"should" Why "should"? Sounds like a Grand Pronouncement from someone who wouldn't want anyone else to be different from him/er. An Anonymous Coward at that.
I host my own domain on a Linux machine in my basement. It has been very educational and offers me lots of flexibility that would be difficult were I to funnel everything through some ISP. Not to mention that I have better up time than any ISP I have ever used.
-kb, the Kent whose static IP address doesn't seem to have been blocked, judging from the 250 in his log from a test message he just sent.
More annoying than 10 (or 11) digit dialing is that the same company that now requires an area code for local calls used to prohibit it!
Verizon (nee Bell Atlantic, nee Nynex, nee New England Telephone) used to insist that calls to ones own area code must not include the area code. When I set up our Tivo I knew 10-digit dialing was coming, but no, I could only get 7-digits to go through. But Tivo is trouble-free enough that I forgot all about it until one day we noticed that we were running out of programming schedule--yup, the switch had happened and I had forgotten to go add the area code.
I don't let anyone install things on my computer. And I am not much interested in my installing their software for them. And I certainly am not attracted to the clueless folks who run cable systems.
But when I subscribed to DSL back when (*not* from the clueless phone company) I was intending to have a Linux machine on that wire, but I had a Macintosh ready to demonstrate whether it worked or not. I guess I would have let him install software over my frowns. But I would have wiped it after investigating.
> My favorite Linux configuration is 1 whatever IDE > drive for the OS, 1 IDE CDROM, and two (RAID-1) > large IDE's for data and configurations.
Why not put Linux on the raid 1? Bootable software raid 1 works. You can have your OS be redundant. You also would need fewer IDE controllers. Assuming the CD use is relatively infrequent, two ide controllers will do you well with each drive effectively getting its own controller, and that means speed.
-kb, the Kent who has become a software raid 1 fan.
I run Red Hat Linux on it and it works quite well, though not everything is supported. (Missing support: winmodem, suspend to disk, and LCD brightness control.)
I like it, it is quite small and light. It is possible that their big laptops are not so tough--but those monsters (by whatever manufacturer) are not suited to being hauled around, they weigh so damn much.
The downside to my Vaio, as others have experienced, is terrible battery life and very high prices for higher capacity batteries.
Release it freely. If it is actually good (or can be made good), use it to become famous, and find employment on that fame. Don't bother spending money patenting it because that would be a waste of money.
First, because there is no shortage of really good encryption available for free, you aren't going to be able to sell it.
Second, because it doesn't work, there is no point in wasting money trying to patent something that is faulty.
How do I know it doesn't work? Because nearly no one can design good cryptography, so chances are yours isn't any good either. And, yours is currently secret; secret cryptography is almost poor. Sure, you might be not be able to see how it is defective, but that only means it is tougher than your ability as a cryptanalyst. Good cryptanalysts are rare. You also seem to say that OTP is vulnerable to known-plaintext attacks, which as I understand it is simply false. A OTP has terrible key distribution problems and there are always attacks outside the strict domain of the encryption, but a one time pad is, if you define the problem as a narrow cryptographic problem, perfect. This makes me doubt your abilities.
Sorry to be so harsh,
-kb, the Kent who tries to know how much he doesn't know about cryptography.
It seems that this question is extremely dependent upon the kind of application.
Are you mostly reading, or also frequently writing this data? Are you searching or doing indexed lookups? Is this a nasty bandwidth hog or a trickle? Is this a zillion parallel transactions or only a few users? What kind of latencies are expected? What reliability is required? What access is needed to historical data?
Consider some concrete examples that are *very* different from each other yet could each total 50TB and would have very different solutions:
- Video-on-demand system for a Hollywood studio deciding that peer-to-peer pirate systems can only be beaten by a legitimate system that is better.
- Online credit card transaction system for, say, Visa.
- SETI data that needs to be collected and searched for messages from extraterrestrials.
- Particle accelerator data that needs to be collected at truly horrendous rates.
- Lexis/Nexis database.
- Google database.
- Echelon data.
- IRS data.
- "Dictionary attack" database for a lone cryto-analyst.
The possibilities go on and on. At the minimum a 50 TB database might be a small number of equipment racks with a single computer attached to them, all totaling maybe $100,000.
And on the other end, I can easily imagine a system where $200,000 of a much larger total might be spent for, say, a terabyte of DRAM.
I can easily imagine a system with less than $5,000 of battery backed up power supplies, and I can imagine a system with hundreds of throusands in generators.
This question has enormous dynamic range.
-kb, the Kent who would enjoy working out solutions for specific instances of this question.
1 million hour MTTF isn't as strange as it first looks. For one drive it means that it ~should~ not fail in its lifetime. (Note, Maxtor also lists a lifetime based on component life that is way shorter than 1 million hours.)
Where this kicks in is if you have a farm of a hundred of these drives. You can expect *one* of those drives to die in a mere 10,000 hours.
I have been running 4 Maxtor 60s in two software raid 1 pairs in two different boxes for a little while now and they still work well.
-kb, the Kent who has a cron job to let him know if one disk manages to die silently.
Mirroring is a great way to survive many hardware problems, but it clearly doesn't help recover from an "rm -rf/".
That's where the other popular question comes in: How do I fill it up? With backups!
Make extra partitions, don't keep them mounted except when backing up or restoring. Sure, a "dd" as root can still break things, so be a little careful. Also, it isn't rare to have more than one computer. Use one to back up another.
Tape is not only having troubles keeping up, and not only is it expensive, but it is *slow*. (What makes IDE seem faster than the blazes? Tape does. An on-disk backup can be used immediately, mount and run.)
Rarely, but I still put one in the home machine I recently built. It might have something to do with the fact my notebook only has a floppy drive for removable media. I need to be able to write floppies, if for no other reason than to be able to install new Linux distros on that notebook.
(I know, I can do it 100% over the net--unless I make a mistake.)
-kb, the Kent who really should update his notebook about now.
First, I want a much better screen--but black and white. Gimme higher resolution, higher contrast, higher reflectivity. Make it look like paper. Make it so good that I mostly don't need any backlight. Make it so good I don't long for a bigger screen. Make the display low in power consumption too.
Second, protect my data. I was a Newton user and the Newton didn't ever lose my data. Part of this was the fact it used flash memory, part was paranoid programming. I liked the result, Palm doesn't do so well.
Third, run Linux and open source apps.
Fourth, be small and tough.
Fifth, have slow wireless data service like the Palm VII/i705 (slow is good because it has deeper reach into buildings and lower battery requirements), but more general purpose (I wanna ssh into my server), and have coverage in more of the country and world.
Sixth, have some expandability. Those little IBM disks are tempting, but flash is fast and tough and lower power--let me choose.
Good point about the possibility of a complete CC# hash lookup table. (It reminds us that security is hard.)
Couldn't this problem be solved by the old Unix "salt" trick? That would be:
For each credit card number to be hashed, pick a high quality random number (the salt), concatenate it with the CC number, hash the whole thing, store the resulting hash and the random number, toss the credit card number.
To verify, do almost the same thing again: concatenate the purported credit card number with the stored salt, hash it, and compare with the stored hash.
Now your hash dictionary can't have a complete CC hash lookup for all such hashes, but only for all hashes that use the same salt as the card you are after. Make the salt a big as the resulting hash to assure the cracking is no easier than the security of your hash. Hell, the random number (probably?) doesn't even need to be high quality, just so you use a lot of different ones.
Read the previous post. The guy/gal said s/he works at the bank that *issues* the card. They need the card numbers so they can print your statement, send you a new card, verify charges from merchants, and generally maintain your credit card account.
-kb, the Kent who can't figure out *how* you got mod-ed up.
And if you are not in US, you don't have to substitute. (Presumably everyone who is not an American knows that fact.) The word was "can", not "must".
Americans, being rather insular (not all countries can equate "overseas" with "foreign"), sometimes need help dealing with the outside world, and they ARE a lot of the slashdot readership, so it is not efficient to tell them to do the conversion. (Isn't it?)
Oh, and not only folks in the USA need to convert from 220 (or 240). Mexico and Canada too. And Japan, they are 100 VAC. (Ever wonder why those little universal switching power supplies go all the way down to 100 volts? Not for California brownouts but for Japan.) I am sure there are others, but I can't recall them right now.
-kb, the American Kent whose daily paper is the International Herald Tribune.
P.S. 110 volts is low in the US. I think 117 is the nominal voltage, and I think of it as 120, for my outlets sometimes measure a smidge higher than that.
Don't remind me how angry I am with Eudora. I have lost mail in Eudora because it couldn't keep its bits straight.
Maybe Eudora is trustworthy now, maybe the search can even be made to work. (What a confusing user interface it had.) But that doesn't mean I am not pissed.
Certainly if you have an imediate problem that Unix and Postgresql address nicely, don't go getting burned with something too innovative. And don't necessarily computerize your card file either. (I miss library card files.)
Conversely, just because "more than 10 years" solves one person's problems doesn't mean there is no room for innovation.
Once upon a time (probably well over 10 years ago) I was involved in (or lt least watching) a Usenet discussion over file name length, you know the one: "My OS is better than yours, nyah, nyah!" One person, in exasperation, asked: "What do you want to do, put the whole file in the name?!"
At the point the light bulb went on for me and I said "Yes, maybe sometimes."
Files systems are for persistent storage of data. When that data is in active use it is seldomly in an unadorned linear order, yet when we store anything we are expected to flatten it to a linear stream and give it a single short name. Why? Tradition.
I suggest it would be good for OSs to offer higher level data storage services.
XML has come up in this thread. In some sense XML exists because we seem to be forever stuck with flat files but our data still insists on having richer structure. XML is a way to force multidimensional rich data into a flat space.
Why does our data misbehave so? Why can't it just be nice and flat like our files are? Damn data.
-kb, the Kent who liked the Macintosh's original resource manager.
Once I was at some internet tradeshow in Boston and every other booth seemed to be showing off their e-mail filtering features, each with one or more enormously complicated dialog box. Features! Features! Features!
My reaction was to want an e-mail reading program that didn't require any filter configuration, though I imagined it would do well to be given a few hints, such as who my boss is, who my mother is, and who my wife is. Other than that, let the program figure it out.
Imagine the canonical, old-fashioned secretary temp. She ('cause that's what the canonical version was) didn't have to know anything domain-specific to sort the morning mail. Magazines go together, bills go together, personal letters go together, etc.
I imagine an automated version for my e-mail. Look at who it is "to" (am I on the list?), look at who is "cc"-ed (am I on that list?), look at who it is from (my boss, wife, or mother?), look at who else it is to (boss, wife, or mother?), look at the thread it is part of (is it responding to something I previously wrote?), look at the content (does it mention me, things I have written, my boss, wife, or mother?). Was it sent to a mailing list? Was it written by someone I have explicitly written to (once or many times?)? Was it written by someone who has previously sent me direct e-mail (once or many times?)? Those ideas are just the obvious ones, think of others. Think of more. (Does it talk about sex, credit card merchant accounts, stock tips, or Nigerian money?)
Now take that and sort it by importance and similarity. Look for a way to present me in a descriptive summary, arranged in a hierarchy with a top-level of, say, 3 to 9 categories, a greatest depth no greater than, say, 4, and keep the sub-branching at intermediate nodes between 3 and 5--but don't max out all those dimensions at once, try to keep the total number of leaf categories to under, say, two dozen. Try to make more important items land higher in the tree and with few siblings, grouped with siblings of similar importance. (Maybe give an importance weight to each e-mail and balance the tree on that scale, that would float e-mails to me from my boss about my mother and wife really high with few siblings.)
This summary needs to be integrated with a complete index of the e-mail so I can see how a message fits into a larger thread, how it fits into previous e-mails.
I (the user) would need to tell the program when to make me a summary of my e-mail (e-mail reading is different when a lot comes in or just a little), and I want to be able to browse through old summaries, including deciding to see composite summaries or, say, the last several days, a week (or three), month, year, or 400 days.
So I think it ends up being a 4-part user interface:
List of summaries (which can be manipulated).
A given summary.
Exhaustive thread/date/subject/sender list (analogous to what every e-mail reader seems to have now). Note that this view could effectively be turned into an exhaustive address book. Frequent (favored) correspondents could be highlighted by me for ease in sending a new e-mail, and also to provide importance hints to the program. This is where I might say who my boss/wife/mother is.
A body of a (or more) specific e-mail being read, written, or old e-mail (sent or received) being reviewed.
And I could go on, but I won't. If anyone wants to write such a thing and wants to hear more, send me an, um, e-mail.
-kb, the Kent who has been saving all his e-mail (including spam!) for a year or so, providing plenty of raw material to test any such program.
So I plug this in and install some software. And I get someone at the other end of the country to do the same. How do I call the other person? How does the person call me? How many of us need static IPs? What kind of NATing breaks it?
-kb, the Kent who is wants to know, but who also had to post something to undo a moderating mistake he made.
Something that wasn't clear to me in the review was that there *are* 2 IDE controllers. That, plus two 3.5 inch bays means one could run a pair of drives in RAID 1, each on its own bus (for more reliability and speed) and still have a CD drive in the large bay, for a box that might be very reliable.
-kb, the Kent who just built himself a RAID 1 mid-tower and would have liked a smaller box.
On my notebook, 2.4.9-ac16 with the preemption patch for me.
I tried several newer but didn't like the way it used memory or its responsiveness or both (I forget the details). I stressed my notebook's memory by launching everything I could and it got weird on me. I think some versions started using up amazing amounts of swap. I think I will stay put until I see a sane settlement of the VM Wars. Hell, I might stay here until I see a 2.5 that I like.
There are a lot of folks happy with the latest 2.4 kernel, but there seem to be some perverse behaviors for other folks.
I believe the usual US standard bit-rate for transmitting compressed HD video over the airwaves is 30Mb/s.
ATSC (the broken US DTV system) has a data rate is 19 Mb/s, using 8VSB modulation.
DVB (used system nearly everywhere else) uses COFDM modulation and can go a bit higher if broadcaster assumes viewers are in good circumstances, a tad lower for rather robust signals, or a lot lower for completely bullet-proof signals (say, after a storm when both transmitter and receiver antennas are damaged). DVB can also do hierarchial modulation where some of the bits are really easy to receive (enough for roughly regular quality TV) but most of them require a better antenna (enough bits for HDTV).
And before some smartarse points out that HD resolution is 1920 x 1080, that's only the active picture area. The full frame including blanking (which is used for things like embedded audio etc.) is 2200 x 1125.
First, digital HD is flexible and doesn't have a single precise dimension or frame rate (hell, some try to call interlaced video high definition). There are acquisition dimensions, post processing dimensions, emission dimensions, and display dimensions, all of which can theoretically (and really should) be different for a single program, and different for different programs. And if watching a 24 fps movie 24 fps is a sensible frame rate, but if watching basketball 60 fps is much more appropriate.
Second, digital TV doesn't transmit analogue blanking areas as part of a raster, and it doesn't put audio in these non-existent blanked picture areas either. Yes, audio has to be there, but once the picture gets very good the number of bits needed for killer 5.1 audio isn't a big part of the budget.
Third, though there are multiple ways to code color, for human consumption they all require 3 values, but your example suggests "chroma" is a single value. An important reason for having a luma and two color signals is to give luma more bits and spend fewer bits for color (which matches the human vision system). In RGB one has to give all the colors the same number of bits.
First, APM itself might not be a good idea for serious servers, but building (and configuring) servers with some consideration of power efficiency would be smart. The power use by server farms is a horrible expense. The cooling costs of server farms is horrible. But up to now it seems that getting a computer to work at all is the only point; how many watts it takes and how many BTUs it dumps is mostly ignored. Being Biggest and Baddest is used to sell, efficiency is not. I expect this will soon change...
Second, most servers are not on server farms. My basement server might be on a DSL connection that is faster than most leased lines of yore, but it is still IO-limited. So it works quite well for me to run a little hacked Think NIC box (www.thinknic.com): I added an otherwise missing hard disk and underclocked (!) the CPU, and the result takes very little power--it has to, the power supply on the thing is too small to draw much. I keep the CRT off when I am not using it. I also bought a little UPS--but the server takes so little power my backup time should be very good. Certainly I am a minimal case, but I suspect that many servers out there are over powered and misused.
Third, why don't computers and related equipment have small builtin UPSs? They already have DC power supplies, and DC is what is needed to charge most batteries. DC is what the computer actually needs, and DC is what batteries produce. Doing some battery backup inside each box would be pretty easy. How much battery does a little ethernet hub need? External UPSs need to make AC from DC (which is never terribly efficient) and they themselves become single points for potential failure. Sure, if you need a survivable facility, buy big UPSs and generators, but the failover and resistance to tripping over power cords would be so much better if each piece of equipment had a few minutes of backup built in. A well maintained generator should be able to start up and be running smoothly within just a few minutes. If the equipment itself could last a dozen minutes or so, there would be no need for any external UPSs other than for a few CRTs. As most power problems are very short, even home users would like a few minutes of backup time.
-kb, the Kent who thinks computers are in a brute-force '50s "muscle car" era and that there is a lot of room for a little design and deployment efficiency.
P.S. Don't forget that most so called "screen savers" are really just entertainments that don't save anything.
Why Does (Nearly) Everybody Want Color?
on
The new Palm VIIx
·
· Score: 1
Don't you people ever go outside? All the color screens I have seen on PDAs and notebooks must emit light to be seen, which means when you are outside, they have to compete with the sun. The result is you can't really see the screen and your batteries are going dead fast.
I was recently sitting outdoors at Starbucks and this guy who had seen all those commercials showing folks with laptops escaping the office decided to try it himself. But when he discovered he couldn't see the screen, he had to go back inside. It was a beautiful day too.
It is little like the clerk in a store who saw me carrying my Razor scooter, thought it looked cool, but asked whether they were available with an engine? Well, sort of, but did he really think I should be dripping gasoline on the bus, or in his store? Or was he thinking I should be carrying some heavy rechargable model with limited range?
Be careful what features you ask for, you might get them. Where can you buy a B&W laptop nowadays? And other than folks running Photoshop or doing video editing, do you really need color enough to always stay indoors?
The Palm is a success not because it is less capable than was the Newton, it is a success because it decided how capable to be, and then nailed it! (The Newton promised the moon and stars, but only achieved a moderate orbit, and at geosynchronous prices.) Keep it simple.
-kb, the Kent who has a Palm VII, and wonders whether the VIIx's "active matrix" screen is much clearer and whether the backlight works better.
Depending on how you look at it, there seem to be three approaches to defining an OS.
1. Bottom Up. These are the BIOS and kernel folks. They want to cover up the hardware and very little more.
2. Sideways. These are the "I fire up my computer, the OS is what I see before I start to do anything."-folks.
3. Top down. Services provided to make writing applications easier and more orderly.
This last one has a certain appeal, for it remembers that computers are extensible, that their key value is in how they can be applied in unanticipated ways.
All seem to have something going for them--if you look at the computer from their perspective. So what do we get if we try to launder out the perspective differences, what is left?
Here is my stab at a definition: The OS is the general purpose software that supports the extension of a computer's abilities.
This usually includes hardware abstraction and resource management.
It usually includes functions that each programmer isn't expected to have to write from scratch: math libraries, a GUI, file systems, networking.
It may include ways to extending the OS itself: facilities for adding shared libraries or drivers.
It may be argued to not exist: My digital cellphone has no visible OS; the folks who programmed it certainly worked with a hardware layer and likely a kernel, but once it is sealed up and handed to me the OS might as well not exist, it becomes a single use appliance and it becomes a bit silly to try point out the OS anymore.
This isn't a perfect definition, but it captures much of people think of when they think "operating system", and it is rich enough to have some conceptual value.
-kb, the Kent who has ideas about what future OSs should do too.
My basic reaction to the Blair Witch Project is that I wish they had told a better story. But that's such a basic and general purpose piece of advice that it almost seems silly to recommend it. Yet so many movies, including this one, would do so much better to remember it.
For those of you who have seen the film, have you told anyone about it? Did you have anything to say other than "These kids, you see, they get lost in the woods looking for the Blair Witch, they're making a documentary, and, um, all this scary stuff happens. But I can't tell you how it ends."
But if you have seen the poster ("a year later their film was found") you know they didn't get out alive. I don't see that there are any spoilers possible with this film because nothing happens beyond some kids being stupid and freaking out and dying.
Why is The Blair Witch Project is so successful? Let me reveal just how old I must be with the following.
People, particularly "kids these days", the ones who go to all those movies, are afraid of the woods. They are also tired of "Oow, what an ugly monster, but not as good as Aliens." The youngerly can relate to seeing other youngerly panicing. And Hollywood's gloss gets predictable.
So put a bunch of panicy kids in the woods, don't show the monster/witch/whatever, use--er--innovative camera techniques that don't look like Hollywood gloss, and I can see why it is a success.
I admire that the film makers realized that what isn't seen is often more frightening than what is seen, however, I wish they had spent more time trying to tell a good story instead of merely trying to be scary, because a good story offers its own suspense.
One thing that can drive an audience crazy (in a good way) is when they know something the character on screen doesn't. When this be something dangerous to the onscreen character the result can be a powerful "Don't do it!!!!"
The problem with Blair Witch is that, being filmed by the characters themselves, there is no way we can see things they don't also see. There is no way we can see the metaphorical knife being raised behind anyone's back. Or is there?
I suggest that two things could have been done.
One would have been to take advantage of the fact that they ran their video camera so much and let them leave it on when they don't intent to or to let them look up over the camera at something else as the camera (and audience) catches something the characters themselves miss.
The second idea would have been near the end to have the camera dropped...and picked up. It could have had a smoother or at least different style of operation. It could have moved in a searching way that would have completely reversed the audience's previous identification with the moving camera. A character could have turned, looked at whatever was now holding the camera, and been completely mortified. The very turn of events would have been so scary that no extra extra emoting would have been needed.
Add to that some careful thinking about how to tell a good story and this film could really have been worthy of all the hype and success.
-kb, the Kent who doesn't pretend to be good at telling a story, but he knows it can be done better than in The Blair Witch Project.
Slashdot Mirror
"should" Why "should"? Sounds like a Grand Pronouncement from someone who wouldn't want anyone else to be different from him/er. An Anonymous Coward at that.
I host my own domain on a Linux machine in my basement. It has been very educational and offers me lots of flexibility that would be difficult were I to funnel everything through some ISP. Not to mention that I have better up time than any ISP I have ever used.
-kb, the Kent whose static IP address doesn't seem to have been blocked, judging from the 250 in his log from a test message he just sent.
More annoying than 10 (or 11) digit dialing is that the same company that now requires an area code for local calls used to prohibit it!
Verizon (nee Bell Atlantic, nee Nynex, nee New England Telephone) used to insist that calls to ones own area code must not include the area code. When I set up our Tivo I knew 10-digit dialing was coming, but no, I could only get 7-digits to go through. But Tivo is trouble-free enough that I forgot all about it until one day we noticed that we were running out of programming schedule--yup, the switch had happened and I had forgotten to go add the area code.
-kb
I don't let anyone install things on my computer. And I am not much interested in my installing their software for them. And I certainly am not attracted to the clueless folks who run cable systems.
But when I subscribed to DSL back when (*not* from the clueless phone company) I was intending to have a Linux machine on that wire, but I had a Macintosh ready to demonstrate whether it worked or not. I guess I would have let him install software over my frowns. But I would have wiped it after investigating.
-kb
> My favorite Linux configuration is 1 whatever IDE
> drive for the OS, 1 IDE CDROM, and two (RAID-1)
> large IDE's for data and configurations.
Why not put Linux on the raid 1? Bootable software raid 1 works. You can have your OS be redundant. You also would need fewer IDE controllers. Assuming the CD use is relatively infrequent, two ide controllers will do you well with each drive effectively getting its own controller, and that means speed.
-kb, the Kent who has become a software raid 1 fan.
I have a Vaio PCG-Z505LE, and it is very tough.
I run Red Hat Linux on it and it works quite well, though not everything is supported. (Missing support: winmodem, suspend to disk, and LCD brightness control.)
I like it, it is quite small and light. It is possible that their big laptops are not so tough--but those monsters (by whatever manufacturer) are not suited to being hauled around, they weigh so damn much.
The downside to my Vaio, as others have experienced, is terrible battery life and very high prices for higher capacity batteries.
-kb, the Kent who thinks small is beautiful.
Release it freely. If it is actually good (or can be made good), use it to become famous, and find employment on that fame. Don't bother spending money patenting it because that would be a waste of money.
First, because there is no shortage of really good encryption available for free, you aren't going to be able to sell it.
Second, because it doesn't work, there is no point in wasting money trying to patent something that is faulty.
How do I know it doesn't work? Because nearly no one can design good cryptography, so chances are yours isn't any good either. And, yours is currently secret; secret cryptography is almost poor. Sure, you might be not be able to see how it is defective, but that only means it is tougher than your ability as a cryptanalyst. Good cryptanalysts are rare. You also seem to say that OTP is vulnerable to known-plaintext attacks, which as I understand it is simply false. A OTP has terrible key distribution problems and there are always attacks outside the strict domain of the encryption, but a one time pad is, if you define the problem as a narrow cryptographic problem, perfect. This makes me doubt your abilities.
Sorry to be so harsh,
-kb, the Kent who tries to know how much he doesn't know about cryptography.
It seems that this question is extremely dependent upon the kind of application.
Are you mostly reading, or also frequently writing this data? Are you searching or doing indexed lookups? Is this a nasty bandwidth hog or a trickle? Is this a zillion parallel transactions or only a few users? What kind of latencies are expected? What reliability is required? What access is needed to historical data?
Consider some concrete examples that are *very* different from each other yet could each total 50TB and would have very different solutions:
- Video-on-demand system for a Hollywood studio deciding that peer-to-peer pirate systems can only be beaten by a legitimate system that is better.
- Online credit card transaction system for, say, Visa.
- SETI data that needs to be collected and searched for messages from extraterrestrials.
- Particle accelerator data that needs to be collected at truly horrendous rates.
- Lexis/Nexis database.
- Google database.
- Echelon data.
- IRS data.
- "Dictionary attack" database for a lone cryto-analyst.
The possibilities go on and on. At the minimum a 50 TB database might be a small number of equipment racks with a single computer attached to them, all totaling maybe $100,000.
And on the other end, I can easily imagine a system where $200,000 of a much larger total might be spent for, say, a terabyte of DRAM.
I can easily imagine a system with less than $5,000 of battery backed up power supplies, and I can imagine a system with hundreds of throusands in generators.
This question has enormous dynamic range.
-kb, the Kent who would enjoy working out solutions for specific instances of this question.
1 million hour MTTF isn't as strange as it first looks. For one drive it means that it ~should~ not fail in its lifetime. (Note, Maxtor also lists a lifetime based on component life that is way shorter than 1 million hours.)
Where this kicks in is if you have a farm of a hundred of these drives. You can expect *one* of those drives to die in a mere 10,000 hours.
I have been running 4 Maxtor 60s in two software raid 1 pairs in two different boxes for a little while now and they still work well.
-kb, the Kent who has a cron job to let him know if one disk manages to die silently.
Mirroring is a great way to survive many hardware problems, but it clearly doesn't help recover from an "rm -rf /".
That's where the other popular question comes in: How do I fill it up? With backups!
Make extra partitions, don't keep them mounted except when backing up or restoring. Sure, a "dd" as root can still break things, so be a little careful. Also, it isn't rare to have more than one computer. Use one to back up another.
Tape is not only having troubles keeping up, and not only is it expensive, but it is *slow*. (What makes IDE seem faster than the blazes? Tape does. An on-disk backup can be used immediately, mount and run.)
-kb
Rarely, but I still put one in the home machine I recently built. It might have something to do with the fact my notebook only has a floppy drive for removable media. I need to be able to write floppies, if for no other reason than to be able to install new Linux distros on that notebook.
(I know, I can do it 100% over the net--unless I make a mistake.)
-kb, the Kent who really should update his notebook about now.
First, I want a much better screen--but black and white. Gimme higher resolution, higher contrast, higher reflectivity. Make it look like paper. Make it so good that I mostly don't need any backlight. Make it so good I don't long for a bigger screen. Make the display low in power consumption too.
Second, protect my data. I was a Newton user and the Newton didn't ever lose my data. Part of this was the fact it used flash memory, part was paranoid programming. I liked the result, Palm doesn't do so well.
Third, run Linux and open source apps.
Fourth, be small and tough.
Fifth, have slow wireless data service like the Palm VII/i705 (slow is good because it has deeper reach into buildings and lower battery requirements), but more general purpose (I wanna ssh into my server), and have coverage in more of the country and world.
Sixth, have some expandability. Those little IBM disks are tempting, but flash is fast and tough and lower power--let me choose.
-kb, the Kent who has been using PDAs for years.
Good point about the possibility of a complete CC# hash lookup table. (It reminds us that security is hard.)
Couldn't this problem be solved by the old Unix "salt" trick? That would be:
For each credit card number to be hashed, pick a high quality random number (the salt), concatenate it with the CC number, hash the whole thing, store the resulting hash and the random number, toss the credit card number.
To verify, do almost the same thing again: concatenate the purported credit card number with the stored salt, hash it, and compare with the stored hash.
Now your hash dictionary can't have a complete CC hash lookup for all such hashes, but only for all hashes that use the same salt as the card you are after. Make the salt a big as the resulting hash to assure the cracking is no easier than the security of your hash. Hell, the random number (probably?) doesn't even need to be high quality, just so you use a lot of different ones.
-kb
Read the previous post. The guy/gal said s/he works at the bank that *issues* the card. They need the card numbers so they can print your statement, send you a new card, verify charges from merchants, and generally maintain your credit card account.
-kb, the Kent who can't figure out *how* you got mod-ed up.
And if you are not in US, you don't have to substitute. (Presumably everyone who is not an American knows that fact.) The word was "can", not "must".
Americans, being rather insular (not all countries can equate "overseas" with "foreign"), sometimes need help dealing with the outside world, and they ARE a lot of the slashdot readership, so it is not efficient to tell them to do the conversion. (Isn't it?)
Oh, and not only folks in the USA need to convert from 220 (or 240). Mexico and Canada too. And Japan, they are 100 VAC. (Ever wonder why those little universal switching power supplies go all the way down to 100 volts? Not for California brownouts but for Japan.) I am sure there are others, but I can't recall them right now.
-kb, the American Kent whose daily paper is the International Herald Tribune.
P.S. 110 volts is low in the US. I think 117 is the nominal voltage, and I think of it as 120, for my outlets sometimes measure a smidge higher than that.
Don't remind me how angry I am with Eudora. I have lost mail in Eudora because it couldn't keep its bits straight.
Maybe Eudora is trustworthy now, maybe the search can even be made to work. (What a confusing user interface it had.) But that doesn't mean I am not pissed.
-kb
In other words, if it ain't broke don't fix it?
Certainly if you have an imediate problem that Unix and Postgresql address nicely, don't go getting burned with something too innovative. And don't necessarily computerize your card file either. (I miss library card files.)
Conversely, just because "more than 10 years" solves one person's problems doesn't mean there is no room for innovation.
Once upon a time (probably well over 10 years ago) I was involved in (or lt least watching) a Usenet discussion over file name length, you know the one: "My OS is better than yours, nyah, nyah!" One person, in exasperation, asked: "What do you want to do, put the whole file in the name?!"
At the point the light bulb went on for me and I said "Yes, maybe sometimes."
Files systems are for persistent storage of data. When that data is in active use it is seldomly in an unadorned linear order, yet when we store anything we are expected to flatten it to a linear stream and give it a single short name. Why? Tradition.
I suggest it would be good for OSs to offer higher level data storage services.
XML has come up in this thread. In some sense XML exists because we seem to be forever stuck with flat files but our data still insists on having richer structure. XML is a way to force multidimensional rich data into a flat space.
Why does our data misbehave so? Why can't it just be nice and flat like our files are? Damn data.
-kb, the Kent who liked the Macintosh's original resource manager.
My reaction was to want an e-mail reading program that didn't require any filter configuration, though I imagined it would do well to be given a few hints, such as who my boss is, who my mother is, and who my wife is. Other than that, let the program figure it out.
Imagine the canonical, old-fashioned secretary temp. She ('cause that's what the canonical version was) didn't have to know anything domain-specific to sort the morning mail. Magazines go together, bills go together, personal letters go together, etc.
I imagine an automated version for my e-mail. Look at who it is "to" (am I on the list?), look at who is "cc"-ed (am I on that list?), look at who it is from (my boss, wife, or mother?), look at who else it is to (boss, wife, or mother?), look at the thread it is part of (is it responding to something I previously wrote?), look at the content (does it mention me, things I have written, my boss, wife, or mother?). Was it sent to a mailing list? Was it written by someone I have explicitly written to (once or many times?)? Was it written by someone who has previously sent me direct e-mail (once or many times?)? Those ideas are just the obvious ones, think of others. Think of more. (Does it talk about sex, credit card merchant accounts, stock tips, or Nigerian money?)
Now take that and sort it by importance and similarity. Look for a way to present me in a descriptive summary, arranged in a hierarchy with a top-level of, say, 3 to 9 categories, a greatest depth no greater than, say, 4, and keep the sub-branching at intermediate nodes between 3 and 5--but don't max out all those dimensions at once, try to keep the total number of leaf categories to under, say, two dozen. Try to make more important items land higher in the tree and with few siblings, grouped with siblings of similar importance. (Maybe give an importance weight to each e-mail and balance the tree on that scale, that would float e-mails to me from my boss about my mother and wife really high with few siblings.)
This summary needs to be integrated with a complete index of the e-mail so I can see how a message fits into a larger thread, how it fits into previous e-mails.
I (the user) would need to tell the program when to make me a summary of my e-mail (e-mail reading is different when a lot comes in or just a little), and I want to be able to browse through old summaries, including deciding to see composite summaries or, say, the last several days, a week (or three), month, year, or 400 days.
So I think it ends up being a 4-part user interface:
List of summaries (which can be manipulated).
A given summary.
Exhaustive thread/date/subject/sender list (analogous to what every e-mail reader seems to have now). Note that this view could effectively be turned into an exhaustive address book. Frequent (favored) correspondents could be highlighted by me for ease in sending a new e-mail, and also to provide importance hints to the program. This is where I might say who my boss/wife/mother is.
A body of a (or more) specific e-mail being read, written, or old e-mail (sent or received) being reviewed.
And I could go on, but I won't. If anyone wants to write such a thing and wants to hear more, send me an, um, e-mail.
-kb, the Kent who has been saving all his e-mail (including spam!) for a year or so, providing plenty of raw material to test any such program.
So I plug this in and install some software. And I get someone at the other end of the country to do the same. How do I call the other person? How does the person call me? How many of us need static IPs? What kind of NATing breaks it?
-kb, the Kent who is wants to know, but who also had to post something to undo a moderating mistake he made.
Something that wasn't clear to me in the review was that there *are* 2 IDE controllers. That, plus two 3.5 inch bays means one could run a pair of drives in RAID 1, each on its own bus (for more reliability and speed) and still have a CD drive in the large bay, for a box that might be very reliable.
-kb, the Kent who just built himself a RAID 1 mid-tower and would have liked a smaller box.
On my notebook, 2.4.9-ac16 with the preemption patch for me.
I tried several newer but didn't like the way it used memory or its responsiveness or both (I forget the details). I stressed my notebook's memory by launching everything I could and it got weird on me. I think some versions started using up amazing amounts of swap. I think I will stay put until I see a sane settlement of the VM Wars. Hell, I might stay here until I see a 2.5 that I like.
There are a lot of folks happy with the latest 2.4 kernel, but there seem to be some perverse behaviors for other folks.
-kb
ATSC (the broken US DTV system) has a data rate is 19 Mb/s, using 8VSB modulation.
DVB (used system nearly everywhere else) uses COFDM modulation and can go a bit higher if broadcaster assumes viewers are in good circumstances, a tad lower for rather robust signals, or a lot lower for completely bullet-proof signals (say, after a storm when both transmitter and receiver antennas are damaged). DVB can also do hierarchial modulation where some of the bits are really easy to receive (enough for roughly regular quality TV) but most of them require a better antenna (enough bits for HDTV).
And before some smartarse points out that HD resolution is 1920 x 1080, that's only the active picture area. The full frame including blanking (which is used for things like embedded audio etc.) is 2200 x 1125.
First, digital HD is flexible and doesn't have a single precise dimension or frame rate (hell, some try to call interlaced video high definition). There are acquisition dimensions, post processing dimensions, emission dimensions, and display dimensions, all of which can theoretically (and really should) be different for a single program, and different for different programs. And if watching a 24 fps movie 24 fps is a sensible frame rate, but if watching basketball 60 fps is much more appropriate.
Second, digital TV doesn't transmit analogue blanking areas as part of a raster, and it doesn't put audio in these non-existent blanked picture areas either. Yes, audio has to be there, but once the picture gets very good the number of bits needed for killer 5.1 audio isn't a big part of the budget.
Third, though there are multiple ways to code color, for human consumption they all require 3 values, but your example suggests "chroma" is a single value. An important reason for having a luma and two color signals is to give luma more bits and spend fewer bits for color (which matches the human vision system). In RGB one has to give all the colors the same number of bits.
-kb
First, APM itself might not be a good idea for serious servers, but building (and configuring) servers with some consideration of power efficiency would be smart. The power use by server farms is a horrible expense. The cooling costs of server farms is horrible. But up to now it seems that getting a computer to work at all is the only point; how many watts it takes and how many BTUs it dumps is mostly ignored. Being Biggest and Baddest is used to sell, efficiency is not. I expect this will soon change...
Second, most servers are not on server farms. My basement server might be on a DSL connection that is faster than most leased lines of yore, but it is still IO-limited. So it works quite well for me to run a little hacked Think NIC box (www.thinknic.com): I added an otherwise missing hard disk and underclocked (!) the CPU, and the result takes very little power--it has to, the power supply on the thing is too small to draw much. I keep the CRT off when I am not using it. I also bought a little UPS--but the server takes so little power my backup time should be very good. Certainly I am a minimal case, but I suspect that many servers out there are over powered and misused.
Third, why don't computers and related equipment have small builtin UPSs? They already have DC power supplies, and DC is what is needed to charge most batteries. DC is what the computer actually needs, and DC is what batteries produce. Doing some battery backup inside each box would be pretty easy. How much battery does a little ethernet hub need? External UPSs need to make AC from DC (which is never terribly efficient) and they themselves become single points for potential failure. Sure, if you need a survivable facility, buy big UPSs and generators, but the failover and resistance to tripping over power cords would be so much better if each piece of equipment had a few minutes of backup built in. A well maintained generator should be able to start up and be running smoothly within just a few minutes. If the equipment itself could last a dozen minutes or so, there would be no need for any external UPSs other than for a few CRTs. As most power problems are very short, even home users would like a few minutes of backup time.
-kb, the Kent who thinks computers are in a brute-force '50s "muscle car" era and that there is a lot of room for a little design and deployment efficiency.
P.S. Don't forget that most so called "screen savers" are really just entertainments that don't save anything.
Don't you people ever go outside? All the color screens I have seen on PDAs and notebooks must emit light to be seen, which means when you are outside, they have to compete with the sun. The result is you can't really see the screen and your batteries are going dead fast.
I was recently sitting outdoors at Starbucks and this guy who had seen all those commercials showing folks with laptops escaping the office decided to try it himself. But when he discovered he couldn't see the screen, he had to go back inside. It was a beautiful day too.
It is little like the clerk in a store who saw me carrying my Razor scooter, thought it looked cool, but asked whether they were available with an engine? Well, sort of, but did he really think I should be dripping gasoline on the bus, or in his store? Or was he thinking I should be carrying some heavy rechargable model with limited range?
Be careful what features you ask for, you might get them. Where can you buy a B&W laptop nowadays? And other than folks running Photoshop or doing video editing, do you really need color enough to always stay indoors?
The Palm is a success not because it is less capable than was the Newton, it is a success because it decided how capable to be, and then nailed it! (The Newton promised the moon and stars, but only achieved a moderate orbit, and at geosynchronous prices.) Keep it simple.
-kb, the Kent who has a Palm VII, and wonders whether the VIIx's "active matrix" screen is much clearer and whether the backlight works better.
Depending on how you look at it, there seem to be three approaches to defining an OS.
1. Bottom Up. These are the BIOS and kernel folks. They want to cover up the hardware and very little more.
2. Sideways. These are the "I fire up my computer, the OS is what I see before I start to do anything."-folks.
3. Top down. Services provided to make writing applications easier and more orderly.
This last one has a certain appeal, for it remembers that computers are extensible, that their key value is in how they can be applied in unanticipated ways.
All seem to have something going for them--if you look at the computer from their perspective. So what do we get if we try to launder out the perspective differences, what is left?
Here is my stab at a definition: The OS is the general purpose software that supports the extension of a computer's abilities.
This usually includes hardware abstraction and resource management.
It usually includes functions that each programmer isn't expected to have to write from scratch: math libraries, a GUI, file systems, networking.
It may include ways to extending the OS itself: facilities for adding shared libraries or drivers.
It may be argued to not exist: My digital cellphone has no visible OS; the folks who programmed it certainly worked with a hardware layer and likely a kernel, but once it is sealed up and handed to me the OS might as well not exist, it becomes a single use appliance and it becomes a bit silly to try point out the OS anymore.
This isn't a perfect definition, but it captures much of people think of when they think "operating system", and it is rich enough to have some conceptual value.
-kb, the Kent who has ideas about what future OSs should do too.
My basic reaction to the Blair Witch Project is that I wish they had told a better story. But that's such a basic and general purpose piece of advice that it almost seems silly to recommend it. Yet so many movies, including this one, would do so much better to remember it.
For those of you who have seen the film, have you told anyone about it? Did you have anything to say other than "These kids, you see, they get lost in the woods looking for the Blair Witch, they're making a documentary, and, um, all this scary stuff happens. But I can't tell you how it ends."
But if you have seen the poster ("a year later their film was found") you know they didn't get out alive. I don't see that there are any spoilers possible with this film because nothing happens beyond some kids being stupid and freaking out and dying.
Why is The Blair Witch Project is so successful? Let me reveal just how old I must be with the following.
People, particularly "kids these days", the ones who go to all those movies, are afraid of the woods. They are also tired of "Oow, what an ugly monster, but not as good as Aliens." The youngerly can relate to seeing other youngerly panicing. And Hollywood's gloss gets predictable.
So put a bunch of panicy kids in the woods, don't show the monster/witch/whatever, use--er--innovative camera techniques that don't look like Hollywood gloss, and I can see why it is a success.
I admire that the film makers realized that what isn't seen is often more frightening than what is seen, however, I wish they had spent more time trying to tell a good story instead of merely trying to be scary, because a good story offers its own suspense.
One thing that can drive an audience crazy (in a good way) is when they know something the character on screen doesn't. When this be something dangerous to the onscreen character the result can be a powerful "Don't do it!!!!"
The problem with Blair Witch is that, being filmed by the characters themselves, there is no way we can see things they don't also see. There is no way we can see the metaphorical knife being raised behind anyone's back. Or is there?
I suggest that two things could have been done.
One would have been to take advantage of the fact that they ran their video camera so much and let them leave it on when they don't intent to or to let them look up over the camera at something else as the camera (and audience) catches something the characters themselves miss.
The second idea would have been near the end to have the camera dropped...and picked up. It could have had a smoother or at least different style of operation. It could have moved in a searching way that would have completely reversed the audience's previous identification with the moving camera. A character could have turned, looked at whatever was now holding the camera, and been completely mortified. The very turn of events would have been so scary that no extra extra emoting would have been needed.
Add to that some careful thinking about how to tell a good story and this film could really have been worthy of all the hype and success.
-kb, the Kent who doesn't pretend to be good at telling a story, but he knows it can be done better than in The Blair Witch Project.