did anybody else notice what was in the bottom screenshot posted at lindows.com? Outlook is open to an email with the subject "Microsoft Warns of 'Critical' Security Hole in XP". The body of the email takes up a good chunk of the screenshot.
Michael Robertson did this with MP3.com too - he subtly taunted the big record labels (and they eventually sued as well). While I wouldn't want to be in his shoes, I'm glad he's doing what he does because somebody needs to be a thorn in the side if the RIAA and Microsoft and he has done it in a very entertaining way.
I'd still like to have site-by-site preferences wihtout having to edit the prefs.js file, but, what can you do?
You could use this handly little preferences toolbar. You can leave pop-ups disabled in general and then when you come across a site that you actually need pop-ups for, simply un-check the checkbox. And don't let the screenshot fool you - it allows you to very quickly turn on/off more than just the 4 preferences you see there (right clicking on the toolbar will give you a big selections of what checkboxes should appear).
It asks for it regardless of whether is declared or not.
Repeating this will not make it true. Your statement goes squarely against what the Mozilla release notes imply. So, thinking that maybe this is undocumented or implemented improperly I decided to test it out for myself. I tested it by using Mozilla 0.9.6 to look at my local copy of Apache (http://192.168.55.111/ for me) which does not have any pages with the necessary <LINK> tag. The results? There were no requests for favicon.ico as you have stated that there should be. Please indicate where you are getting your information from that Mozilla always loads favicon.ico because this contridicts both the Mozilla release notes and direct testing. Here are all my log entries from today (note the lack of ".ico" files):
on every first page visit to a site it requests favicon.ico
Take a look at the release notes for 0.9.6. They say that to define an icon for a page you need to use the <LINK REL="icon"> tag in your document and I don't see anything that would indicate that Mozilla will be requesting icons automatically as you have implied. To me, it sounds like Mozilla will only request an icon if the page defines the <LINK> tag (which would indicate that's what the author intended) or if the user bookmarks a page (coming in 0.9.7). That seems like a pretty non-intrusive way to handle things and doesn't sound like it will skew stats at all.
Suppose for a moment that Disney has been convicted of lacing their movies with illegal, subliminal messages to trick viewers into purchasing Disney products. Suppose that to make ammends they offer to donate $1B worth of "educational" videos to schools but that these "educational" videos also contain the subliminal messages. Would you support the Disney "settlement" in this case? Sure kids may learn a little more with the new videos, but as a side effect the original problem of subliminal messages not only persists but is actually made much worse through the expansion of their audience into these schools. Now replace "Disney" with "Microsoft" and "subliminal messages" with "anti-competitive behaviour" and you have the situation with Microsoft.
This would be a nightmare for me with my "age detector" website because of all the people who would ask for refunds. Even if I opt out the the program and keep my pages free, I'm sure I would still get a ton of email from people asking for their penny back. I already get a ton of email from irate "customers" despite the fact that I'm not selling anything. I get around 20,000 visitors a day and if I had to deal with refund requests for even a small percentage of that it would become a serious burden. I imagine that there are plenty of other sites out there who would want to keep there pages free but would be forced to step up "customer service" if users started assuming that sites are pay per view by default.
I'm curious about the statement that on that page that all Universal CDs will have copy protection as of this month. Are they not bothering to list individual Universal CDs because of this blanket statement? Does this include CDs produced under another label and distributed by Universal?
I ask because I don't see the new Garbage on the list and this CD is distributed by Universal even though the label it's produced under is something different. I walked into the store last week with the intention to buy it until I saw "Universal" on the back, which lost them a sale. I'd still like to buy it if the CD isn't actually corrupted, so does anybody out there have this CD? Does it have the intentional corruption on it?
I don't think the question asked was what percentage of your budget is going to software licenses.
CNet wrote:
Goldman Sachs... asked about highest and lowest spending priorities.
and
Areas like supply-chain management software and Linux servers rank near the bottom of spending priorities.
It sure sounds to me like they asked about the budget and based their conclusions on that. This definitely biases the conclusions in favor of higher priced products because they inherently require a higher budget.
Slashdot Mirror
They did this once before, though. If I remember correctly, when AOL made the switch to IE, Netscape had a strong majority of the browser market and IE was still of the very poor quality that you can expect from early Microsoft releases. Websites were targetted to Netscape at the time, there were pages that didn't render right in IE, and yet AOL made the switch.
The other thing to consider is that Mozilla's rendering is downright excellent these days. I haven't had any problems rendering sites with it for a long time now. Are users really going to be getting a lot of error messages after the switch? And even if they do, why would they blame AOL? Years of Windows use has conditioned people to expect errors all the time which they can't do anything about so they shrug their shoulders and move on.
AOL has been testing Mozilla with their Compuserve users for awhile now and the tests have reportedly gone well. I don't think this is a bluff.
Ah, now there's a way to defeat the Borg that was never tried on Star Trek... offer up Pooh as bait and then after his consciousness has been assimilated, point the Borg at the HoneyNet project.
I don't get what you're saying. In the example, I didn't "reach" Boston - Boston was the starting point (San Francisco was the end point). If I were travelling faster than the speed of sound, how is it that my sound "I'm leaving" would get there first when I am, by definition, travelling faster than it? I wasn't suggesting that the sound is sent across a phone or sent via radio waves, I'm talking about somebody observing to the actual, original sound.
Couldn't this same logic be used to prove that nothing can move faster than the speed of sound? Say I hop in my supersonic jet, shout "I'm leaving", fly from Boston to San Francisco, and then say "I'm here". Somebody standing in San Francisco will hear me say "I'm here" before they hear "I'm leaving". Following the same argument you used, this should make faster than sound travel impossible because the person standing in San Francisco will observe B before A even though A happened before B. Of course, we all know that supersonic travel is possible, so this shows that observations of occurrences do not need to follow chronological order.
"Impossible" is a pretty strong word. I'd bet that a few months of assembly code optimization could produce software fast enough. However, that is an awful lot of development time to spend on a single feature, which is why waiting for hardware support may be a better option.
The other option is that this doesn't need to be done in real time. What if you could instruct your TiVo to work on time-compressing certain shows whenever its CPU is idle? It then becomes irrelevant if the software can't time-compress the MPEG stream in real time. It will throw out whatever frames it can before you start watching so that you can potentially get the full advantages of the time-compression we're talking about, but if you start watching before it's done with the compression it can just let you watch the uncompressed version.
I'm curious now - what does the TiVo use its CPU for when it would otherwise be idle? It seems like it could be put to good use on some sort of space/time optimization feature like this.
What's to stop them from using software to rewrite the MPEG stream before it hits the MPEG decoder chip? Yes, that would probably take a substantial development effort on TiVo's part, but it should be possible. The trick is to get enough people to ask for it so that TiVo sees it as worth the hassle.
Yes! This would be an excellent feature. Please request this from TiVo - they are asking for feature suggestions. I requested this very feature a few months ago, and if enough people chime in with the same request it might just catch their attention.
To answer your question, my guess is that no PVRs offer this feature simply because PVRs have only been around for a relatively short amount of time and they just haven't had enough time to add all the features that somebody would want yet.
That's what I do too, but the one thing I do above and beyond this is to set up a filter to automatically delete mail from the sender's domain or address if I receive more than one piece of spam from the same sender. If everybody started doing this, then spammers would have a huge disincentive to send spam because they would no longer be able to send the recipient any mail at all (which they may eventually have a legitimate need to do). Yes, they could change their "From" address, but if they are attempting to masquerade as a legitimate business (you know, the types that preface their messages with "This is not spam" and have a presumably stable web presence), they will have a much harder time defending themselves in court if there is a record of their addresses jumping around. This obviously doesn't apply to all spammers, but it applies to enough for my "Trash" folder to have been automatically filled by 830 pieces of unread mail over the last three months or so.
That works out to a decline of programmers per computer of around 25% (compounded) per year, or a halving every 2.4 years - not quite as dramatic as Moore's Law, which would be every 1.5 years. To use the original author's date of 2015, projecting out this rate of decline would mean that the number of programmers per computer should be roughly 2.37% of what it is today. Of course, this doesn't necessarily mean that the absolute number of programmers will fall that far or even at all. If we have 42 computers in 2015 for every one that we have today, then that would require the same number of programmers as today (if the number of programmers per computer continues to decline at the same rate). Who knows how many things will have embedded computers in 2015 - in any case, I seriously doubt that there will be less computers in 2015, so even holding the number of computers constant and thereby using 2.37% as a lower bound, the number of programmers left in 2015 would still not be as dire as the original poster predicted.
I did this with my own resume awhile back (http://www.twmacinta.com/resume/resume_gen.php). Initially, my thinking was close to yours and I thought about making a generalized resume system that reads from a db. What I did, instead, was to store all of my resume information as string literals (inside of arrays) within a PHP program. This is obviously not something you would want to do if you were writing a generic resume generation program for others to use, but since I am the only user of the code itself I opted for this route because it took very little time to develop. I would recommend this route to other developers because updating your resume data is pretty easy (just add another string to the array) and supporting new formats is just a matter of writing some PHP to transform the arrays into that format. All of the formats on my page are automatically generated except for Word and PDF, for which I have written intermediate HTML formats that can be very easily converted into Word and PDF by hand.
Anyway, PBS tends to re-run Nova episodes quite a bit where I live, so check your local listings - you might be able to catch it again real soon if you missed it the first time.
Microsoft got caught ages ago with its hand in the cookie jar doing exactly that with the Barkto indcident.
How exactly did he do that? I'd argue just the opposite - MP3.com helped legitimize the use of MP3's for indie artists wishing to publish their own music (that's not piracy). Their other service, Beam-It, was what they were sued over and that wasn't piracy either - all parties involved had paid for the CDs of the music they listened to, so who was the pirate?
You may very well be right that Robertson likes to play up the David vs. Goliath thing in media stunts, but I have to disagree with you when you imply that MP3.com was a bad thing. It was (and still is) a great boon for indie artists and music lovers in general, and I'm optimistic that Lindows can be equally benefical to Linux (though I admit their odds of success seem rather far out).
For as long as I can remember, MP3.com has supported Linux as a desktop client for their service. My point is that I don't think that Robertson's interest in Linux has recently materialized just to make a quick buck. Of course, I could be wrong, but even if I am that's not necessarily a bad thing. He reportedly registered the domain name mp3.com before knowing what mp3s were (after noticing that 'mp3' was a frequently searched for term), and like I said before, I think MP3.com has had a net positive effect on music.
Oh man... did anybody else notice what was in the bottom screenshot posted at lindows.com? Outlook is open to an email with the subject "Microsoft Warns of 'Critical' Security Hole in XP". The body of the email takes up a good chunk of the screenshot.
Michael Robertson did this with MP3.com too - he subtly taunted the big record labels (and they eventually sued as well). While I wouldn't want to be in his shoes, I'm glad he's doing what he does because somebody needs to be a thorn in the side if the RIAA and Microsoft and he has done it in a very entertaining way.
Repeating this will not make it true. Your statement goes squarely against what the Mozilla release notes imply. So, thinking that maybe this is undocumented or implemented improperly I decided to test it out for myself. I tested it by using Mozilla 0.9.6 to look at my local copy of Apache (http://192.168.55.111/ for me) which does not have any pages with the necessary <LINK> tag. The results? There were no requests for favicon.ico as you have stated that there should be. Please indicate where you are getting your information from that Mozilla always loads favicon.ico because this contridicts both the Mozilla release notes and direct testing. Here are all my log entries from today (note the lack of ".ico" files):
192.168.55.111 - - [21/Nov/2001:09:44:05 -0500] "GET / HTTP/1.1" 200 4716 "-" "Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.6) Gecko/20011120" /icons/back.gif HTTP/1.1" 200 216 "http://192.168.55.111/" "Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.6) Gecko/20011120" /icons/blank.gif HTTP/1.1" 200 148 "http://192.168.55.111/" "Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.6) Gecko/20011120" /icons/folder.gif HTTP/1.1" 200 225 "http://192.168.55.111/" "Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.6) Gecko/20011120" /icons/unknown.gif HTTP/1.1" 200 245 "http://192.168.55.111/" "Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.6) Gecko/20011120" /icons/tar.gif HTTP/1.1" 200 219 "http://192.168.55.111/" "Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.6) Gecko/20011120" /icons/text.gif HTTP/1.1" 200 229 "http://192.168.55.111/" "Mozilla/5.0 (X11; U; Linux i686; en-US; rv:0.9.6) Gecko/20011120"
192.168.55.111 - - [21/Nov/2001:09:44:06 -0500] "GET
192.168.55.111 - - [21/Nov/2001:09:44:06 -0500] "GET
192.168.55.111 - - [21/Nov/2001:09:44:06 -0500] "GET
192.168.55.111 - - [21/Nov/2001:09:44:06 -0500] "GET
192.168.55.111 - - [21/Nov/2001:09:44:06 -0500] "GET
192.168.55.111 - - [21/Nov/2001:09:44:06 -0500] "GET
Take a look at the release notes for 0.9.6. They say that to define an icon for a page you need to use the <LINK REL="icon"> tag in your document and I don't see anything that would indicate that Mozilla will be requesting icons automatically as you have implied. To me, it sounds like Mozilla will only request an icon if the page defines the <LINK> tag (which would indicate that's what the author intended) or if the user bookmarks a page (coming in 0.9.7). That seems like a pretty non-intrusive way to handle things and doesn't sound like it will skew stats at all.
Suppose for a moment that Disney has been convicted of lacing their movies with illegal, subliminal messages to trick viewers into purchasing Disney products. Suppose that to make ammends they offer to donate $1B worth of "educational" videos to schools but that these "educational" videos also contain the subliminal messages. Would you support the Disney "settlement" in this case? Sure kids may learn a little more with the new videos, but as a side effect the original problem of subliminal messages not only persists but is actually made much worse through the expansion of their audience into these schools. Now replace "Disney" with "Microsoft" and "subliminal messages" with "anti-competitive behaviour" and you have the situation with Microsoft.
How about a t-shirt with a picture of an all-powerful attack-penguin on it? Come on, you know you want one.
I take it that you haven't installed XP yet.
This would be a nightmare for me with my "age detector" website because of all the people who would ask for refunds. Even if I opt out the the program and keep my pages free, I'm sure I would still get a ton of email from people asking for their penny back. I already get a ton of email from irate "customers" despite the fact that I'm not selling anything. I get around 20,000 visitors a day and if I had to deal with refund requests for even a small percentage of that it would become a serious burden. I imagine that there are plenty of other sites out there who would want to keep there pages free but would be forced to step up "customer service" if users started assuming that sites are pay per view by default.
I ask because I don't see the new Garbage on the list and this CD is distributed by Universal even though the label it's produced under is something different. I walked into the store last week with the intention to buy it until I saw "Universal" on the back, which lost them a sale. I'd still like to buy it if the CD isn't actually corrupted, so does anybody out there have this CD? Does it have the intentional corruption on it?
If they can't come to a resolution maybe they could take a page from the Emacs/XEmacs playbook and fork the storyline.
I'm sorry, I know that was bad but I couldn't resist.
CNet wrote:
and
It sure sounds to me like they asked about the budget and based their conclusions on that. This definitely biases the conclusions in favor of higher priced products because they inherently require a higher budget.