***The Prius batteries are well known to last 200K miles and more. And only the military Hummers last 300K miles the commercial version doesn't even come close.***
I think I have this right. There are really two commercial versions of the Hummer. One is very expensive and is the military version with some military stuff removed and a few ammenities added. Probably will last as long as the military version. The other is a typical GM SUV with a lot of Hummer-like sheet metal wrapped around it. Basically, a Suburban in a Hummer costume?
Will a GM SUV last for 300K miles? A lot depends on whether GM has actually overcome the quality problems that plauged them from about 1960 until the 1990s. They may have. I drove a low mileage GM SUV from Seattle to Vermont a few months ago (cheaper than a moving van) and it really seemed to be a far better vehicle than the godawful junkheaps I occasionally rented on business trips during the dark decades. But even if they have, I don't think the commercial SUV is likely to go 300K miles routinely unless it is used mostly for highway driving. So, I think 150000 miles might be a better estimate of Hummer mileage.
And $3.25 per mile for a Prius. That's $325,000 for 100000 miles, right? Not very plausible no matter how you figure costs.
... Maybe if you buy the vehicle with a payday loan from Benny the Bonebreaker.
Indeed it does. It failed frequently when I first saw it 46 years ago. And it still failed frequently when last I had to deal with it 3 years ago. Was it better in 2004.? Yes, a bit. Was it good enough to be a sole backup? No way. We went through five DDS tape drives in six years BTW. A pretty sorry performance any way you look at it.
*** and while some drives (TRAVAN) are pretty awful, they make tape systems today that are extremely reliable***
I doubt it. They make drives that vendors claim to be extremely reliable and some people believe those claims to be valid. I personally don't think the vendor measured bit-error-rates which indicate about one tape failure per geologic epoch have any real world connection with actual drive and tape cartridge failure modes or failure rates. Note that we are talking about the failure of one of those "extremely reliable" tapes.
***And what would you suggest?***
What I did was find a junker PC; install Linux on the original small drive; and add a big secondary drive (the BIOS in the 486SX33 was less than pleased with the idea of booting from a many GB drive). Then I simply backed up the server to the big hard drive over the network every night.
As a bonus, it was much easier and faster to recover files from the hard drive with cp rather than fighting with BACKUP EXEC (a truly abominable piece of software if you ask me) to get the files back from tape.
Since the world would not have actually ended had we lost all our data I settled for using a Tape for Offsite backup. Had the data been more important, I would have shunted removable hard drives back and forth. You're right, they are a bit fragile, but they will probably last as long as a tape cartridge and they don't cost that much more. In our case, they would probably have outlasted the tape DRIVES and would have been lots cheaper to replace. Would I use them for critical data? You bet I would. Are they overkill for data you'd like to keep but could afford to lose? Possibly.
I agree that saving more than will fit on the current biggest cheap hard drive is a problem. (It'd be even more of a mess with DVDs I should think). Maybe tape has some reasonable use if terabytes of data need to be backed up every night. But I have to tell you that the prospect of recovering terabytes of data from tape ought to make any sensible IT type really nervous.
And I haven't even mentioned RAID arrays. That's because I don't know much about them. I'd sure as hell look into them as an alternative/supplement to tape if I had to deal with huge amounts of critical data.
At a guess, Clear Channel doesn't really care how they get their advertising to listeners. They figure they can compete with two guys in a basement operations. What they don't want to compete with is sophisticated operations backed with real money and based in the Bahamas or Cayman Islands or Madagascar for that matter who are targeting listeners in their markets with ads for Joe's Bar and Grill and giving Joe cheaper rates than Clear Channel is getting.
Therefore, Clear Channel doesn't want to drive US Internet Radio offshore. Probably not the reason, but at least it's sort of plausible.
BTW, Clear Channel is said to be selling off its operations in small markets. Maybe if you can figure out why (I can't) the reason for their siding with the forces of light and virtue on Internet Radio royalty rates might be clearer.
***Hm. Tapes with a proven shelf life of many, many years, or DVDs where a single scratch can render 4GB of data worthless. I wonder which enterprises (or governments) should chose?***
Neither? The historical record is pretty clear. Depending on tape for your only copy of critical data is roughly equivalent to removing the bulb from the Check Engine light in your car. You'll probably get away with it, but you have only yourself to blame if it turns out that you have made an expensive mistake.
And no, I don't think writable DVDs are highly reliable storage either. Even if the write is OK, I don't think anyone knows what their lifetime for digital data is going to be. Judging from writable CDs, maybe not all that many years
.
Maybe "Both" is OK if a high probability of recovering the data from one media or the other is good enough.
***You reach for your back up tapes only to find out that the information on the tapes is unreadable.***
They depended solely on TAPE backup?... sigh... they aren't alone in that. But it's time proven really, really bad idea. If I had a dollar for every tape drive that I have encountered that was apparently working but was either creating unreadable tapes, or was omitting important files because of misconfiguration, I'd be a wealthy man. Tapes are fine for "we'll try to get it back, but we may not be able to" type storage. And there is a valid need for that kind of storage. But for the only copy of really important data?
Sure, other backup media can have problems also, but IMO tapes never have been and probably never will be reliable enough to be the sole backup media for anything really important.
And where, pray tell, was the off-site backup? Billions of dollars worth of Acoounts Payable data, and they don't have a copy off site? What is their plan if there is a fire, flood, explosion, etc?
***GIMP is great until you need to, say, draw a straight line. Then you get bounced around their website to a highly sarcastic tutorial [gimp.org] that makes it sound like ANY moron should know their weird shift-clicking technique with no explanation. And then OSS people say it's unfair that they have a rep for not being "user-friendly."/rant***
I just made my first attempt at using GIMP the other day. Overall, it seemed to work pretty well. But I had a lot more difficulty drawing a straight line than I thought I should have had. I assumed that the problem was my stupidity or maybe misconfiguration.
Knuth analyzes Linked Lists in some depth in Volume 3 of The Art of Computer Programming. Unfortunately, my copy isn't accessible at the moment.. But scanning on line, it seems that he only addresses single and doubly linked lists plus a "skip list" that allows rapid traversal of the list to an area of interest.
However, he does finish up with "A binary tree can be seen as a type of linked list where the elements are themselves linked lists of the same nature. The result is that each node may include a reference to the first node of one or two other linked lists, which, together with their contents, form the subtrees below that node."
I'm not sure that any of that addresses what the patent addresses. OTOH, I'd be hard put to guess what a triply linked list might be good for. Additional links intended to speed up access to the list clearly have prior art.
***Many people have roasted GWB for his apparent glaring shortcomings. But I bet his one lasting legacy will be his judicial appointments to the supreme court that may reign in copyright gone amok.***
I think you are assuming that when Lessig says "conservative" he means the same thing as when the news media says "conservative". Not so I strongly suspect. It is important to keep in mind that the Republican Party of today and its leader GWB are NOT political conservatives in any classical sense of the word -- their rhetoric notwithstanding. In fact and practice, they are probably better described as reactionaries or neofascists.
Chief Justice Roberts may in fact be what Lessing would call a judical conservative. Samuel Alito looks to be (yet another) right wing nut case. I would not assume that because these guys were appointed by someone who claims to be a "conservative", they will rule in a judicially conservative manner.
It's not clear to me where the neoconservative perception of reality (which is frequently really, really wierd IMHO) is going to come down in Viacom vs U-Tube. I wouldn't bet on it coinciding with congressional will, common sense, or any other rational critereon.
Do Roberts and Alito have a track record on Copyright? Do you happen to know what it is?
Your natural inclination will be to take too much. When I set off to see the world in 1965, I took 3 or 4 suitcases. I shipped all but one of them back from Sydney. When I went off a decade later to spend a year in Japan, I took one suitcase and an athletic bag. More than enough.
If I were setting out to backpack someplace, I'd take what will fit in a backpack -- underware, socks, toothbrush, toothpaste, a real razor with some blades, a bar of soap, a towel. a couple of shirts and a spare pair of pants. I'd wear a good sturdy light jacket and carry at least one heavy wool sweater and one light cotton one. Then I'd throw in a pocket knife with a screwdriver blade, a versatile, compact pair of pliers, about a meter of soft steel wire, a little string. a compact, battery powered FM radio. A bit of thread, a needle and a couple of spare buttons might be a good idea. If I could get ahold of some broad spectrum antibiotics that aren't temperature sensitive, I'd take a dozen or so tablets along and a dozen asprin. A few bandaids. Maybe a few matches. Some sort of cheap sturdy camera that doesn't need wall power. And maybe a cell phone if I could figure out what will work in foreign climes, and I could figure out how to charge it without carrying two pounds of transformers and adapters. Two or three books -- at least one of them serious enough so you can resonably reread it once or twice while your are waiting for a bus or ferry that runs only when the weather and stars are propitious.
Hell, I'd even leave the duct tape at home.
Computer? Not this decade. Maybe in some future time when a serious PC will run for a week off an AAA cell. occupies about 20cc of space and weighs 200g or so.
GPS? Sure... If you are going to hitchike across the Nullarbor plain, Sahara, or explore Tibet. Otherwise, plan on asking people where you are.
***Trees aren't a big deal for TV signals though. I know several people that live in remote wooded areas that can get decent TV reception with reasonable antennas that don't extend above the trees.***
Good point. I was thinking of smaller antennae and higher frequencies of course. Silly me.
Still though, TV coverage often isn't all that good in hilly country with or without trees -- "one and a half channels"
***Is Tivoization [wikipedia.org] part of the consideration?***
Probably, but so is the fact that GPLed TIVO source code is (presumably) accessible even if you can't induce TIVO's hardware to run a modified version. One important use of source code that is rarely discussed is that examining the code is sometimes the only way to figure out why something that should work doesn't. Typically, it will turn out that you just have to do things a little different -- in some fashion that no sentient creature is likely to guess -- in order to do what you want to do.
***Several analysts said a TV-spectrum system might make the most sense in rural areas, where high-speed Internet access via phone or cable lines is expensive to deploy. Small companies might build some towers, beam white-space spectrum to farm homes and cabins, and connect it to an Internet provider, they said.***
A few years ago when we were looking at ways to bring broadband to a rural school in Vermont, I trecked up to the highest point we could reasonably put an antenna. What I saw was trees -- hundreds of trees. Maybe thousands of trees. It was pretty clearly going to take us several intermediate relays to get to a place where we could connect to existing broadband. And each intermediate was going to need power and access and probably a tower to get above the trees. Scratch that idea.
I think that using TV frequencies for broadband wireless may be a workable idea in the plains and Great Basin. I've managed to raise a cell phone signal in some pretty unlikely places out in the west. But I don't think it is going to work very well in areas East of the Mississippi since most of the potential users are going to be in valleys and surrounded by trees. And no, cell phones didn't work at the school although there was a spot out at the end of the driveway and a couple of hundred yards down the road where one could raise a couple of bars if you held the phone just right.
(Thanks to a peculiarity in the local regulatory structure, we were finally able to get a T1 at reasonable rates.)
***Flash is now fairly cheap, it's in widespread use and it's a known quantity. Good luck trying to replace it.***
According to the article, this PRAM stuff can be configured to be flash compatible. IF it works out -- big IF, most of these wonders don't -- it might turn out to be a faster, drop in, replacement for flash.. If so, and if it isn't significantly more expensive than flash,it wouldn't be suprising to see PRAM replace flash in new products over the period of a few years.
***Frankly, I'm perplexed that anyone would pass on the opportunity to try out a free (as in beer) OS. Except gaming junkies, of course, but I think that with the maturing PC userbase they've become less relevant. Or maybe I'm just getting old...***
I suspect that the average PC user would have a vague discomfort deep in his belly that the free new OS might not work, and that the old OS might not work either when the new OS was removed.
***Linux may not be just a programmer's OS, but the Ubuntu flavor of Linux, IMO, isn't a very good programmer's OS at all. I think it crossed that fine line between control and ease of use.***
What's the problem? Either run Ubuntu from the ISO for a while or install it in a temporary partition. Then you can devote your spare time for the next three weeks to installing Slackware package by package in another partition one missing library at a time. Given any luck at all, you should have Slack running perfectly before the first (American) Football game of the 2007 season. Lest anyone think I'm making fun of Slackware, that's not my intent. I'm actually quite fond of Slack. I just picked it because if has a reputation for being somewhat purer than some other popular PC Unixes.
Seriously Unbuntu (or Kunbuntu actually) has the usual six consoles and python, perl and gcc seem to be there. In what way is it not a programmer's OS?
***Have you looked at the HTML code on Google's front page? It's not built to be valid, it's built to load fast.Have you looked at the HTML code on Google's front page? It's not built to be valid, it's built to load fast.***
Not faulting Google. Fact is their web page is minimal and they deliver a usable search page on every browser, I've ever tried which is no small accomplishment. However, I'm not sure whether they understand standards but think other things are more important -- or whether they just hacked out an effective web page by trial and error. My experience with my own web pages has been that the my standards problems are due to not thinking, or occasionally to not understanding (e.g. it never crossed my mind that ampersands would be translated within PRE/PRE brackets). They wouldn't affect performance. Not sure about Google. They are clearly smarter than I am, so as far as I'm concerned they can do whatever they want with no whinging from me.
First of all, writing a simple GUI application using say Python and TKinter is probably easier than writing a web application. I'm sure the same is true of Ruby, Perl, etc. Or Visual Basic for that matter although VB's database interface (at least in VB3) was so obtuse that I decided to find another language. All of those languages will handle the Event interfaces relatively gracefully.
Second even the localhost (127.0.0.1) interface is likely to be a bit jerky.
Third, No two browsers will render HTML beyond the "hello world" level consistently. Conceptually, that shouldn't matter, but if your input boxes don't appear or line up with inappropriate material in the page display, you can end up tinkering with your application well beyond what you originally envisioned.
Fourth, Browsers cache web pages. They don't always figure out that the page you have requested has changed. It looks to me like NOCACHE statements in HTML pretty much don't work. They may work when used in the HTTP (1.0 or later, right?) header, but getting them there may be non-trivial. This is not a big deal if you are the only user and understand caching since all browsers allow you to force a page reread. But it is not going to work out well with ordinary users.
I'd say that there is a place for simple web applications. But there are a lot of situations where alternative solutions are probably going to be more usable or simpler than a web browser, server, and CGI.
So, CGI is a perfectly OK tool, and maybe it belongs in the toolkit. But it's by no means universally the best solution.
*** 'software engineering' may be an academic discipline, but 'professional' (in their execution) software engineers are few and far between***
Software engineering is nearly totally lacking in discipline. (and yes, that's a slightly different meaning of the word). It's also lacking in any meaningful theoretical underpinnings in most areas. Exceptions, and there aren't many, would be a few things like cryptology that are rarely controversial.
Example 1: Given an interrupt driven, priority scheduled OS and a non-trivial defined set of tasks, can anyone tell me how often a given task will execute? Will it execute at all? What is the longest possible time between executions? Answer. AFAIK, None of those things can be calculated if the CPU loading is heavy. I wouldn't dwell on this too much before your next airplane trip.
The seminal work that is supposed to tell us what we know about programming computers is surely Knuth's "The Art of Computer Programming". It's really very good. But only half of it has ever been published and that half is decades out of date.
I looked at computer science extensively in the early 1980s when I still had 20 years of work in my future. I thought, that it was time to actually learn the trade rather than just coping with the problem at hand. My conclusion after three years of reading and cogitating. There are computer scientists -- people who are seeking knowledge. There are isolated islands of knowledge. There is not yet any connected body of knowledge that actually represents a coherent "Computer Science".
I don't think things have changed any in the past two decades.
IMO, Software Engineering is no more engineering than 'Political Science' is 'science'. It can't be. There is no meaningful body of knowledge and theory underlying it. For that matter, the body of practice is pretty chaotic. Seems to me that today -- on for the forseeable future -- certifying "Professional Software Engineers" would be as much a waste of time as licensing "Professional Phrenological Engineers" or "Certified Psychics"
Just my opinion. Feel free to differ. But try to back up your thoughts with reference to some body of serious paper and -- if possible -- some math that a paracticioner could use meaningfully to predict performance, reliability, stability, costs, etc.
***Why are they charging that amount for a patch?***
Because they are now obligated to fix a substantial set of problems if the patch doesn't work right. These time of day things are notoriously touchy with unexpected consequences.
The problem is due to congressional dementia, not to any action of Microsoft's, so MS really is not obligated to fix it for free.
Nonetheless, I agree this is something that Microsoft could have done for free. The revenue is probably not all that great, and MS could use some customer goodwill. It's not like MS is struggling to make payroll, and it's not like they are widely loved by their customers -- even those beyond the Slashdot universe.
***Those guys can't even put down proper HTML, I'm not sure i'd trust them to write a whole web-based "OS" in XML***
In their defense:
The guy doing their web page is probably not one of the folks doing the applications.
It appears that a LOT of Web Page developers are totally unaware of standards -- or don't care. My guess is that seriously non-compliant web pages probably outnumber those that are valid on major web sites. Hell, many of them don't even have a DOCTYPE spec and can't be validated.
Last time I looked, the Google home page threw about 50 HTML errors when fed to the W3C validator.
That said, if I were these guys, I'd fix the HTML.
Does this represent a threat to Google or Microsoft? Not any time Soon
But then, it's not that long ago that Google was just two guys doodling on scrap paper.
A few problems have to be overcome including internet latency and the tendancy of everyone to cache stuff they should not be putting in caches (If your PC's memory cache worked like Internet caches do, you'd be lucky to get a Solitaire hand dealt before the PC crashed.)
And I doubt this is a threat to Google because they will do the same thing it if it works out.
My impression is that what's good about this specific scheme is that only data is sent over the network, so the annoying latency issues many of us have with Google spreadsheets and Writely should be less of a problem.
What's bad is that the data is stored on someone's servers. Security will be an issue. So will availability. And loss of data. And...
Another problem is that networked "OS"es may not be acceptable for a lot of users because they are just plain too damn slow. A few years ago I slapped together a networked application running on a server here at home for keeping notes together. Worked, sorta. But even though I owned the network and the application was built into server code, not run via CGI, it was too slow to be usable. The problem looked to be latency, not slow processing.
The few serious attempts I've seen at using HTTP/browsers to do real jobs varied from awful to marginal. IMHO even things like SAIL suck. I'd rather update the/etc files directly. Hell, even ed/EDLINE would be faster and more satisfactory.
Maybe the problems can be overcome with brains, technology, and money. Maybe they can't.
Back on topic. Is this stuff a threat to Microsoft? You just bet it is. MS makes most of its money off OK, but overpriced, products that do way more than most customers need (Exception--Xbox which may eventually be a real, money making operation with a bright future). Furthermore, adding more features and charging more for new versions of Windows/Office is probably an unsustainable strategy. We're already seeing geeks and a few organizations walking away from Microsoft. I think that is only going to become more common and some of them may well go to schemes like this.
***That is the important point here. There is no reason for many of these programs to be asking for 'administrative' access to do any of this shit. MS can't just cut it off b/c it will break most of it's install base. This is a way to guide software companies into writing programs with a thought to security, rather than just doing it the 'easy way'.***
That's really a very good point. Microsoft REALLY needs to set up a decent security model that developers can understand and accomodate. And I think this is one of the very rare cases where backwards compatibility is not all that important because it is likely that most legacy programs are going to be insecure. There has to be an "insecure mode" to handle them since many useful/necessary programs aren't going to be updated for years -- if ever. But new software should comply with the security model.
I'm far from convinced that the unix/1970s mainframe model is optimal for a single user PERSONAL computer. It's sure better than nothing, but it's mostly designed to keep users in a multiuser system from examining/altering each other's files. It may be overkill in some respects and inadquate in others. I also have real doubt whether the "services" architecture in both unix and NT is really all that secure. It seems to me (I'm by no means a security expert) that a very high percentage of attacks on Windows and Linux are attacks on communictions service providers that either run with admin priviliges or whose privilege level can be escalated. How, exactly, does file oriented "security" protect against services problems?
Anyway, I think, MS needs to go off and come up with a new, well thought out, security model. Then they need to test it -- extensively. Finally, they need to lay it out for developers -- both inside and outside the company. I don't think the can do that in less than three or four years. Maybe more. And developers need years -- not weeks or months -- to respond.
So, I'd guess that real security in Windows may well be two major releases away. Maybe in 2011 (which will inevitably slip to 2012 or 2013).
Slashdot Mirror
I think I have this right. There are really two commercial versions of the Hummer. One is very expensive and is the military version with some military stuff removed and a few ammenities added. Probably will last as long as the military version. The other is a typical GM SUV with a lot of Hummer-like sheet metal wrapped around it. Basically, a Suburban in a Hummer costume?
Will a GM SUV last for 300K miles? A lot depends on whether GM has actually overcome the quality problems that plauged them from about 1960 until the 1990s. They may have. I drove a low mileage GM SUV from Seattle to Vermont a few months ago (cheaper than a moving van) and it really seemed to be a far better vehicle than the godawful junkheaps I occasionally rented on business trips during the dark decades. But even if they have, I don't think the commercial SUV is likely to go 300K miles routinely unless it is used mostly for highway driving. So, I think 150000 miles might be a better estimate of Hummer mileage.
And $3.25 per mile for a Prius. That's $325,000 for 100000 miles, right? Not very plausible no matter how you figure costs.
Indeed it does. It failed frequently when I first saw it 46 years ago. And it still failed frequently when last I had to deal with it 3 years ago. Was it better in 2004.? Yes, a bit. Was it good enough to be a sole backup? No way. We went through five DDS tape drives in six years BTW. A pretty sorry performance any way you look at it.
*** and while some drives (TRAVAN) are pretty awful, they make tape systems today that are extremely reliable***
I doubt it. They make drives that vendors claim to be extremely reliable and some people believe those claims to be valid. I personally don't think the vendor measured bit-error-rates which indicate about one tape failure per geologic epoch have any real world connection with actual drive and tape cartridge failure modes or failure rates. Note that we are talking about the failure of one of those "extremely reliable" tapes.
***And what would you suggest?***
What I did was find a junker PC; install Linux on the original small drive; and add a big secondary drive (the BIOS in the 486SX33 was less than pleased with the idea of booting from a many GB drive). Then I simply backed up the server to the big hard drive over the network every night.
As a bonus, it was much easier and faster to recover files from the hard drive with cp rather than fighting with BACKUP EXEC (a truly abominable piece of software if you ask me) to get the files back from tape.
Since the world would not have actually ended had we lost all our data I settled for using a Tape for Offsite backup. Had the data been more important, I would have shunted removable hard drives back and forth. You're right, they are a bit fragile, but they will probably last as long as a tape cartridge and they don't cost that much more. In our case, they would probably have outlasted the tape DRIVES and would have been lots cheaper to replace. Would I use them for critical data? You bet I would. Are they overkill for data you'd like to keep but could afford to lose? Possibly.
I agree that saving more than will fit on the current biggest cheap hard drive is a problem. (It'd be even more of a mess with DVDs I should think). Maybe tape has some reasonable use if terabytes of data need to be backed up every night. But I have to tell you that the prospect of recovering terabytes of data from tape ought to make any sensible IT type really nervous.
And I haven't even mentioned RAID arrays. That's because I don't know much about them. I'd sure as hell look into them as an alternative/supplement to tape if I had to deal with huge amounts of critical data.
Therefore, Clear Channel doesn't want to drive US Internet Radio offshore. Probably not the reason, but at least it's sort of plausible.
BTW, Clear Channel is said to be selling off its operations in small markets. Maybe if you can figure out why (I can't) the reason for their siding with the forces of light and virtue on Internet Radio royalty rates might be clearer.
Neither? The historical record is pretty clear. Depending on tape for your only copy of critical data is roughly equivalent to removing the bulb from the Check Engine light in your car. You'll probably get away with it, but you have only yourself to blame if it turns out that you have made an expensive mistake.
And no, I don't think writable DVDs are highly reliable storage either. Even if the write is OK, I don't think anyone knows what their lifetime for digital data is going to be. Judging from writable CDs, maybe not all that many years .
Maybe "Both" is OK if a high probability of recovering the data from one media or the other is good enough.
They depended solely on TAPE backup? ... sigh ... they aren't alone in that. But it's time proven really, really bad idea. If I had a dollar for every tape drive that I have encountered that was apparently working but was either creating unreadable tapes, or was omitting important files because of misconfiguration, I'd be a wealthy man. Tapes are fine for "we'll try to get it back, but we may not be able to" type storage. And there is a valid need for that kind of storage. But for the only copy of really important data?
Sure, other backup media can have problems also, but IMO tapes never have been and probably never will be reliable enough to be the sole backup media for anything really important.
And where, pray tell, was the off-site backup? Billions of dollars worth of Acoounts Payable data, and they don't have a copy off site? What is their plan if there is a fire, flood, explosion, etc?
I just made my first attempt at using GIMP the other day. Overall, it seemed to work pretty well. But I had a lot more difficulty drawing a straight line than I thought I should have had. I assumed that the problem was my stupidity or maybe misconfiguration.
Maybe not. Thanks.
However, he does finish up with "A binary tree can be seen as a type of linked list where the elements are themselves linked lists of the same nature. The result is that each node may include a reference to the first node of one or two other linked lists, which, together with their contents, form the subtrees below that node."
I'm not sure that any of that addresses what the patent addresses. OTOH, I'd be hard put to guess what a triply linked list might be good for. Additional links intended to speed up access to the list clearly have prior art.
I think you are assuming that when Lessig says "conservative" he means the same thing as when the news media says "conservative". Not so I strongly suspect. It is important to keep in mind that the Republican Party of today and its leader GWB are NOT political conservatives in any classical sense of the word -- their rhetoric notwithstanding. In fact and practice, they are probably better described as reactionaries or neofascists.
Chief Justice Roberts may in fact be what Lessing would call a judical conservative. Samuel Alito looks to be (yet another) right wing nut case. I would not assume that because these guys were appointed by someone who claims to be a "conservative", they will rule in a judicially conservative manner.
It's not clear to me where the neoconservative perception of reality (which is frequently really, really wierd IMHO) is going to come down in Viacom vs U-Tube. I wouldn't bet on it coinciding with congressional will, common sense, or any other rational critereon.
Do Roberts and Alito have a track record on Copyright? Do you happen to know what it is?
If I were setting out to backpack someplace, I'd take what will fit in a backpack -- underware, socks, toothbrush, toothpaste, a real razor with some blades, a bar of soap, a towel. a couple of shirts and a spare pair of pants. I'd wear a good sturdy light jacket and carry at least one heavy wool sweater and one light cotton one. Then I'd throw in a pocket knife with a screwdriver blade, a versatile, compact pair of pliers, about a meter of soft steel wire, a little string. a compact, battery powered FM radio. A bit of thread, a needle and a couple of spare buttons might be a good idea. If I could get ahold of some broad spectrum antibiotics that aren't temperature sensitive, I'd take a dozen or so tablets along and a dozen asprin. A few bandaids. Maybe a few matches. Some sort of cheap sturdy camera that doesn't need wall power. And maybe a cell phone if I could figure out what will work in foreign climes, and I could figure out how to charge it without carrying two pounds of transformers and adapters. Two or three books -- at least one of them serious enough so you can resonably reread it once or twice while your are waiting for a bus or ferry that runs only when the weather and stars are propitious.
Hell, I'd even leave the duct tape at home.
Computer? Not this decade. Maybe in some future time when a serious PC will run for a week off an AAA cell. occupies about 20cc of space and weighs 200g or so.
GPS? Sure ... If you are going to hitchike across the Nullarbor plain, Sahara, or explore Tibet. Otherwise, plan on asking people where you are.
Good point. I was thinking of smaller antennae and higher frequencies of course. Silly me.
Still though, TV coverage often isn't all that good in hilly country with or without trees -- "one and a half channels"
Probably, but so is the fact that GPLed TIVO source code is (presumably) accessible even if you can't induce TIVO's hardware to run a modified version. One important use of source code that is rarely discussed is that examining the code is sometimes the only way to figure out why something that should work doesn't. Typically, it will turn out that you just have to do things a little different -- in some fashion that no sentient creature is likely to guess -- in order to do what you want to do.
A few years ago when we were looking at ways to bring broadband to a rural school in Vermont, I trecked up to the highest point we could reasonably put an antenna. What I saw was trees -- hundreds of trees. Maybe thousands of trees. It was pretty clearly going to take us several intermediate relays to get to a place where we could connect to existing broadband. And each intermediate was going to need power and access and probably a tower to get above the trees. Scratch that idea.
I think that using TV frequencies for broadband wireless may be a workable idea in the plains and Great Basin. I've managed to raise a cell phone signal in some pretty unlikely places out in the west. But I don't think it is going to work very well in areas East of the Mississippi since most of the potential users are going to be in valleys and surrounded by trees. And no, cell phones didn't work at the school although there was a spot out at the end of the driveway and a couple of hundred yards down the road where one could raise a couple of bars if you held the phone just right.
(Thanks to a peculiarity in the local regulatory structure, we were finally able to get a T1 at reasonable rates.)
According to the article, this PRAM stuff can be configured to be flash compatible. IF it works out -- big IF, most of these wonders don't -- it might turn out to be a faster, drop in, replacement for flash.. If so, and if it isn't significantly more expensive than flash,it wouldn't be suprising to see PRAM replace flash in new products over the period of a few years.
I suspect that the average PC user would have a vague discomfort deep in his belly that the free new OS might not work, and that the old OS might not work either when the new OS was removed.
AFAICS, that'd be an entirely reasonable concern.
What's the problem? Either run Ubuntu from the ISO for a while or install it in a temporary partition. Then you can devote your spare time for the next three weeks to installing Slackware package by package in another partition one missing library at a time. Given any luck at all, you should have Slack running perfectly before the first (American) Football game of the 2007 season. Lest anyone think I'm making fun of Slackware, that's not my intent. I'm actually quite fond of Slack. I just picked it because if has a reputation for being somewhat purer than some other popular PC Unixes.
Seriously Unbuntu (or Kunbuntu actually) has the usual six consoles and python, perl and gcc seem to be there. In what way is it not a programmer's OS?
Was there some other kind of jumper?
Not faulting Google. Fact is their web page is minimal and they deliver a usable search page on every browser, I've ever tried which is no small accomplishment. However, I'm not sure whether they understand standards but think other things are more important -- or whether they just hacked out an effective web page by trial and error. My experience with my own web pages has been that the my standards problems are due to not thinking, or occasionally to not understanding (e.g. it never crossed my mind that ampersands would be translated within PRE /PRE brackets). They wouldn't affect performance. Not sure about Google. They are clearly smarter than I am, so as far as I'm concerned they can do whatever they want with no whinging from me.
First of all, writing a simple GUI application using say Python and TKinter is probably easier than writing a web application. I'm sure the same is true of Ruby, Perl, etc. Or Visual Basic for that matter although VB's database interface (at least in VB3) was so obtuse that I decided to find another language. All of those languages will handle the Event interfaces relatively gracefully.
Second even the localhost (127.0.0.1) interface is likely to be a bit jerky.
Third, No two browsers will render HTML beyond the "hello world" level consistently. Conceptually, that shouldn't matter, but if your input boxes don't appear or line up with inappropriate material in the page display, you can end up tinkering with your application well beyond what you originally envisioned.
Fourth, Browsers cache web pages. They don't always figure out that the page you have requested has changed. It looks to me like NOCACHE statements in HTML pretty much don't work. They may work when used in the HTTP (1.0 or later, right?) header, but getting them there may be non-trivial. This is not a big deal if you are the only user and understand caching since all browsers allow you to force a page reread. But it is not going to work out well with ordinary users.
I'd say that there is a place for simple web applications. But there are a lot of situations where alternative solutions are probably going to be more usable or simpler than a web browser, server, and CGI.
So, CGI is a perfectly OK tool, and maybe it belongs in the toolkit. But it's by no means universally the best solution.
Software engineering is nearly totally lacking in discipline. (and yes, that's a slightly different meaning of the word). It's also lacking in any meaningful theoretical underpinnings in most areas. Exceptions, and there aren't many, would be a few things like cryptology that are rarely controversial.
Example 1: Given an interrupt driven, priority scheduled OS and a non-trivial defined set of tasks, can anyone tell me how often a given task will execute? Will it execute at all? What is the longest possible time between executions? Answer. AFAIK, None of those things can be calculated if the CPU loading is heavy. I wouldn't dwell on this too much before your next airplane trip.
The seminal work that is supposed to tell us what we know about programming computers is surely Knuth's "The Art of Computer Programming". It's really very good. But only half of it has ever been published and that half is decades out of date.
I looked at computer science extensively in the early 1980s when I still had 20 years of work in my future. I thought, that it was time to actually learn the trade rather than just coping with the problem at hand. My conclusion after three years of reading and cogitating. There are computer scientists -- people who are seeking knowledge. There are isolated islands of knowledge. There is not yet any connected body of knowledge that actually represents a coherent "Computer Science".
I don't think things have changed any in the past two decades.
IMO, Software Engineering is no more engineering than 'Political Science' is 'science'. It can't be. There is no meaningful body of knowledge and theory underlying it. For that matter, the body of practice is pretty chaotic. Seems to me that today -- on for the forseeable future -- certifying "Professional Software Engineers" would be as much a waste of time as licensing "Professional Phrenological Engineers" or "Certified Psychics"
Just my opinion. Feel free to differ. But try to back up your thoughts with reference to some body of serious paper and -- if possible -- some math that a paracticioner could use meaningfully to predict performance, reliability, stability, costs, etc.
Because they are now obligated to fix a substantial set of problems if the patch doesn't work right. These time of day things are notoriously touchy with unexpected consequences.
The problem is due to congressional dementia, not to any action of Microsoft's, so MS really is not obligated to fix it for free.
Nonetheless, I agree this is something that Microsoft could have done for free. The revenue is probably not all that great, and MS could use some customer goodwill. It's not like MS is struggling to make payroll, and it's not like they are widely loved by their customers -- even those beyond the Slashdot universe.
Is it going to cost another $4000 per product to back the change out?
In their defense:
That said, if I were these guys, I'd fix the HTML.
But then, it's not that long ago that Google was just two guys doodling on scrap paper.
A few problems have to be overcome including internet latency and the tendancy of everyone to cache stuff they should not be putting in caches (If your PC's memory cache worked like Internet caches do, you'd be lucky to get a Solitaire hand dealt before the PC crashed.)
And I doubt this is a threat to Google because they will do the same thing it if it works out.
My impression is that what's good about this specific scheme is that only data is sent over the network, so the annoying latency issues many of us have with Google spreadsheets and Writely should be less of a problem.
What's bad is that the data is stored on someone's servers. Security will be an issue. So will availability. And loss of data. And ...
Another problem is that networked "OS"es may not be acceptable for a lot of users because they are just plain too damn slow. A few years ago I slapped together a networked application running on a server here at home for keeping notes together. Worked, sorta. But even though I owned the network and the application was built into server code, not run via CGI, it was too slow to be usable. The problem looked to be latency, not slow processing.
The few serious attempts I've seen at using HTTP/browsers to do real jobs varied from awful to marginal. IMHO even things like SAIL suck. I'd rather update the /etc files directly. Hell, even ed/EDLINE would be faster and more satisfactory.
Maybe the problems can be overcome with brains, technology, and money. Maybe they can't.
Back on topic. Is this stuff a threat to Microsoft? You just bet it is. MS makes most of its money off OK, but overpriced, products that do way more than most customers need (Exception--Xbox which may eventually be a real, money making operation with a bright future). Furthermore, adding more features and charging more for new versions of Windows/Office is probably an unsustainable strategy. We're already seeing geeks and a few organizations walking away from Microsoft. I think that is only going to become more common and some of them may well go to schemes like this.
I dunno. Since Vista seems to offer not one single feature that most people want, could it not be viewed as costing $109 too much?
That's really a very good point. Microsoft REALLY needs to set up a decent security model that developers can understand and accomodate. And I think this is one of the very rare cases where backwards compatibility is not all that important because it is likely that most legacy programs are going to be insecure. There has to be an "insecure mode" to handle them since many useful/necessary programs aren't going to be updated for years -- if ever. But new software should comply with the security model.
I'm far from convinced that the unix/1970s mainframe model is optimal for a single user PERSONAL computer. It's sure better than nothing, but it's mostly designed to keep users in a multiuser system from examining/altering each other's files. It may be overkill in some respects and inadquate in others. I also have real doubt whether the "services" architecture in both unix and NT is really all that secure. It seems to me (I'm by no means a security expert) that a very high percentage of attacks on Windows and Linux are attacks on communictions service providers that either run with admin priviliges or whose privilege level can be escalated. How, exactly, does file oriented "security" protect against services problems?
Anyway, I think, MS needs to go off and come up with a new, well thought out, security model. Then they need to test it -- extensively. Finally, they need to lay it out for developers -- both inside and outside the company. I don't think the can do that in less than three or four years. Maybe more. And developers need years -- not weeks or months -- to respond.
So, I'd guess that real security in Windows may well be two major releases away. Maybe in 2011 (which will inevitably slip to 2012 or 2013).
Isn't THAT a cheery thought? Hope that I'm wrong.