You'd be shocked to find that these machines are still being made, keeping up with the latest technology, and faster and more reliable than the best cluster systems.
They arn't so much faster and more reliable then the latest cluster systems so much as actually being the latest cluster systems. Oh, and being far more reliable (lots of check logic, ECC on the cache, and register files, not just main memory). Not so much the faster part though. The CPUs only run a few hundrad Mhz (last year 300Mhz to 400Mhz was extreamly fast for them). Of corse they have dedicated IO processors, and small tens of CPUs is a common size.
Most of the innovations in clustering in the micro world is re-inventing what mainframes have allready been doing. Then again so was having caches, and out-of-order execution. That's not to say micros didn't invent something, or that finding the right time to re-use what had gone before isn't hard in and of itself.
Re:A cautionary tale of Web design
on
Boo No More
·
· Score: 2
IIRC, the original boo.com site was overblown with too many images, JavaScript, probably Flash and lots of other junk.
It definitly had Flash. It had a little virtual you that would put on the clothes and let you pan around. It had a little virtual Ms. Boo that would comment (allways positave empty comments). And it was dog slow. And not from anything on our end.
Oh, and the virtual you? You can't tell what the damm clothes look like on it anyway. Your better off looking at the models (so you can see how the fabric really drapes).
Even if boo.com failed for other reasons, I like to think of the total lack of usability testing as a contributing factor. Or total lack to listen to the testing.
If the "bubble being over" means really bad sites like this dry up, I'm all for it.
This wouldn't work on every company, because for some of them, the phrase "publicly traded" is a joke -- they're privately held by other corporations/individuals who bought the stock and will never sell it. OTOH, many of those companies who own companies have a lot of public shares outstanding....
True for tracking stocks (like AWE), not so true for other stocks. It is true for tracking stocks because there is no way for the parent compony to actually sell the asset, not because they wouldn't want too. They may even spin off the unit if the tracking stock goes high enough. Maybe.
For a non-tracking stock, you better beleve that a corp will sell off it's shares in another once the price excedes their percieved value, and the owning corp has reason to beleve the price has peaked (i.e. it's on it's way down now), or (the seller) needs a (profit) boost.
They have their own stock holders to answer too, and that is one of the (few) things a corp must must must do.
Remember though, perceved value isn't just money. Apple may not want to sell shares of Earthlink (assuming it has any) because as long as it owns a big part of them it can exert extra presure to make sure Earthlink keeps supporting Apple products. That is the same non-monitory value you are intrested in, so it's not supprising other have allready discovered it.
What I don't understand is why make the secondary cache smaller than the primary cache?
The L2 duron cache is exclusave, that is does not contain cache lines from the L1 cache. Many other processors don't work this way, in part because it is simpler to make a multi-processer system if only the L2 caches need to do bus-snooping. My guess is the Duron won't be too happy in a multi-processor system for this reason.
It'll still make a decent single CPU system, especally at the dirt cheap prices listed in the last article.
Do you really think SCO's shareholders would read that interview on slashdot to see if he says something they don't like ?
Not very many of them. But some may. And reporters and stock analists read slashdot (I don't know if it is just a few, or lots).
But not many need too. If CNNFN's reporter sees a nice jucy quote, it'll go somewhere SCO's shareholders do read. If it is east-bumfuck's reporter, and they print it, you can bet other reporters will reprint it later.
What if the offender had been another open source project? If they had used a BSD license instead of the GPL, for instance. Would they have been forced to change to the GPL? Would anyone have cared?
Well, if the offender was going to BSDL, or MPL their code, they would probbably realise that it would be examined by people who knew the borrowed code was GPLed, and not have made the "mistake".
In fact they would almost certonally understand the "rules" and send mail to the orignal author saying "I'm doing a BSDL'ed project, can I re-release some of your code from foo.c under the BSDL?".
Of corse the original author is free to say "no, I don't want anyone to take it from your code and slap it into Windows NT". So, they could be mostly outta-luck.
They could BSDL their code except for the foo.c code which remains under GPL. Or they could re-write it. Lots of OSS is written for the fun of writing, so maybe they woulnd't even want to borrow from foo.c, they might want to write a better foo.c, and mail the GPL'ed code author "Too bad you wouldn't let us re-licence that code, but here is another better version we wrot from scratch, feel free to ship it with your code".
If they went and violated the GPL anyway, I expect they would have gotten the same polite treatment nVidia got. Probbably.
I'll take you to court (without being able to plead damages, because no money was lost.)
What makes you think money is the only kind of damage? If I claim an Arthor Clark short story is mine when I read it to friends, it damages him. He does not get the gain in reputation his work has earned him. If I were to call someone else a wife beater, there would be no money involved, but there would be damages (their reputation, and possably my face just after they punch me).
If you break into my home and steal my VCR, and yet leave me $200 (the purchase price) there is no monatary damage, yet it leaves me with a feeling of insecurity. You could break in and maybe not leave money. Or take a life rather then properity next time.
Damages are not just money.
A lawyer would be able to convince a jury. Easy.
Now whether a lawyer would be able to convince a jury that this wasn't an honest mistake, or that 200 lines of stolen code in a 10,000 project is a super-huge deal, well, I don't know about that.
I would rather have the GPL tested against a violation that is clearly not a mistake, and is clearly a large chunk of code. And best of all, on a clearly wealthy and unrepentent violator, since US juries seem to wack them the hardest.
Re:nVidia has incentive to remain closed-source...
on
GPL Violation - NVIDIA
·
· Score: 2
Then why not just make it public domain?
Public Domain grants users some rights that BSDL doesn't. For example BSDL specifically disclaims any warranty, and makes a big deal out of it. "Public Domain" does not. So if you PD some software, and it doesn't work you stand a large chance of being sued, and a larger chance of being found liable. In my opinion a very small chance either way, but why take the risk?
Now in my opinion the original author who put his code under the BSDL because he figured he couldn't prevent violations is giving up. Personally I would have MPL'ed it and put an attachment on saying "there is a flat $100 fee to gain commercial license to any or all of this code". Most companies would rather pay $100 for something then steal.
That's not to say that people that put things under the BSDL under the belief that that license does the most good are wrong. If you BSDL something because you think it is good solid code, and you would rather see it used as the basis of a commercial work then a buggy chunk of crap, and have no desire to be payed (directly at least) for the code, then the BSDL is fine.
If you think the path to better software (or whatever it is you want) is better served by the GPL, or the MPL, well go for it. I'm kind of fond of the MPL myself, especially the bit about patents (it would be nice if a future GPL would adopt that section).
You don't use the GPL in order to restrict people from making money off your work. You do it in order to see that any improvements other people make to your code are also open.
No, that is what the "Lesser GPL" (formerly Library GPL) does. The (greater?) GPL both keeps improvements free, and prevents most use in non-open software (in fact, not just non-open, but non-GPL).
You can not use GPL'ed code in a open source project that uses a non-GPL license. It's the same for MPL'ed code too. Or to be a bit more accurate, you can't use GPL/MPL'ed code as anything other then a distinct program. I could have a non-GPL'ed juke box that uses a GPL'ed mp3 player if I pipe my songs to it. If I want to link it in (so I can fast forward and rewind) I'm out of luck with a GPL'ed MP3 player.
If that is the intent of the author of the GPL code, then all is well with the world. After all it is their work, and they have some rights over how it should be used. If it was the intent of the author that GPLing the code allows almost unrestricted sharing, then they have made a grave mistake. They should have used a different license, like the LGPL. Or put in an escape clause for integration by other Open Source software.
Re:Bad wording, and benchmark link
on
Athlons Sold Out
·
· Score: 2
The key words here are three months. If the new chips are going to be out next month (as some AMD luvers here are saying), we would be hearing all about them right now.
Well, it's less then a week later, and we are hearing about them. So I guess they could be out next month. Or a bit after.
So soone I guess we will know how the perform. I didn't see any AMD talk about the Thunderbird, so maybe the Spitfire isn't faster then the existing K7. Or maybe they decided it won't be embarasing if it is. Or maybe the Tbird will make a "suppise release".
I was using their description. I agree that it's a rather dubious one. However their processor is hard to categorise - I don't think I've ever encountered anything quite like it before.
I admit it is their description. I was going to let it slide on that basis alone. But then I figured what the heck. We raise a stink in almost every thread where a media article mis-uses the term "hacking". Why should we let hardware marketers mis-use the term RISC just as badly?
Also for a somewhat similar CPU look at the Motorola 68F3XX series.
Of course the addressing modes are very valuable for the kind of work this chip is intended for.
The addressing modes don't really make life simpler. They just let you do "LOAD ((R1)+R2), R3" rather then "LOAD (R1), R4; ADD R2, R4, R4; LOAD (R4), R3" (assuming target register is the last one). It doesn't save any cycles, even on CPUs where the double indirect plus displacement is faster then the two loads and add, the whole CPU is slower at everything for having the indirect mode!
Go back and read about the horror of the "fast" VAX in the Mashy essay I posted the link to.
Maybe this is why it is hard to categorise - the design (including the instrcution set) is optimsed towards doing I/O and networking manipulation.
The parts of the instruction set that make it CISCy aren't really needed for I/O. I don't object to the in-register byte/bit flopping. I don't object to the integrated Ethernet or serials. I definitely don't object to the integrated net booting.
To be honest I don't even object to the parts that make it CISCy, only that they chose to mis-lead people by saying "RISC" rather then "CISC and damn proud!". It's not the mistake that kills you, it;s the cover-up:-)
100 mips
Thanks. I couldn't remember.
Perhaps not - but it isn't really a classical CISC either. Perhaps someone needs to invent a new acronym ?
Well since it is "standard CISC with a bunch of peripherals stuck inside it to appeal to embedded network apps", I would go with "embedded CISC", or "embedded network CISC".
The Data sheet seems to push the "RISC" angle as an explanation for the low power consumption - and it certainly achieves pretty good results there.
The Motorola ColdFire is a CISC that goes up to around 200Mhz at very low power. The Motorola DragonBall is a CISC that is definitely the low-power brains behind current Palm Pilots.
CISC doesn't mean power hungry. Hell thanks to Intel and AMD CISC doesn't even mean slow-as-a-dog.
The "bottom line" for me is that, on paper, it allows me to build a solution with a hardware cost around half that I can achieve any other way.
Well, that's because it is a kick ass little CISC CPU.
I think perhaps my application is unusual - I don't want any displays or user interface (the customer explicitly wants a "grey box screwed to the wall that they can just forget about") - so many single board computers on the market would just sit there with half the chips never doing anything - but I can't be the only person who needs to do this kind of thing ?
To have totally no UI is a bit unusual. But to have no raster pixel display (LCD or otherwise) isn't uncommon. But most applications are either in such huge volumes that it's better to do a design just for that application (so you can save $0.03 by leaving off an address pin you don't need, or other stuff like that), or such low volume, that they are better off buying a board with parts they don't need.
I think you can normally get the "unneeded" parts down to serial ports by looking at the "BASIC Stamp" market, and the small SBCs that replace high-end stamps.
Or look at Chuck Moore's stuff. Sure it has the NTSC video generator, but are (well, will be, someday) dirt cheap. Oh, and it's definitely a CPU design that I have a hard time pegging as RISC or CISC. So you could win on that one:-) Besides it's so cool to fit the computer inside the mouse.
They've designed their own custom RISC processor with built-in network support and quite a lot of very neat I/O ( A dream come true for telecomms embedded people).
It's not very RISCy. It looks more like a almost-68000 with delay slot branches (if I remember right). It has a bunch of indirect addressing modes not found on most (or any) RISCs. Of corse it doens't have to be a RISC. It runs pretty quick (50Mhz? 100Mhz?), as you say it has lots of good I/O features and built-in networking.
It is a pretty cool CPU for the right target. But it ain't a RISC.
People who want to know more about a cool embeded CPU with lots of networking and I/O features should definitly checkout their little CISC CPU.
I'm not sure. I lay about six feet from my 32" TV, but that's because that's about how tall I am. Sometimes I sit about nine feet from it, but that's where the couch is. I really don't have a good place for a 54" projection set. It's much deeper then my current TV, so I would have to put it closer, and I don't think it would stand up to dog slobber.
For me the 28" HDTV VVega ultra-flat glass tube sounds better. I would lay about six feet from it, or sit about 8 feet from it.
I bought a nice GemStar 19"
monitor for $300. Not quite a trinitron, but it'll give me 1600x1200 and 75hz. I sit less than two feet away
from this nice big screen. And let me tell you. It takes up a lot more of my field of vision than a 54" tv
would from my couch.
I have a very nice desk chair. To gloat a bit, it's an Areon. It's not as comfey as my couch. After a day of sitting (at the office, or at home) it's not really as comfey as laying on the floor either.
My wife would also be a little upset if only one person could watch TV at once, especally if I kept kikcing her off it to read slashdot.
And considering I can have a DVD player, DVD burner, internet access, chat, and
video game access. I think that it more than surpasses a 54" HDTV for what it does.
My car provides warmth, has a six speaker stereo, can haul 4 people and a little baggage at 130MPH with the top down. It can muss my hair. I can cook a pizza on it's engine block. It is definitly surpasses an HDTV system at what it does.
Regretabaly the HDTV surpasses both my car, and your PC at being a easy to use appliance that can be viewed by several people in the comfort of a tipical living room.
Now if you wanted a car, or a computer, then your better off getting one of them then a HDTV. But if you want a TV, a PC is only a limited substitute. Under the right set of circumstances it is a quite acceptable substitute.
Unless you're at a
party.
I would question the wisdom of asking a girlfriend to sit at your desk and watch a movie with you. Then again you'll both have to sit close together, so it may work out.
You're not going to get 20 people to watch your 19" monitor for the super bowl. (not that nerds
watch such things!!!)
Geeks definitly watch the superbowl. But pretty much only for the comercials.
In Case you haven't noticed, hdtv sets are rather expensive.
Yep. The Sony 38" (or is it 36") ultra-flat glass VVega is quite pricey at $6000. The 50" projections sets are somewhat more so.
The average computer nowadays, however, is
less half the price and has a monitor more than capable of displaying any format of hdtv you throw at it..
basically you could have a 19" hdtv/dvd player for around 2,000...
There are a lot of cheep monitors that won't do the more agressave HDTV reslutions, but yes any monitor over about $150 will do nicely. Of corse that's a bit unfair because I expect if Sony, et al thought the HDTV market would buy 19" HDTV sets they could make them cost somewhat less then computers.
Find me a 32" PC monitor for cheep. Howabout a 38"? Or a 54"? Remember to discount the size to get it to the same aspect ratio as HDTV (or use that amazingly costly Apple Cinima display).
It would be wondeful if I could make a $600 PC (or even $2000) PC system do the work of a $6000 HDTV, but it's just not possable yet.
Of corse if all you have for HDTV is $600, and 19" sounds good to you, go for it. I admit some HDTV probbably beats no HDTV. Then again Quake III might beat any HDTV just now, given the lack of much programming!
(I'm all for bargins, and trade-offs, just know what you are trading off!)
Re:Bad wording, and benchmark link
on
Athlons Sold Out
·
· Score: 2
That was bad wording on my part, I apolgize. I should have said, maybe: Marketwise , the K7 is obsolete. As in, it's time for them to market the poo out of the Spitfires and Thunderbirds.
Not if anyone at AMD remembers the failure of Osborne Computers. Don't market your new product when you rely on your old one for income. AMD does not make more money off the Thunderbird or Spitfire then the current K7s, so they have no intrest in selling you a Thunderbird in three months by not selling a K7 now.
By waiting to sell a product they lose money in at least two diffrent ways. They don't get the time value of the money (intrest, or the ability to spend it on infastructure, etc.). They also run the risk that the consumer will lose intrest in the product and buy something else (an Intel CPU, or a nice coffee table).
The fact that wholesale they are sold out of K7s doesn't change this too much. If they market the crap out of non-existant products they could damage the OEMs ability to sell the K7, which could cause cancled orders (making them no longer sold out, and AMDs problem), and will cause hard feelings between AMD and the OEMs. AMD doesn't need any bad blood there.
As many quality hardware sites speculate, the Spitfire will still outperform the current Athlon in many applications, mostly games. Unfortantly, a good site I frequent (Ace's hardware) had an article/link/write up on why the Spitfire and T-bird CPUs are better CPUs then current Athlons (As in, what they changed, how the on-die cache will help/hurt, etc) but they don't keep a backlog of articles that I can see.
I havn't seen Ace's Spitfire review. I have seen some reviews that look faked. So be careful. I'll beleve the next round of benchmarks when I see them on AMD's site (or SPECs). Before that I'll be intrested to see them, but I won't base any important choices off of them. Ace's is moderatly technical, not intensley so like, say comp.arch (anyone remember Usenet?), but it ain't bad.
I also note that the Spitfire could out do the existing K7. In fact I totally expect it will on things that don't need more then 256K cache. But I'm not sure what those applications will be. It won't (er, may not) be the same as are happy on Intel's 256K cache because of the many noted diffrences between the two cache systems, and CPUs.
Intel isn't going to be bought instead, because Intel can't even supply their chips.
Exactly, the article pretty much says AMD is out of K6's because Intel underproduced the Cellerons.
Why would you want an Athlon anymore, anyway? Spitfire (Low-end) and Thunderbirds (High-end) are on their way in a month and some, which are much better then the current Athlon.
Anyone who can't/won't wait another month and change for their new machine. I bought a new machine three months agoish because my old one picked the wrong time to kick the bucket. A friend just got a new (non-K7) machine because they were tired of his three year old machine.
Athlons now are what the origional K6 was. Obsolete.
No. The K6's successer is allready on hte market. The K7's isn't. What's selling now is allways worse then what's selling in two months (in CPUs at least). Just because there is an extra large discontunity coming up doesn't mean you can point to the future product and say the current one is history.
The off-chip cache K7 will be obsolete when the on-chip cache ones come to maket. Not before. Otherwise everything is obsolte. Why buy a K7 when a K8 is surely going to come out? Why buy a obsolete P-III when the 1.5Ghz Willmette is "only a year away"? Why would anyone buy a 21264 when the 21364 will be avilable in december? Why would anyone buy a 2000 VW bug when the 2001 VW will surely be out any month now?
Granted this is a bad time to buy if you can wait 2 to 3 months, but if you can't, well, progress allways marches on. In two years your obsolete K7 will not look noticably more quaint then a Spitfire/Thunderbird. On a side note, those Spitfire's are suppost to be pretty cheap, and perform better then the origional K7?
Well they have less but faster cache, much like the Intel CuMine vs. the um, Kamtai. Let's see, the 550B vs the 550E is that what they are called now?
The Intel part has a half sized twice as large cache with a 4 times wider cache bus (256bit vs. 64bit) , and 8-way associatave vs. 4-way associtave. There are still things that are faster on the "big cache" versions, but they are not common I think.
The only thing we know for sure about the Spitfire is it has half the cache of the current K7. It's a good guess that it has a wider (cache) bus, but that's not a given. There have been no hints that the Spitfire's cache has changed to a more associatave structure, and since that isn't a no-brainer to do, it may well not have. Oh, and we know the Spitfire's cache should run full speed rather then one half to 2/5ths of CPU speed. The larger L1 cache sizes on the K7 also make a direct comparisin with the P-III non-trivial.
So we don't know for sure that the Spitfire will outperform the existing K7. We have an existance proof in the P-III E vs B that it could. But the changes arn't identical. The results could differ by quite a bit.
That said, I think the Spitfire will be a really good CPU. If it gets priced similar to the K6's or the Celerons, then it's going to be a great bargin.
Also since the Spitfire is intended to replace the current K7, I expect it will do at least as well. That may mean it has to have a large on-die cache. Or maybe it will have a higher associtavity, or some novel approch, but it's design goal is to replace the current K7, so it will be at least as good. The design goal of the Spitfire is to replace Intel's Celeron, it may "accidentally" be better thn the current K7, but that wasn't a direct design goal.
P.S. my appologies if I swapped the Tbird and Spitfire's roles. I cna never quite remember which code name is what in this bisness.
Sure I do. I just happen to think that Unix does, in fact, suck - and I did so long before I found the Handbook. I've spoken to at least one of the authors in length, and he also thinks that Unix sucks. So, it seems to be a consensus... after all, why do you think Ritchie and the gang have left Unix for Plan 9?;)
Plan 9 strikes me as "more Unix then Unix". Everything is a file. Everything. No more ioctl, or fcntl, just have a few extra "control devices". (I know, you put the smiley there, but I had to respond....)
My big problem with the Unix Haters Handbook, is not that they hate Unix. I rather hate parts of it too. It's just that the whole seems to be better then all the other stuff that is out there, at least that I can get!
My beef with it, is that while it did discribe Unixs failings (and sometimes things I didn't think of as failings) at great length, it seldom described a system that did it better. It did for error messages (I think it liked the VMS error system), and it did for a hand full of other things, but for less then half.
To me that just makes it a bitch session. Which is fine if that's what you want. I would rather have something point me at a "right way", something for Unix to grow, or at least something to pine for that can't be retrofitted. I woulnd't buy a book telling me my car sucked because it only has a 190HP engine. I might buy one if it told me how other componies managed a 240HP engine in the same space. At least if I were mechanically inclined;-)
(and yes, I have the tunes project page up in another window, that looks more intresting then the UHB)
. But the one I really wanted to get my hands on was the ?NCR?AMD??? 32032,[...]
National Semiconductors. Diffrent form NCR. I think they are still in bisness. They bought Symbios a year a two ago if I remember right. They make nice SCSI controlers (that happen to have little 16/32bit CPUs), not sure what else. They might have bought Cyrex, and then sold it.
Unfortunately, the maker of the chip never managed to ship them in volume or with adequate performance.
Volume might have been a problem. I don't recall. Performance would have been Ok, they had decent clock rates for the time, and decent cycle counts. I think what killed them was tons of eratta for each stepping of the CPU. That would be "hardware bugs". Many time appearing and dissapaering over diffrent steppings. It made some instructions useless, some addressing modes useless, and many combos of the two useless. (i.e. it might have had "load the value at R1, add it to the value at R2 and store if greater then zero", but you coulnd't use it on 30% of the CPUs, so it was worthless).
I think it (or a decendent) actually got used in the PC532, which was a homebrew computer (probbably the last ever sold as a scematic and bag of parts, sodder together, not plugging cables and bords in!). They can run NetBSD now (and for the last N years). There may be as many as 200 of them in the world.
Just from a theretical point of view, how difficult do you think it would be to take those servers down from terrorist activity. I mean could the internet be taken down if 12 explosions at the right time/place where detonated?
Assuming you can figure out where they all are form the IP addresses in the root.cache file, and traceroute, or other similar tools, and maybe a bit of social engenering, it shouldn't be any harder then any other 12 randomly selected machines. (i.e. you may get unlucky and some are in phone COs and you need to get into a somewhat secure area, or blow through a lot of concrete in the internal walls behing the office bilding facade).
That wouldn't take out "the Internet", just much of name service. It would suck a lot. As caches started timing out things would start to suck a lot more.
However there are unoffical secondaries (not listed), and I assume other backup sets of the data. "All" that would be required would be to set up another root server (or 12), and route the old root serve's machine's IP address to the new ones. Wait less then five minutes for routing to converge, and all is right with name service again. Regretabably the loss of life involved in "12 explosions" would be far harder to "correct".
Beats me how long it would take to fix. If there is a real drill for it, maybe under an hour. If there is no drill for it, it could be much longer since the "12 explosions" probbably will cause lots of confusion.
I'm not sure why you used the analogy you did. In the event of a natural disaster, a piece of Big Iron is just as fallible as a PC. Depends on the disaster. The S80 could probabbly fare pretty well in an earthquake. I remember an add DEC use to run about their "High Availability" VAX/VMS systems. A picture of a machine room after an earthquake. Machines that had ripped the bolts out of the racks and were on their sides. The HA VAX had it's disk lights going, even with part of the machine in a pool of water (I assume from some thing else's cooling). The S80 could have a lot of it's CPU and memory boards unseated (or destroyed) and should keep on chugging (it might have to auto-reboot). No PC's I know of would. Unless you count the old Sequents as PCs just because they use 80386s and 80486s. This makes me curious -- what would happen if the root A server got totalled? What gets failed over onto? If the primary fails the secondarys can still give answers (I think secondaries can even give authoritatave answers in most cases). The failure would have to last days before a Bad Thing (other then excess load) happened. Check your/etc/namedb/root.cache for details.
Except that I can't find out how DES was developed. Oh sure I have the source, but it contains tables and tables and tables of magic numbers! The choice of the values for the S boxes and why and how those choices were made is **still** classified!
When the diffrential-cryptoanalsis attack came out in the mid-90s DES was one of the few cyphers that were extreamly resistant to it. DES with the old pre-NSA-change S-Boxes was very weak against the "new" attack.
There were many people who beleved that
NSA "must have" known about the "new" attack for 20+ years
The S-Box change was to make sure DES was resistant against that attack
There is no reason to beleve that the S-Box change made DES weaker. The again there is no reason to beleve that it didn't. Or wasn't breakable by some other means the NSA knew about, and they wanted to fix the S-Boxes because they thought some other goverment knew how to attack them.
It's a hard field to make a right choice in, only in part because paranoia is actually a good mindset to be in when doing crypto think. I beleve 3DES is relitavly safe. But I don't know. I'll feel much better after we have 5+ years of experiance with whatever cypher wins the AES selection process.
Netbooting is still important. If I want a quiet machine hooked to my stereo to play MP3s netbooting would let me avoid a hard drive (assuming I fetch the MP3s over the net, which is more then fast enough). Using a StrongARM, or other low power CPU would let me avoid the need for a CPU fan as well, so no moving noisy parts at all.
Netbooting is also useful if you are bringing up a new system. To be useful the new system will need networking (even if only to get new boot images). A netbooted system won't need the disk drivers to work right away. Or the console drivers. Or anything but the ethernet (or a serial driver and PPP...).
Netbooting is useful for field upgrades as well. Simpler then getting a floppy or CD in the machine if it isn't anywhere near you (or other people!). I have lots of computers I do work on that are nowhere near me (some 30 miles away, some 3000). Many of them are in unmaned locations.
Netbooting is useful as a last resort on a machine that has blown it's drive and is nowhere near people that can fix it, at least if the image can be re-written, or if the system can really be used diskless (like it has blown a boot but not data disk, or is doing CPU bound tasks like playing part of a render farm). To be honest I havn't had this happen in the last 10 years. But it could happen.
Lastly, look at the i-opener. Sure it has 16M of FLASH, but you don't want to write that very offen. Unless you cram a hard drive into it you'll alsmot certonally use the 16M of FLASH as a glorifyed bootloader, NFS mounting the bulk of your system. Possabbably even/. Sure you could use a USB Zip (or Orb) drive, but that'll be as slow as NFS over USB ethernet anyway (well caching works somewhat better, unless you use NQNFS, or CODA, or RVD, but still...).
So while netbooting isn't as useful as it was thought to be 10 years ago (and to be honest it was overhyped then!), it is really really valuable in some situations.
The bottom line is that this guy bought an Internet appliance with the intention of using it in ways the manufacturer didn't intend. It's not their fault they assumed he was a normal customer. If you're going to purchase things and use them in ways the manufacturer didn't intend, you have to expect that there are going to be things attached along with that purchase that you don't necessarily need.
Let's say I was buying a car. For some reason the dealer for this brand of car sells them really cheap, say $4000 rather then $15,000. But they charge $200 per scheduled service rather then $19.99 like Jiffy Lube. You ask the sales person if you have to take it to them, or if you can take it elsewhere.
If they say "you have to take it to us, it's in the contract", then your argument is right. Suck it up. You agreed to it, you deal with it.
If they say "oh, well you can take it wherver you want, but they way it is built we are really the only ones that can service it, so it'll be a bad idea...but there is no contract", then your argument is crap. They told you you'll end up coming back to them, but they said there is no agreement in the contract. You can do what you like. (and this is what Netpliance told some people, including me)
In the middle ground if they said "you have to get the oild changed here, but if you choose not to drive it, then you don't need the oil changed (and there is no contract)", well I figure your still wrong. After all they said it ain't in the contract. (this sounds more like what netpliance told the guy in this article)
Now in the real world, it doesn't matter as much what they said, if there is a contract that says the other thing.
Netpliance is fully within their rights to make a contract that says "you have to pay us $21 a month for service for X months or pay us Y dollars to terminate the deal". They are totally within their rights to do this to all new customers. They can't do it to people who have allready bought. I expect this includes people who have payed, but not yet recieved the product.
Or do you really think Ford should be able to force you to get oil changes at a Ford dealer on a car you bought months before they decided this was the "new policy"?
Great deal, or not, nobody can retroactavly change contracts. Except a court of law, and normally they don't unless the terms of the contract are really really absurdly unfair. By absurdly unfair, I mean things like "after you work for us, you can't work for anyone else" (and even those are upheld from time to time -- if they are really "for a huge amount of money you can't do a really similar thing for some amount of time", like millions and years, or thousands and a few months).
But they could make similar systems intended for home networkers that cost more, rather than ignore that market. Add a small harddrive and either an ethernet card or an external USB port for an ethernet adaptor. That's what many people are doing with them, and from the i-opener plus parts cost, probably spending $200-250.
If they sold one with that functionality -- a low-powered, cute, network-friendly thing -- for anything under $350, they would probably fly out the door.
They have made a modest amount of noise about "working with the open source people" and about "seeing a new market identifyed". They may well be planning a somewhat beefyer machine with a much larger price tag. I'm not sure they would "fly off the shelves" if they were the same machine plus a small (I think 2G is the smallest in production) portable drive, and an ethernet (it would cost less to have an ethernet then the existing modem, but only by a few bucks), and no QNX or even SanDisk flash part.
But I'm not sure you could make such a device, and profit from it at $250.
You have, maybe $20 of CPU, $50 of memory, the cheapest desktop SS7 motherboard is about $30, a small in-production hard disk is about $100. I'm not sure about the flat-panel, a better quality one at the exact same size is $99, so $50 might be reasonable. That's about $250 right there, and we havn't payed anyone to put it together! Or for the keybord, the plastic "case", the little metal stand, or any money for the design and other NREs. Oh, and profit, both for netpliance, and for the distrbution channel.
Even if Netpliance does, it takes months to bring a machine to market. Even if it is just a fairly minor respin of an existing product. Even if you skimp on "focus group" "market testing".
Remember this thing is essensally a cheap portable. None of those are on the market at signifagantly less then $1,000. Let alone $500. Let alone $300. Even with this portable being made of more trailing edge parts then any i have seen in a while, it still costs a fair amount of money!
P.S. the existing one has a USB port, I assume for "future expansion". So you could put a USB ethernet in it (assuming you were running a patched Linux, or a FreeBSD 4.0 or newer, or the right revs of NetBSD, or...)
P.P.S. I think Netpliance it justifyed in changing their "no contract policy", but not retroactivaly. In fact I'm almost sure doing it retroactavly breaks some laws. I think they are also justifyed (legally and moraly) in making the new units harder to hack, but I think it is a mistake. Either the percentage of units ordered to be hacked is too low to bother with, or they are about to get a wave of returns, possabably broken returns.
One really cool thing about Cisco is that their IOS isn't all copyrighted - Juniper and Redback both use large parts of Cisco's software on their equipment, and the interfaces work the same way.
IOS is quite copyrighted, and if Cisco thought Juniper or Redback used any IOS code they would sue in half a heartbeat.
IOS's "look and feel" isn't copyrighted. Which is good, since it is pretty much the TENEX LnF anyway. Juniper made their product have a similar LnF because the coustmer base allready has Cisco experiance, and maybe because many of the original Juniper folks were ex-Cisco folks. Redback did the same LnF because they figured the customers would expect it. Redback also added "virtual routers", which they are very proud of.
Thhe other thing about networking equipment, it has to obey the protocol. Cisco hasn't gone out and written it's own version of IP, frame relay,[...]
Cisco has gone and done their own Frame-Relay like Framing ("Cisco HDLC"), their own ethernet VLAN stuff. But I don't beleve it was done for "embrace and extend" reasons. Their VLAN stuff was out before 801.Q was a standard. Their HDLC is much lighter weight then Frame Relay.
But it's not all rosey. They have patented their "Hot Swap IP address" thing (I forget if it is two machines that share a ether MAC, or if they both have a MAC and the third MAC is passed between them). That's right patented.
Cisco hasn't shot itself in the foot yet, and I doubt they intend to. In contrast to Microsoft, Cisco has become the networking giant because their products actually work. Redback and Juniper have come into the game in niche positions, and Cisco has left them there, rather than trying to kill them off.
Cisco has left Juniper in the "super fast super dense router niche" because they can't dislodge them. I don't know if that is because they see selling an ISP-only router (few non-nationwide ISPs need an M40 let alone an M160), or if the BFR really is the best they could do.
Similar with Redback, they just didn't have a product that worked as well in that space last year, and I don't know how hard they are trying.
Now both of these things may be sound bisness moves. Being the biggest-baddest-router is a lot like being a SuperComputer, and you will note few SuperComputer componies make the really big bucks. There is much more money in selling PCs, or even "wussy little Unix things like the Sun E10K".
Slashdot Mirror
They arn't so much faster and more reliable then the latest cluster systems so much as actually being the latest cluster systems. Oh, and being far more reliable (lots of check logic, ECC on the cache, and register files, not just main memory). Not so much the faster part though. The CPUs only run a few hundrad Mhz (last year 300Mhz to 400Mhz was extreamly fast for them). Of corse they have dedicated IO processors, and small tens of CPUs is a common size.
Most of the innovations in clustering in the micro world is re-inventing what mainframes have allready been doing. Then again so was having caches, and out-of-order execution. That's not to say micros didn't invent something, or that finding the right time to re-use what had gone before isn't hard in and of itself.
It definitly had Flash. It had a little virtual you that would put on the clothes and let you pan around. It had a little virtual Ms. Boo that would comment (allways positave empty comments). And it was dog slow. And not from anything on our end.
Oh, and the virtual you? You can't tell what the damm clothes look like on it anyway. Your better off looking at the models (so you can see how the fabric really drapes).
Even if boo.com failed for other reasons, I like to think of the total lack of usability testing as a contributing factor. Or total lack to listen to the testing.
If the "bubble being over" means really bad sites like this dry up, I'm all for it.
True for tracking stocks (like AWE), not so true for other stocks. It is true for tracking stocks because there is no way for the parent compony to actually sell the asset, not because they wouldn't want too. They may even spin off the unit if the tracking stock goes high enough. Maybe.
For a non-tracking stock, you better beleve that a corp will sell off it's shares in another once the price excedes their percieved value, and the owning corp has reason to beleve the price has peaked (i.e. it's on it's way down now), or (the seller) needs a (profit) boost.
They have their own stock holders to answer too, and that is one of the (few) things a corp must must must do.
Remember though, perceved value isn't just money. Apple may not want to sell shares of Earthlink (assuming it has any) because as long as it owns a big part of them it can exert extra presure to make sure Earthlink keeps supporting Apple products. That is the same non-monitory value you are intrested in, so it's not supprising other have allready discovered it.
The L2 duron cache is exclusave, that is does not contain cache lines from the L1 cache. Many other processors don't work this way, in part because it is simpler to make a multi-processer system if only the L2 caches need to do bus-snooping. My guess is the Duron won't be too happy in a multi-processor system for this reason.
It'll still make a decent single CPU system, especally at the dirt cheap prices listed in the last article.
Not very many of them. But some may. And reporters and stock analists read slashdot (I don't know if it is just a few, or lots).
But not many need too. If CNNFN's reporter sees a nice jucy quote, it'll go somewhere SCO's shareholders do read. If it is east-bumfuck's reporter, and they print it, you can bet other reporters will reprint it later.
Intresting news travels. Far beyond slashdot.
Well, if the offender was going to BSDL, or MPL their code, they would probbably realise that it would be examined by people who knew the borrowed code was GPLed, and not have made the "mistake".
In fact they would almost certonally understand the "rules" and send mail to the orignal author saying "I'm doing a BSDL'ed project, can I re-release some of your code from foo.c under the BSDL?".
Of corse the original author is free to say "no, I don't want anyone to take it from your code and slap it into Windows NT". So, they could be mostly outta-luck.
They could BSDL their code except for the foo.c code which remains under GPL. Or they could re-write it. Lots of OSS is written for the fun of writing, so maybe they woulnd't even want to borrow from foo.c, they might want to write a better foo.c, and mail the GPL'ed code author "Too bad you wouldn't let us re-licence that code, but here is another better version we wrot from scratch, feel free to ship it with your code".
If they went and violated the GPL anyway, I expect they would have gotten the same polite treatment nVidia got. Probbably.
What makes you think money is the only kind of damage? If I claim an Arthor Clark short story is mine when I read it to friends, it damages him. He does not get the gain in reputation his work has earned him. If I were to call someone else a wife beater, there would be no money involved, but there would be damages (their reputation, and possably my face just after they punch me).
If you break into my home and steal my VCR, and yet leave me $200 (the purchase price) there is no monatary damage, yet it leaves me with a feeling of insecurity. You could break in and maybe not leave money. Or take a life rather then properity next time.
Damages are not just money.
A lawyer would be able to convince a jury. Easy.
Now whether a lawyer would be able to convince a jury that this wasn't an honest mistake, or that 200 lines of stolen code in a 10,000 project is a super-huge deal, well, I don't know about that.
I would rather have the GPL tested against a violation that is clearly not a mistake, and is clearly a large chunk of code. And best of all, on a clearly wealthy and unrepentent violator, since US juries seem to wack them the hardest.
Public Domain grants users some rights that BSDL doesn't. For example BSDL specifically disclaims any warranty, and makes a big deal out of it. "Public Domain" does not. So if you PD some software, and it doesn't work you stand a large chance of being sued, and a larger chance of being found liable. In my opinion a very small chance either way, but why take the risk?
Now in my opinion the original author who put his code under the BSDL because he figured he couldn't prevent violations is giving up. Personally I would have MPL'ed it and put an attachment on saying "there is a flat $100 fee to gain commercial license to any or all of this code". Most companies would rather pay $100 for something then steal.
That's not to say that people that put things under the BSDL under the belief that that license does the most good are wrong. If you BSDL something because you think it is good solid code, and you would rather see it used as the basis of a commercial work then a buggy chunk of crap, and have no desire to be payed (directly at least) for the code, then the BSDL is fine.
If you think the path to better software (or whatever it is you want) is better served by the GPL, or the MPL, well go for it. I'm kind of fond of the MPL myself, especially the bit about patents (it would be nice if a future GPL would adopt that section).
No, that is what the "Lesser GPL" (formerly Library GPL) does. The (greater?) GPL both keeps improvements free, and prevents most use in non-open software (in fact, not just non-open, but non-GPL).
You can not use GPL'ed code in a open source project that uses a non-GPL license. It's the same for MPL'ed code too. Or to be a bit more accurate, you can't use GPL/MPL'ed code as anything other then a distinct program. I could have a non-GPL'ed juke box that uses a GPL'ed mp3 player if I pipe my songs to it. If I want to link it in (so I can fast forward and rewind) I'm out of luck with a GPL'ed MP3 player.
If that is the intent of the author of the GPL code, then all is well with the world. After all it is their work, and they have some rights over how it should be used. If it was the intent of the author that GPLing the code allows almost unrestricted sharing, then they have made a grave mistake. They should have used a different license, like the LGPL. Or put in an escape clause for integration by other Open Source software.
Well, it's less then a week later, and we are hearing about them. So I guess they could be out next month. Or a bit after.
So soone I guess we will know how the perform. I didn't see any AMD talk about the Thunderbird, so maybe the Spitfire isn't faster then the existing K7. Or maybe they decided it won't be embarasing if it is. Or maybe the Tbird will make a "suppise release".
I admit it is their description. I was going to let it slide on that basis alone. But then I figured what the heck. We raise a stink in almost every thread where a media article mis-uses the term "hacking". Why should we let hardware marketers mis-use the term RISC just as badly?
Also for a somewhat similar CPU look at the Motorola 68F3XX series.
The addressing modes don't really make life simpler. They just let you do "LOAD ((R1)+R2), R3" rather then "LOAD (R1), R4; ADD R2, R4, R4; LOAD (R4), R3" (assuming target register is the last one). It doesn't save any cycles, even on CPUs where the double indirect plus displacement is faster then the two loads and add, the whole CPU is slower at everything for having the indirect mode!
Go back and read about the horror of the "fast" VAX in the Mashy essay I posted the link to.
The parts of the instruction set that make it CISCy aren't really needed for I/O. I don't object to the in-register byte/bit flopping. I don't object to the integrated Ethernet or serials. I definitely don't object to the integrated net booting.
To be honest I don't even object to the parts that make it CISCy, only that they chose to mis-lead people by saying "RISC" rather then "CISC and damn proud!". It's not the mistake that kills you, it;s the cover-up :-)
Thanks. I couldn't remember.
Well since it is "standard CISC with a bunch of peripherals stuck inside it to appeal to embedded network apps", I would go with "embedded CISC", or "embedded network CISC".
The Motorola ColdFire is a CISC that goes up to around 200Mhz at very low power. The Motorola DragonBall is a CISC that is definitely the low-power brains behind current Palm Pilots.
CISC doesn't mean power hungry. Hell thanks to Intel and AMD CISC doesn't even mean slow-as-a-dog.
Well, that's because it is a kick ass little CISC CPU.
To have totally no UI is a bit unusual. But to have no raster pixel display (LCD or otherwise) isn't uncommon. But most applications are either in such huge volumes that it's better to do a design just for that application (so you can save $0.03 by leaving off an address pin you don't need, or other stuff like that), or such low volume, that they are better off buying a board with parts they don't need.
I think you can normally get the "unneeded" parts down to serial ports by looking at the "BASIC Stamp" market, and the small SBCs that replace high-end stamps.
Or look at Chuck Moore's stuff. Sure it has the NTSC video generator, but are (well, will be, someday) dirt cheap. Oh, and it's definitely a CPU design that I have a hard time pegging as RISC or CISC. So you could win on that one :-) Besides it's so cool to fit the computer inside the mouse.
It's not very RISCy. It looks more like a almost-68000 with delay slot branches (if I remember right). It has a bunch of indirect addressing modes not found on most (or any) RISCs. Of corse it doens't have to be a RISC. It runs pretty quick (50Mhz? 100Mhz?), as you say it has lots of good I/O features and built-in networking.
It is a pretty cool CPU for the right target. But it ain't a RISC.
People who want to know more about a cool embeded CPU with lots of networking and I/O features should definitly checkout their little CISC CPU.
Anyone that thinks it is a RISC chould check out page 17 of the ETRAX 100 Data Sheet, and then maybe the John Mashey "What Is RISC" essay. Enjoy.
I'm not sure. I lay about six feet from my 32" TV, but that's because that's about how tall I am. Sometimes I sit about nine feet from it, but that's where the couch is. I really don't have a good place for a 54" projection set. It's much deeper then my current TV, so I would have to put it closer, and I don't think it would stand up to dog slobber.
For me the 28" HDTV VVega ultra-flat glass tube sounds better. I would lay about six feet from it, or sit about 8 feet from it.
I have a very nice desk chair. To gloat a bit, it's an Areon. It's not as comfey as my couch. After a day of sitting (at the office, or at home) it's not really as comfey as laying on the floor either.
My wife would also be a little upset if only one person could watch TV at once, especally if I kept kikcing her off it to read slashdot.
My car provides warmth, has a six speaker stereo, can haul 4 people and a little baggage at 130MPH with the top down. It can muss my hair. I can cook a pizza on it's engine block. It is definitly surpasses an HDTV system at what it does.
Regretabaly the HDTV surpasses both my car, and your PC at being a easy to use appliance that can be viewed by several people in the comfort of a tipical living room.
Now if you wanted a car, or a computer, then your better off getting one of them then a HDTV. But if you want a TV, a PC is only a limited substitute. Under the right set of circumstances it is a quite acceptable substitute.
I would question the wisdom of asking a girlfriend to sit at your desk and watch a movie with you. Then again you'll both have to sit close together, so it may work out.
Geeks definitly watch the superbowl. But pretty much only for the comercials.
Yep. The Sony 38" (or is it 36") ultra-flat glass VVega is quite pricey at $6000. The 50" projections sets are somewhat more so.
There are a lot of cheep monitors that won't do the more agressave HDTV reslutions, but yes any monitor over about $150 will do nicely. Of corse that's a bit unfair because I expect if Sony, et al thought the HDTV market would buy 19" HDTV sets they could make them cost somewhat less then computers.
Find me a 32" PC monitor for cheep. Howabout a 38"? Or a 54"? Remember to discount the size to get it to the same aspect ratio as HDTV (or use that amazingly costly Apple Cinima display).
It would be wondeful if I could make a $600 PC (or even $2000) PC system do the work of a $6000 HDTV, but it's just not possable yet.
Of corse if all you have for HDTV is $600, and 19" sounds good to you, go for it. I admit some HDTV probbably beats no HDTV. Then again Quake III might beat any HDTV just now, given the lack of much programming!
(I'm all for bargins, and trade-offs, just know what you are trading off!)
Not if anyone at AMD remembers the failure of Osborne Computers. Don't market your new product when you rely on your old one for income. AMD does not make more money off the Thunderbird or Spitfire then the current K7s, so they have no intrest in selling you a Thunderbird in three months by not selling a K7 now.
By waiting to sell a product they lose money in at least two diffrent ways. They don't get the time value of the money (intrest, or the ability to spend it on infastructure, etc.). They also run the risk that the consumer will lose intrest in the product and buy something else (an Intel CPU, or a nice coffee table).
The fact that wholesale they are sold out of K7s doesn't change this too much. If they market the crap out of non-existant products they could damage the OEMs ability to sell the K7, which could cause cancled orders (making them no longer sold out, and AMDs problem), and will cause hard feelings between AMD and the OEMs. AMD doesn't need any bad blood there.
I havn't seen Ace's Spitfire review. I have seen some reviews that look faked. So be careful. I'll beleve the next round of benchmarks when I see them on AMD's site (or SPECs). Before that I'll be intrested to see them, but I won't base any important choices off of them. Ace's is moderatly technical, not intensley so like, say comp.arch (anyone remember Usenet?), but it ain't bad.
I also note that the Spitfire could out do the existing K7. In fact I totally expect it will on things that don't need more then 256K cache. But I'm not sure what those applications will be. It won't (er, may not) be the same as are happy on Intel's 256K cache because of the many noted diffrences between the two cache systems, and CPUs.
Crap. Sorry for the typos everyone. I hit "submit" rather then preview. I wish we could delete our own posts.
Is there a good reason we can't?
Exactly, the article pretty much says AMD is out of K6's because Intel underproduced the Cellerons.
Anyone who can't/won't wait another month and change for their new machine. I bought a new machine three months agoish because my old one picked the wrong time to kick the bucket. A friend just got a new (non-K7) machine because they were tired of his three year old machine.
No. The K6's successer is allready on hte market. The K7's isn't. What's selling now is allways worse then what's selling in two months (in CPUs at least). Just because there is an extra large discontunity coming up doesn't mean you can point to the future product and say the current one is history.
The off-chip cache K7 will be obsolete when the on-chip cache ones come to maket. Not before. Otherwise everything is obsolte. Why buy a K7 when a K8 is surely going to come out? Why buy a obsolete P-III when the 1.5Ghz Willmette is "only a year away"? Why would anyone buy a 21264 when the 21364 will be avilable in december? Why would anyone buy a 2000 VW bug when the 2001 VW will surely be out any month now?
Granted this is a bad time to buy if you can wait 2 to 3 months, but if you can't, well, progress allways marches on. In two years your obsolete K7 will not look noticably more quaint then a Spitfire/Thunderbird. On a side note, those Spitfire's are suppost to be pretty cheap, and perform better then the origional K7?
Well they have less but faster cache, much like the Intel CuMine vs. the um, Kamtai. Let's see, the 550B vs the 550E is that what they are called now?
The Intel part has a half sized twice as large cache with a 4 times wider cache bus (256bit vs. 64bit) , and 8-way associatave vs. 4-way associtave. There are still things that are faster on the "big cache" versions, but they are not common I think.
The only thing we know for sure about the Spitfire is it has half the cache of the current K7. It's a good guess that it has a wider (cache) bus, but that's not a given. There have been no hints that the Spitfire's cache has changed to a more associatave structure, and since that isn't a no-brainer to do, it may well not have. Oh, and we know the Spitfire's cache should run full speed rather then one half to 2/5ths of CPU speed. The larger L1 cache sizes on the K7 also make a direct comparisin with the P-III non-trivial.
So we don't know for sure that the Spitfire will outperform the existing K7. We have an existance proof in the P-III E vs B that it could. But the changes arn't identical. The results could differ by quite a bit.
That said, I think the Spitfire will be a really good CPU. If it gets priced similar to the K6's or the Celerons, then it's going to be a great bargin.
Also since the Spitfire is intended to replace the current K7, I expect it will do at least as well. That may mean it has to have a large on-die cache. Or maybe it will have a higher associtavity, or some novel approch, but it's design goal is to replace the current K7, so it will be at least as good. The design goal of the Spitfire is to replace Intel's Celeron, it may "accidentally" be better thn the current K7, but that wasn't a direct design goal.
P.S. my appologies if I swapped the Tbird and Spitfire's roles. I cna never quite remember which code name is what in this bisness.
Plan 9 strikes me as "more Unix then Unix". Everything is a file. Everything. No more ioctl, or fcntl, just have a few extra "control devices". (I know, you put the smiley there, but I had to respond....)
My big problem with the Unix Haters Handbook, is not that they hate Unix. I rather hate parts of it too. It's just that the whole seems to be better then all the other stuff that is out there, at least that I can get!
My beef with it, is that while it did discribe Unixs failings (and sometimes things I didn't think of as failings) at great length, it seldom described a system that did it better. It did for error messages (I think it liked the VMS error system), and it did for a hand full of other things, but for less then half.
To me that just makes it a bitch session. Which is fine if that's what you want. I would rather have something point me at a "right way", something for Unix to grow, or at least something to pine for that can't be retrofitted. I woulnd't buy a book telling me my car sucked because it only has a 190HP engine. I might buy one if it told me how other componies managed a 240HP engine in the same space. At least if I were mechanically inclined ;-)
(and yes, I have the tunes project page up in another window, that looks more intresting then the UHB)
National Semiconductors. Diffrent form NCR. I think they are still in bisness. They bought Symbios a year a two ago if I remember right. They make nice SCSI controlers (that happen to have little 16/32bit CPUs), not sure what else. They might have bought Cyrex, and then sold it.
Volume might have been a problem. I don't recall. Performance would have been Ok, they had decent clock rates for the time, and decent cycle counts. I think what killed them was tons of eratta for each stepping of the CPU. That would be "hardware bugs". Many time appearing and dissapaering over diffrent steppings. It made some instructions useless, some addressing modes useless, and many combos of the two useless. (i.e. it might have had "load the value at R1, add it to the value at R2 and store if greater then zero", but you coulnd't use it on 30% of the CPUs, so it was worthless).
I think it (or a decendent) actually got used in the PC532, which was a homebrew computer (probbably the last ever sold as a scematic and bag of parts, sodder together, not plugging cables and bords in!). They can run NetBSD now (and for the last N years). There may be as many as 200 of them in the world.
Assuming you can figure out where they all are form the IP addresses in the root.cache file, and traceroute, or other similar tools, and maybe a bit of social engenering, it shouldn't be any harder then any other 12 randomly selected machines. (i.e. you may get unlucky and some are in phone COs and you need to get into a somewhat secure area, or blow through a lot of concrete in the internal walls behing the office bilding facade).
That wouldn't take out "the Internet", just much of name service. It would suck a lot. As caches started timing out things would start to suck a lot more.
However there are unoffical secondaries (not listed), and I assume other backup sets of the data. "All" that would be required would be to set up another root server (or 12), and route the old root serve's machine's IP address to the new ones. Wait less then five minutes for routing to converge, and all is right with name service again. Regretabably the loss of life involved in "12 explosions" would be far harder to "correct".
Beats me how long it would take to fix. If there is a real drill for it, maybe under an hour. If there is no drill for it, it could be much longer since the "12 explosions" probbably will cause lots of confusion.
I'm not sure why you used the analogy you did. In the event of a natural disaster, a piece of Big Iron is just as fallible as a PC. Depends on the disaster. The S80 could probabbly fare pretty well in an earthquake. I remember an add DEC use to run about their "High Availability" VAX/VMS systems. A picture of a machine room after an earthquake. Machines that had ripped the bolts out of the racks and were on their sides. The HA VAX had it's disk lights going, even with part of the machine in a pool of water (I assume from some thing else's cooling). The S80 could have a lot of it's CPU and memory boards unseated (or destroyed) and should keep on chugging (it might have to auto-reboot). No PC's I know of would. Unless you count the old Sequents as PCs just because they use 80386s and 80486s. This makes me curious -- what would happen if the root A server got totalled? What gets failed over onto? If the primary fails the secondarys can still give answers (I think secondaries can even give authoritatave answers in most cases). The failure would have to last days before a Bad Thing (other then excess load) happened. Check your /etc/namedb/root.cache for details.
When the diffrential-cryptoanalsis attack came out in the mid-90s DES was one of the few cyphers that were extreamly resistant to it. DES with the old pre-NSA-change S-Boxes was very weak against the "new" attack.
There were many people who beleved that
There is no reason to beleve that the S-Box change made DES weaker. The again there is no reason to beleve that it didn't. Or wasn't breakable by some other means the NSA knew about, and they wanted to fix the S-Boxes because they thought some other goverment knew how to attack them.
It's a hard field to make a right choice in, only in part because paranoia is actually a good mindset to be in when doing crypto think. I beleve 3DES is relitavly safe. But I don't know. I'll feel much better after we have 5+ years of experiance with whatever cypher wins the AES selection process.
Netbooting is still important. If I want a quiet machine hooked to my stereo to play MP3s netbooting would let me avoid a hard drive (assuming I fetch the MP3s over the net, which is more then fast enough). Using a StrongARM, or other low power CPU would let me avoid the need for a CPU fan as well, so no moving noisy parts at all.
Netbooting is also useful if you are bringing up a new system. To be useful the new system will need networking (even if only to get new boot images). A netbooted system won't need the disk drivers to work right away. Or the console drivers. Or anything but the ethernet (or a serial driver and PPP...).
Netbooting is useful for field upgrades as well. Simpler then getting a floppy or CD in the machine if it isn't anywhere near you (or other people!). I have lots of computers I do work on that are nowhere near me (some 30 miles away, some 3000). Many of them are in unmaned locations.
Netbooting is useful as a last resort on a machine that has blown it's drive and is nowhere near people that can fix it, at least if the image can be re-written, or if the system can really be used diskless (like it has blown a boot but not data disk, or is doing CPU bound tasks like playing part of a render farm). To be honest I havn't had this happen in the last 10 years. But it could happen.
Lastly, look at the i-opener. Sure it has 16M of FLASH, but you don't want to write that very offen. Unless you cram a hard drive into it you'll alsmot certonally use the 16M of FLASH as a glorifyed bootloader, NFS mounting the bulk of your system. Possabbably even /. Sure you could use a USB Zip (or Orb) drive, but that'll be as slow as NFS over USB ethernet anyway (well caching works somewhat better, unless you use NQNFS, or CODA, or RVD, but still...).
So while netbooting isn't as useful as it was thought to be 10 years ago (and to be honest it was overhyped then!), it is really really valuable in some situations.
Let's say I was buying a car. For some reason the dealer for this brand of car sells them really cheap, say $4000 rather then $15,000. But they charge $200 per scheduled service rather then $19.99 like Jiffy Lube. You ask the sales person if you have to take it to them, or if you can take it elsewhere.
If they say "you have to take it to us, it's in the contract", then your argument is right. Suck it up. You agreed to it, you deal with it.
If they say "oh, well you can take it wherver you want, but they way it is built we are really the only ones that can service it, so it'll be a bad idea...but there is no contract", then your argument is crap. They told you you'll end up coming back to them, but they said there is no agreement in the contract. You can do what you like. (and this is what Netpliance told some people, including me)
In the middle ground if they said "you have to get the oild changed here, but if you choose not to drive it, then you don't need the oil changed (and there is no contract)", well I figure your still wrong. After all they said it ain't in the contract. (this sounds more like what netpliance told the guy in this article)
Now in the real world, it doesn't matter as much what they said, if there is a contract that says the other thing.
Netpliance is fully within their rights to make a contract that says "you have to pay us $21 a month for service for X months or pay us Y dollars to terminate the deal". They are totally within their rights to do this to all new customers. They can't do it to people who have allready bought. I expect this includes people who have payed, but not yet recieved the product.
Or do you really think Ford should be able to force you to get oil changes at a Ford dealer on a car you bought months before they decided this was the "new policy"?
Great deal, or not, nobody can retroactavly change contracts. Except a court of law, and normally they don't unless the terms of the contract are really really absurdly unfair. By absurdly unfair, I mean things like "after you work for us, you can't work for anyone else" (and even those are upheld from time to time -- if they are really "for a huge amount of money you can't do a really similar thing for some amount of time", like millions and years, or thousands and a few months).
They have made a modest amount of noise about "working with the open source people" and about "seeing a new market identifyed". They may well be planning a somewhat beefyer machine with a much larger price tag. I'm not sure they would "fly off the shelves" if they were the same machine plus a small (I think 2G is the smallest in production) portable drive, and an ethernet (it would cost less to have an ethernet then the existing modem, but only by a few bucks), and no QNX or even SanDisk flash part.
But I'm not sure you could make such a device, and profit from it at $250.
You have, maybe $20 of CPU, $50 of memory, the cheapest desktop SS7 motherboard is about $30, a small in-production hard disk is about $100. I'm not sure about the flat-panel, a better quality one at the exact same size is $99, so $50 might be reasonable. That's about $250 right there, and we havn't payed anyone to put it together! Or for the keybord, the plastic "case", the little metal stand, or any money for the design and other NREs. Oh, and profit, both for netpliance, and for the distrbution channel.
Even if Netpliance does, it takes months to bring a machine to market. Even if it is just a fairly minor respin of an existing product. Even if you skimp on "focus group" "market testing".
Remember this thing is essensally a cheap portable. None of those are on the market at signifagantly less then $1,000. Let alone $500. Let alone $300. Even with this portable being made of more trailing edge parts then any i have seen in a while, it still costs a fair amount of money!
P.S. the existing one has a USB port, I assume for "future expansion". So you could put a USB ethernet in it (assuming you were running a patched Linux, or a FreeBSD 4.0 or newer, or the right revs of NetBSD, or...)
P.P.S. I think Netpliance it justifyed in changing their "no contract policy", but not retroactivaly. In fact I'm almost sure doing it retroactavly breaks some laws. I think they are also justifyed (legally and moraly) in making the new units harder to hack, but I think it is a mistake. Either the percentage of units ordered to be hacked is too low to bother with, or they are about to get a wave of returns, possabably broken returns.
IOS is quite copyrighted, and if Cisco thought Juniper or Redback used any IOS code they would sue in half a heartbeat.
IOS's "look and feel" isn't copyrighted. Which is good, since it is pretty much the TENEX LnF anyway. Juniper made their product have a similar LnF because the coustmer base allready has Cisco experiance, and maybe because many of the original Juniper folks were ex-Cisco folks. Redback did the same LnF because they figured the customers would expect it. Redback also added "virtual routers", which they are very proud of.
Cisco has gone and done their own Frame-Relay like Framing ("Cisco HDLC"), their own ethernet VLAN stuff. But I don't beleve it was done for "embrace and extend" reasons. Their VLAN stuff was out before 801.Q was a standard. Their HDLC is much lighter weight then Frame Relay.
But it's not all rosey. They have patented their "Hot Swap IP address" thing (I forget if it is two machines that share a ether MAC, or if they both have a MAC and the third MAC is passed between them). That's right patented.
Cisco has left Juniper in the "super fast super dense router niche" because they can't dislodge them. I don't know if that is because they see selling an ISP-only router (few non-nationwide ISPs need an M40 let alone an M160), or if the BFR really is the best they could do.
Similar with Redback, they just didn't have a product that worked as well in that space last year, and I don't know how hard they are trying.
Now both of these things may be sound bisness moves. Being the biggest-baddest-router is a lot like being a SuperComputer, and you will note few SuperComputer componies make the really big bucks. There is much more money in selling PCs, or even "wussy little Unix things like the Sun E10K".