Yeah, works great for me. It re-writes caller-id based on the names in my Mac's address book, recognizes certain callers and uses distinctive ring to tell what's happening, blacklists other numbers. I'm using it to send long-distance calls out via VoIP while still sending local calls via the POTS line that I can't get rid of. Great fun. Details on my blog.
This venturing a bit far from where we started--the point I was making is that an advanced Bayes filter can (in at least some situations) do a better job then SpamAssassin, even though SA includes a wide variety of filtering tools. I'm not trying to claim that SA is useless or that tablesaws are the end-all-be-all of woodworking tools (I don't actually use my tablesaw all that much:-).
For initially seeding the bayes table, I can certainly see SA's other filters as useful. Personally, I have a few thousand test cases sitting around that I can use for Bayes trainging, but random users don't generally work that way (personal experience: they keep everything, unsorted, in one big inbox, never hit delete, and expect me to keep it backed up for them. And then complain that my mail server is slow accessing their inbox).
However, in a lot of cases, I'd rather use one focused tool then one that can be customized and tuned into doing the job. I don't generally use perl to copy files--I use cp. When filtering my own pile of spam, I'll stick to a simple Bayesean filter because it's much more effective for me and I don't have to spend any time tweaking the weights of different filters. It's been a while since I used SA, but balancing the different filters was a pain, and it was always generating false positives. I'm sure it's better now, but I'm just amazed at how well SpamProbe has worked for me--it's been at least 4 months since I saw a false positive, and I only rarely see spam in my inbox. It's just a nicely tuned, nicely balanced tool that doesn't take a lot of effort to do a good job. And *that*'s what I'm really looking for in tools.
Actually, that *is* my argument. I used to carry around a multi-tool with a billion little blades and driver tips. In the end, I realized that all I really ever used was the #1 Phillips screwdriver, and the one on the multitool sucked--it was too short to reach into cases, and too hard to pull out. I was better served with a simple, cheap screwdriver. Some times single-use tools are much more powerful then jack-of-all-trades tools. And sometimes the uber-tool wins. Complexity is the enemy, do what you can to simplify the tools that you use.
Yeah, but if the Bayes classifier is much more accurate then any of the other classifiers (that's been my experience), then why not just turn all of the other ones off and go solely by the Bayes filter. Averaging 'great' with a lot of 'okay's doesn't do a whole lot to improve the 'great'.
My personal problem with SA is that it's really just a muddled average of a bunch of guessed-at filters for recognizing spam. The individual filters aren't very accurate, but the idea is that the average across a bunch of filters will be more accurate then any individual filter.
Bayes-based filters, on the other hand, directly calculate the probability of specific words appearing in spam vs. non-spam messages. Newer versions calculate the probability of short phrases, HTML tags, and mail headers as well. There's no guesswork involved (unlike SA)--if you feed them enough of yesterday's spam, then they're going to be really good with today and tomorrow's spam. The spammers keep evolving, so sooner or later messages will get through, but the filters keep evolving, too, and it's really hard to beat a good filter these days.
I've been using SpamProbe for almost 6 months, and it's amazingly accurate. I haven't had a false positive in months, and I only see a couple false negatives per month.
Done properly, software RAID 5 can be as fast (or even faster) then hardware RAID 5. After all, in most cases, "hardware" RAID is really just software RAID on an embedded CPU. I've dealt with way too many hardware RAID cards that slow to an utter crawl when doing RAID 5. On the other hand, Linux's software RAID is pretty zippy; I've seen 200MB/sec on the right hardware.
There can be performance issues if you're short of CPU on your system, or if you're low on RAM, or if you're nearing the limits of your system bus, but none of those are usually the case on modern storage servers--you don't usually need to run them that close to the edge.
The one thing that hardware RAID can buy for you is a non-volatile cache; that lets you get away with caching writes, and just generally buys you better performance and consistency. One of these days, someone will add journalled RAID-5 to Linux, though, and then you'll be able to accomplish the same thing with a PCI-based battery-backed RAM card for cheap.
So there has been a standard for almost four years, but how many HAVi enabled electronics devices do you see down at your local Generic Big Box Electronics Store? Zero would be a reasonable estimate.
Nope, not zero. Most of Mitsubishi's big-screen TVs have HAVi. I think their HDTV VCR is HAVi also. You can wire them together with firewire and then your TV's remote can control the VCR (as well as send and receive video) over firewire. The Mitsu TVs actually want to know all about your A/V system, including which boxes are wired to which audio and video ports on which other boxes. Then, everything (supposedly) works right if you want to have your HDTV sat receiver feed video straight to the TV, but feed audio into your (non-HDTV-capable) receiver. As long as you use the TV's remote and on-screen display, switching to the satellite receiver will automatically flip the receiver to the right input and then display the video direct from the satellite.
It's actually a bit cooler then that -- I ordered one of the HAVi books used from someone on Amazon (only $5), and it's kind of an interesting spec. It really wants to stream MPEGs over Firewire, and have devices hand around Java applets for UIs. So, your HAVi TV and HAVi satellite receiver could (in an ideal world) work by having the satellite receiver send the TV an applet that contains all of the satellite guide functions, and then the TV runs the applet on-screen for you. When you tell the applet to change the channel, it sends a HAVi control message to the satellite box, telling the box to change channels and start sending a new MPEG channel.
Just reducing the number of remotes is a very small piece of HAVi.
Did you read the specs? It records 640x480 MP4 at 1.5 or 3 Mbps. Assuming they actually mean Mbps not MBps, that works out to around an hour of high-quality video. Not too bad, but it depends on the video quality.
But that's not all you're paying for. You're not counting the cost of storage and printing. DIY-enlargements work out to a couple bucks per 8x10 (about the same as having film enlarged to the same size). And what about ink longevity?
You have a lot of good points. I've spent about a hundred on film storage since I took up photography (negative/slide sheets, storage boxes, etc), and I've bought an extra 80 GB hard drive to store pictures from the digital camera. I have a 35mm film scanner, and I was using it to scan and print negatives before I got the D60, so a lot of the printing comparisons break down. Anyway, 90% of the time, I get prints by burning a CD and dropping it off at the local Costco with a Frontier, so the print cost and longevity are the same as film prints. I have a decent inkjet, and I use it occasionally, but you can't really compare it to conventional prints--I can print any size up to 13x19, and I can spend hours tweaking it to look the way I want it to look. 95% of the time, there's something wrong with machine prints, either from digital or film, but it's ususally too much of a pain to get it fixed. A DIY printing solution gives you a lot more control, at the cost of taking longer. So, I use both--if I'm not feeling super-critical, I let Costco handle it. Otherwise I do it myself.
Also by storage I don't just mean the cost of a single hard drive, regardless of size. You've got backups, transfers to other media, etc., to worry about. And 20 years from now my negatives will still be in the box on the bookshelf, available for prints and enlargements. Where will your photoshop files be?
Good points, and mostly looking for a good solution. Except I've had a hard time finding several pages of negatives for the last year, and I have a bunch of other negatives that I cheaped out and had Costco reprint (rather then scanning them and giving them the files), and now they're scratched. Film and Digital *both* have storage issues. They're just not the same issues:-).
It all depends on your definition of "nice film-based SLR." I was $3-4k into Canon film cameras before I bought my D60; I don't think that's uncommon--one lens now, a new flash later, then a new body, it all adds up over the years.
So, adding a $2,200 D60 wasn't a *huge* step, price-wise. I've had it around 6 months, and I've shot around 7,000 frames with it. Assuming for the moment that I'd have shot the same number of frames had I been using film, that averages out to $0.35/frame, which is in the same general range as film and processing (that's $10 for 36 exposures).
Assuming that I've got at least another couple years of functional use in the camera, the per-frame cost should drop down under a dime. Plus, I get instant feedback (nice when fiddling with lighting problems) and it's easier for me to sort, edit, and produce prints with digital then it is with film.
So, with six months of use, you can start to argue that it's paid for itself. Add another couple years of use, and it'll be hard to argue that it would have been cheaper to use film. So, even if it has no resale value in 3 years, it'll still have been a good move, financially speaking.
If you're setting up, say, a server, why on earth would you pay a penny more on graphics than you have to for basic diagnostics and BIOS access?
Because it's worth $20-$50 or so to have the video on-board. It takes up less space, has less impact on cooling, and is less likely to pop out during shipping.
That's pretty much it, though. Good 3D is utterly worthless for 99% of all server applications.
Where can my wife take a memory card and have someone extract 36 of the pictures and print them out for 8 bucks in one hour? It's not going to happen soon
You mean, besides Costco or Walmart? Several of the local Costcos have the ability to print from flash cards or CDs in their one hour lab. They charge $0.20/print. I've heard that Walmarts are similar, but my skin crawls every time I set foot in one, so I tend to avoid the experience.
IMHO, the only problem with digital cameras for 95% of the uses out there today is the up-front cost of the equipment (plus maybe the PC software to interface with it. Get a Mac with iPhoto:-). A cheap Nikon 2500 produces crappy 4x6s just about as well as most of the cheap P&S cameras out there. My D60 seems to produce better prints then I could usually get out of my EOS 3. Not always, but usually. Plus, with the D60, I can shoot as much as I want without worrying about the hideous cost of film and quality processing, and I can go anywhere and get prints without worrying about cheap labs scratching my negatives. If I want a really good print, then I can go to a good lab. If I want a pile of 4x6s for family, then I can stop by Costco on the way to work. That, combined with the instant feedback that you get from the display (hmm, the histogram looks overexposed, or the subject blinked for all four shots, or whatever) means that I'm taking better pictures.
The 1Ds looks like it's going to be quite a bit better then the D60 in pretty much every way, probably encroaching into MF 's territory in many ways due to its lower noise. Is it perfect? Of course not. Does it have higher resolution, defined in terms of the ability to resolve really small details in images? Nope. Does it matter for prints under 11x14 or so? Not really. Frankly, I wouldn't be suprised if the 1Ds produces 16x20-ish portraits that compare favorably to MF prints in the same size.
Is film dead? I don't know, is portrait painting dead? Did color film kill black and white film? I supposed it depends on how you look at things -- I wouldn't have a clue how to find a portrait painter, except maybe the yellow pages in a major city. There aren't a whole lot of them out there. Black and white is nice, and usually better then color for pictures of people, frankly, but try finding a 1 hour (or even 24-hour) lab that can give you a pile of 4x6 prints from a roll of real black and while film for under $20. Most likely, what we're going to see over the next five or so years is the gradual disappearance of most of the less-popular professional 35mm film, along with most of the slower (100 and maybe 200 speed) consumer film. Frankly, 100 speed consumer film doesn't make any sense anyway -- with decent film, you shouldn't be able to tell the difference between 100 and 400 film in a 4x6, and most P&S cameras don't let in enough light to use ISO 100 95% of the time. You'll probably be able to buy 400 or 800-speed consumer film in the checkout line of your favorite grocery store for another 10+ years. Specialty stuff is already disappearing, though. I'd be amazed if more then 1/3 of the film types available today are still available in 6 or 7 years, and very little new film is going to be introduced from now on out.
That was with ext3. Reiser is even worse on large partitions, from what I've heard. Something about it having performance issues with filesystems over 1TB.
The point is that you can build a 2+TB system for well under $10k, using 160GB IDE drives and 3ware cards. I have 5 of them, and I've actually had problems -- my first partitioning attempt gave me a 2.06 TB RAID, which mke2fs decided was only 60 GB:-(.
The next round of storage servers that I buy will probably be even bigger, and it'd be nice to be able to use them as one big partition. Pity that I'll have to wait for 2.6 for that.
The reason that this announcement doesn't make sense is because it leaves out the non-Intel half of the Xeon server platform -- the ServerWorks "Grand Champion" chipset.
Go look at the specs for this beast -- it supports something like 6 PCI-X (133 MHz/64-bit) PCI *busses* and quad-channel DDR memory, with a 6.4 GB/sec memory bandwidth. They've been demoing it at conferences since early this year.
So, given that, the P4 makes a lot more sense in a server. Frankly, I would order one *now* if they were available, because I have a project that really needs to be able to manage multiple gigabit Ethernet pipes, and none of the existing x86 solutions (4x P3 Xeon, 2x Athlon, 2x Tualatin, 2x P4 w/ i860) really have the bus bandwidth that I'd like. They'll probably all work, but the P4 Xeon with the serverworks chipset would probably scale twice as far due to faster busses and faster memory.
Don't say "if you need that kind of performance, don't use x86." We're in the middle of switching to Linux/x86 from Solaris/Sparc for this, because low-end (under $50k) Sun hardware just isn't competetive with high-end PCs.
More seriously, a lot of routers are limited to 128 MB or 256 MB of RAM. Right now, 128 MB is barely enough for a full routing table under IOS. Scale forward, and you'll see that 256 MB is only good for a year or two, tops.
I agree with your comment, but your math is *WAY* off:-). Using your numbers, here's what I get:
DVD - $19 for 5.2GB = $3.65/GB IDE - $238 for 60GB = $3.97/GB CDR - $38 for 64GB = $0.59/GB
Of course, that ignores drive costs for the removable media. It also assumes that ~60 GB of data online with an IDE drive is comperable to 60 GB of data offline on a bunch of CD-R or DVD-RAM discs. It usually isn't. IMHO, the best mix for cost-sensitive users is probably a big IDE drive or two and a couple boxes of CDs for storing things that you don't need access to more then once or twice per month.
PDF isn't really that bad -- Adobe released the complete specs to it when they first released Acrobat. It's completely possible to write your own PDF reader/writer. Like xpdf and Ghostscript:-).
That's not to say that other formats aren't better, but PDF is way better then Word.doc files, which tend to be seen as the real alternative by a lot of companies.
Re:A bad book by Banks' standards
on
Inversions
·
· Score: 1
I hated Song of Stone, and wasn't very fond of The Wasp Factory, but thought that Whit was fascinating. As others have mentioned, The Crow Road and Espedair Street are probably his best non-SF books. Of his SF books, Player of Games and Excession are my favorites.
I guess I'm just jealous because I didn't have the cashish to buy all these great domains back in '93 when i thought about it.:)
Domains were still free in '93. If you'd have registered them then, you wouldn't have had to pay anything until 95-97, when NSI started billing for legacy domains.
Also, the seller has been involved with Linux for a *long* time. He was the second maintainer of the Linux networking code, after Ross Biro (who wrote the first implmentation) and before Alan Cox. According to whois, he registered this domain in 1994, which was right around the time that he was the Linux networking maintainer.
I really don't think this counts as cyber-squatting.
Re:Are we approaching microwave frequencies?
on
700 MHz Athlon
·
· Score: 1
As other posters have mentioned, microwaves ovens use frequencies around 2.4 GHz, which PCs aren't going to hit for a few years. So you don't need to worry about using your PC as a microwave oven quite yet. If you really want to cook food with your PC, you'll have to use it like the toaster-oven it is:-).
There still won't be anything to worry about when PCs hit 2.4 GHz -- take a look at most of the new high-end cordless phones that are hitting the market in the US. They all use the 2.4 GHz band. They're just a lot lower power than your oven.
PCI can support 64-bit addressing even in 32-bit slots. Most "high-end" PCI chips (like Ultra-3 controllers, gigabit Ethernet, and Fibre Channel) can do 64-bit addressing even when they're plugged into a 32-bit slot. They just have to do two address cycles instead of one.
Of course, 64-bit addressing doesn't help you much when you're using a PC with a 1 GB memory limit:-(.
Slashdot Mirror
Yeah, works great for me. It re-writes caller-id based on the names in my Mac's address book, recognizes certain callers and uses distinctive ring to tell what's happening, blacklists other numbers. I'm using it to send long-distance calls out via VoIP while still sending local calls via the POTS line that I can't get rid of. Great fun. Details on my blog.
This venturing a bit far from where we started--the point I was making is that an advanced Bayes filter can (in at least some situations) do a better job then SpamAssassin, even though SA includes a wide variety of filtering tools. I'm not trying to claim that SA is useless or that tablesaws are the end-all-be-all of woodworking tools (I don't actually use my tablesaw all that much :-).
For initially seeding the bayes table, I can certainly see SA's other filters as useful. Personally, I have a few thousand test cases sitting around that I can use for Bayes trainging, but random users don't generally work that way (personal experience: they keep everything, unsorted, in one big inbox, never hit delete, and expect me to keep it backed up for them. And then complain that my mail server is slow accessing their inbox).
However, in a lot of cases, I'd rather use one focused tool then one that can be customized and tuned into doing the job. I don't generally use perl to copy files--I use cp. When filtering my own pile of spam, I'll stick to a simple Bayesean filter because it's much more effective for me and I don't have to spend any time tweaking the weights of different filters. It's been a while since I used SA, but balancing the different filters was a pain, and it was always generating false positives. I'm sure it's better now, but I'm just amazed at how well SpamProbe has worked for me--it's been at least 4 months since I saw a false positive, and I only rarely see spam in my inbox. It's just a nicely tuned, nicely balanced tool that doesn't take a lot of effort to do a good job. And *that*'s what I'm really looking for in tools.
Actually, that *is* my argument. I used to carry around a multi-tool with a billion little blades and driver tips. In the end, I realized that all I really ever used was the #1 Phillips screwdriver, and the one on the multitool sucked--it was too short to reach into cases, and too hard to pull out. I was better served with a simple, cheap screwdriver. Some times single-use tools are much more powerful then jack-of-all-trades tools. And sometimes the uber-tool wins. Complexity is the enemy, do what you can to simplify the tools that you use.
Yeah, but if the Bayes classifier is much more accurate then any of the other classifiers (that's been my experience), then why not just turn all of the other ones off and go solely by the Bayes filter. Averaging 'great' with a lot of 'okay's doesn't do a whole lot to improve the 'great'.
My personal problem with SA is that it's really just a muddled average of a bunch of guessed-at filters for recognizing spam. The individual filters aren't very accurate, but the idea is that the average across a bunch of filters will be more accurate then any individual filter.
Bayes-based filters, on the other hand, directly calculate the probability of specific words appearing in spam vs. non-spam messages. Newer versions calculate the probability of short phrases, HTML tags, and mail headers as well. There's no guesswork involved (unlike SA)--if you feed them enough of yesterday's spam, then they're going to be really good with today and tomorrow's spam. The spammers keep evolving, so sooner or later messages will get through, but the filters keep evolving, too, and it's really hard to beat a good filter these days.
I've been using SpamProbe for almost 6 months, and it's amazingly accurate. I haven't had a false positive in months, and I only see a couple false negatives per month.
Done properly, software RAID 5 can be as fast (or even faster) then hardware RAID 5. After all, in most cases, "hardware" RAID is really just software RAID on an embedded CPU. I've dealt with way too many hardware RAID cards that slow to an utter crawl when doing RAID 5. On the other hand, Linux's software RAID is pretty zippy; I've seen 200MB/sec on the right hardware.
There can be performance issues if you're short of CPU on your system, or if you're low on RAM, or if you're nearing the limits of your system bus, but none of those are usually the case on modern storage servers--you don't usually need to run them that close to the edge.
The one thing that hardware RAID can buy for you is a non-volatile cache; that lets you get away with caching writes, and just generally buys you better performance and consistency. One of these days, someone will add journalled RAID-5 to Linux, though, and then you'll be able to accomplish the same thing with a PCI-based battery-backed RAM card for cheap.
So there has been a standard for almost four years, but how many HAVi enabled electronics devices do you see down at your local Generic Big Box Electronics Store? Zero would be a reasonable estimate.
Nope, not zero. Most of Mitsubishi's big-screen TVs have HAVi. I think their HDTV VCR is HAVi also. You can wire them together with firewire and then your TV's remote can control the VCR (as well as send and receive video) over firewire. The Mitsu TVs actually want to know all about your A/V system, including which boxes are wired to which audio and video ports on which other boxes. Then, everything (supposedly) works right if you want to have your HDTV sat receiver feed video straight to the TV, but feed audio into your (non-HDTV-capable) receiver. As long as you use the TV's remote and on-screen display, switching to the satellite receiver will automatically flip the receiver to the right input and then display the video direct from the satellite.
It's actually a bit cooler then that -- I ordered one of the HAVi books used from someone on Amazon (only $5), and it's kind of an interesting spec. It really wants to stream MPEGs over Firewire, and have devices hand around Java applets for UIs. So, your HAVi TV and HAVi satellite receiver could (in an ideal world) work by having the satellite receiver send the TV an applet that contains all of the satellite guide functions, and then the TV runs the applet on-screen for you. When you tell the applet to change the channel, it sends a HAVi control message to the satellite box, telling the box to change channels and start sending a new MPEG channel.
Just reducing the number of remotes is a very small piece of HAVi.
Did you read the specs? It records 640x480 MP4 at 1.5 or 3 Mbps. Assuming they actually mean Mbps not MBps, that works out to around an hour of high-quality video. Not too bad, but it depends on the video quality.
Ahh. I see. Combining the two (l7 and lvs) shouldn't be rocket science. Give it a few months.
You mean like LVS?
But that's not all you're paying for. You're not counting the cost of storage and printing. DIY-enlargements work out to a couple bucks per 8x10 (about the same as having film enlarged to the same size). And what about ink longevity?
You have a lot of good points. I've spent about a hundred on film storage since I took up photography (negative/slide sheets, storage boxes, etc), and I've bought an extra 80 GB hard drive to store pictures from the digital camera. I have a 35mm film scanner, and I was using it to scan and print negatives before I got the D60, so a lot of the printing comparisons break down. Anyway, 90% of the time, I get prints by burning a CD and dropping it off at the local Costco with a Frontier, so the print cost and longevity are the same as film prints. I have a decent inkjet, and I use it occasionally, but you can't really compare it to conventional prints--I can print any size up to 13x19, and I can spend hours tweaking it to look the way I want it to look. 95% of the time, there's something wrong with machine prints, either from digital or film, but it's ususally too much of a pain to get it fixed. A DIY printing solution gives you a lot more control, at the cost of taking longer. So, I use both--if I'm not feeling super-critical, I let Costco handle it. Otherwise I do it myself.
Also by storage I don't just mean the cost of a single hard drive, regardless of size. You've got backups, transfers to other media, etc., to worry about. And 20 years from now my negatives will still be in the box on the bookshelf, available for prints and enlargements. Where will your photoshop files be?
Good points, and mostly looking for a good solution. Except I've had a hard time finding several pages of negatives for the last year, and I have a bunch of other negatives that I cheaped out and had Costco reprint (rather then scanning them and giving them the files), and now they're scratched. Film and Digital *both* have storage issues. They're just not the same issues :-).
It all depends on your definition of "nice film-based SLR." I was $3-4k into Canon film cameras before I bought my D60; I don't think that's uncommon--one lens now, a new flash later, then a new body, it all adds up over the years.
So, adding a $2,200 D60 wasn't a *huge* step, price-wise. I've had it around 6 months, and I've shot around 7,000 frames with it. Assuming for the moment that I'd have shot the same number of frames had I been using film, that averages out to $0.35/frame, which is in the same general range as film and processing (that's $10 for 36 exposures).
Assuming that I've got at least another couple years of functional use in the camera, the per-frame cost should drop down under a dime. Plus, I get instant feedback (nice when fiddling with lighting problems) and it's easier for me to sort, edit, and produce prints with digital then it is with film.
So, with six months of use, you can start to argue that it's paid for itself. Add another couple years of use, and it'll be hard to argue that it would have been cheaper to use film. So, even if it has no resale value in 3 years, it'll still have been a good move, financially speaking.
I suppose it all depends on how much you shoot.
Because it's worth $20-$50 or so to have the video on-board. It takes up less space, has less impact on cooling, and is less likely to pop out during shipping.
That's pretty much it, though. Good 3D is utterly worthless for 99% of all server applications.
You mean, besides Costco or Walmart? Several of the local Costcos have the ability to print from flash cards or CDs in their one hour lab. They charge $0.20/print. I've heard that Walmarts are similar, but my skin crawls every time I set foot in one, so I tend to avoid the experience.
IMHO, the only problem with digital cameras for 95% of the uses out there today is the up-front cost of the equipment (plus maybe the PC software to interface with it. Get a Mac with iPhoto :-). A cheap Nikon 2500 produces crappy 4x6s just about as well as most of the cheap P&S cameras out there. My D60 seems to produce better prints then I could usually get out of my EOS 3. Not always, but usually. Plus, with the D60, I can shoot as much as I want without worrying about the hideous cost of film and quality processing, and I can go anywhere and get prints without worrying about cheap labs scratching my negatives. If I want a really good print, then I can go to a good lab. If I want a pile of 4x6s for family, then I can stop by Costco on the way to work. That, combined with the instant feedback that you get from the display (hmm, the histogram looks overexposed, or the subject blinked for all four shots, or whatever) means that I'm taking better pictures.
The 1Ds looks like it's going to be quite a bit better then the D60 in pretty much every way, probably encroaching into MF 's territory in many ways due to its lower noise. Is it perfect? Of course not. Does it have higher resolution, defined in terms of the ability to resolve really small details in images? Nope. Does it matter for prints under 11x14 or so? Not really. Frankly, I wouldn't be suprised if the 1Ds produces 16x20-ish portraits that compare favorably to MF prints in the same size.
Is film dead? I don't know, is portrait painting dead? Did color film kill black and white film? I supposed it depends on how you look at things -- I wouldn't have a clue how to find a portrait painter, except maybe the yellow pages in a major city. There aren't a whole lot of them out there. Black and white is nice, and usually better then color for pictures of people, frankly, but try finding a 1 hour (or even 24-hour) lab that can give you a pile of 4x6 prints from a roll of real black and while film for under $20. Most likely, what we're going to see over the next five or so years is the gradual disappearance of most of the less-popular professional 35mm film, along with most of the slower (100 and maybe 200 speed) consumer film. Frankly, 100 speed consumer film doesn't make any sense anyway -- with decent film, you shouldn't be able to tell the difference between 100 and 400 film in a 4x6, and most P&S cameras don't let in enough light to use ISO 100 95% of the time. You'll probably be able to buy 400 or 800-speed consumer film in the checkout line of your favorite grocery store for another 10+ years. Specialty stuff is already disappearing, though. I'd be amazed if more then 1/3 of the film types available today are still available in 6 or 7 years, and very little new film is going to be introduced from now on out.
That was with ext3. Reiser is even worse on large partitions, from what I've heard. Something about it having performance issues with filesystems over 1TB.
The point is that you can build a 2+TB system for well under $10k, using 160GB IDE drives and 3ware cards. I have 5 of them, and I've actually had problems -- my first partitioning attempt gave me a 2.06 TB RAID, which mke2fs decided was only 60 GB :-(.
The next round of storage servers that I buy will probably be even bigger, and it'd be nice to be able to use them as one big partition. Pity that I'll have to wait for 2.6 for that.
The reason that this announcement doesn't make sense is because it leaves out the non-Intel half of the Xeon server platform -- the ServerWorks "Grand Champion" chipset.
Go look at the specs for this beast -- it supports something like 6 PCI-X (133 MHz/64-bit) PCI *busses* and quad-channel DDR memory, with a 6.4 GB/sec memory bandwidth. They've been demoing it at conferences since early this year.
So, given that, the P4 makes a lot more sense in a server. Frankly, I would order one *now* if they were available, because I have a project that really needs to be able to manage multiple gigabit Ethernet pipes, and none of the existing x86 solutions (4x P3 Xeon, 2x Athlon, 2x Tualatin, 2x P4 w/ i860) really have the bus bandwidth that I'd like. They'll probably all work, but the P4 Xeon with the serverworks chipset would probably scale twice as far due to faster busses and faster memory.
Don't say "if you need that kind of performance, don't use x86." We're in the middle of switching to Linux/x86 from Solaris/Sparc for this, because low-end (under $50k) Sun hardware just isn't competetive with high-end PCs.
Clearly you've never bought RAM from Cisco :-).
More seriously, a lot of routers are limited to 128 MB or 256 MB of RAM. Right now, 128 MB is barely enough for a full routing table under IOS. Scale forward, and you'll see that 256 MB is only good for a year or two, tops.
I agree with your comment, but your math is *WAY* off :-). Using your numbers, here's what I get:
DVD - $19 for 5.2GB = $3.65/GBIDE - $238 for 60GB = $3.97/GB
CDR - $38 for 64GB = $0.59/GB
Of course, that ignores drive costs for the removable media. It also assumes that ~60 GB of data online with an IDE drive is comperable to 60 GB of data offline on a bunch of CD-R or DVD-RAM discs. It usually isn't. IMHO, the best mix for cost-sensitive users is probably a big IDE drive or two and a couple boxes of CDs for storing things that you don't need access to more then once or twice per month.
PDF isn't really that bad -- Adobe released the complete specs to it when they first released Acrobat. It's completely possible to write your own PDF reader/writer. Like xpdf and Ghostscript :-).
That's not to say that other formats aren't better, but PDF is way better then Word .doc files, which tend to be seen as the real alternative by a lot of companies.
I hated Song of Stone, and wasn't very fond of The Wasp Factory, but thought that Whit was fascinating. As others have mentioned, The Crow Road and Espedair Street are probably his best non-SF books. Of his SF books, Player of Games and Excession are my favorites.
I guess I'm just jealous because I didn't have the cashish to buy all these great domains back in '93 when i thought about it. :)
Domains were still free in '93. If you'd have registered them then, you wouldn't have had to pay anything until 95-97, when NSI started billing for legacy domains.
Also, the seller has been involved with Linux for a *long* time. He was the second maintainer of the Linux networking code, after Ross Biro (who wrote the first implmentation) and before Alan Cox. According to whois, he registered this domain in 1994, which was right around the time that he was the Linux networking maintainer.
I really don't think this counts as cyber-squatting.
As other posters have mentioned, microwaves ovens use frequencies around 2.4 GHz, which PCs aren't going to hit for a few years. So you don't need to worry about using your PC as a microwave oven quite yet. If you really want to cook food with your PC, you'll have to use it like the toaster-oven it is :-).
There still won't be anything to worry about when PCs hit 2.4 GHz -- take a look at most of the new high-end cordless phones that are hitting the market in the US. They all use the 2.4 GHz band. They're just a lot lower power than your oven.
PCI can support 64-bit addressing even in 32-bit slots. Most "high-end" PCI chips (like Ultra-3 controllers, gigabit Ethernet, and Fibre Channel) can do 64-bit addressing even when they're plugged into a 32-bit slot. They just have to do two address cycles instead of one.
Of course, 64-bit addressing doesn't help you much when you're using a PC with a 1 GB memory limit :-(.