...broken mirrors...
A nice idea until Murphy's law strikes.
Not an issue. There are a couple of solutioons:
3 way mirror... Break off the third mirror, back it up, then re=attach it before the next backup.
Primary is raid-5. Secondary is a simple concatonation (I presume Windows can handle this sort of setup.) Once again, break off secondary to do backup.
Filesystem (such as Veritas) which allows snapshot filesystems. Snapshot systems keep track of old data when new data is written to the filesystem. This allows you to have a static snapshot of the filesystem (at least, until your snapshot reserve gets filled).
Bulletproof server (oh hold on -- that's SPAM, not RAID).
Note that 3-way mirrors are an especially good idea for read-mostly filesystems, since extra mirrors will increase read bandwidth -- but decrease write bandwidth (unless done with hardware support).
It was the '64k segments' thing that made the 80[12]86 into pigs for programming. I've done assembly language programming on at least 6 different architectures (including raw machine language programming for the 6809). When I learned how the 8086 was designed, I swore I'd never learn that processor (and I've kept that promise).
Among the grottiness of the 80x86 (x
but due to the design of the MMU only 48-bits worth will be available (currently its about 40 bitsbut due to the design of the MMU only 48-bits worth will be available (currently its about 40 bits)
Not a big deal. Building a single-image system with 1Terrabyte (40 bits) of ram would be rather like building a 2GB memory module for the original Mac-II [I calculated that you could camouflage the system as a desk. The entire top of the 'MemDesk(tm)' would be filled with memory modules. One pedestal of the desk would be the cooling unit and the other would be a 16Kilowatt power supply.])
Just for the sake of curiosity, what is the value of several human lives in comparison to a robotic system? I'm sure the astronauts would like to know what monetary value you've affixed to them.
A number of Astronauts have said that they'd happily take on the risks associated with a Hubble repair mission -- enough to fill the mision crew (as I understand it).
They know the risks far better than I do, and they'll be living the risks. If it's good enough for them, it's good enough for me.
Life is not without risks. We take on risk everytime we do amything from driving the car to going to sleep in our beds. The question is: "Is this worth the risk?". The answer to that question is unique to each individual. I would never do some of the things that Evil Kenevil did, and I'm sure that there are some things that I've done that he would baulk at (probably with good reason).
...and figure roughly 10000 atoms of silicon per bit, that means that if we were to mine all of the silicon out of the continents, make RAM out of all of it, and put all that RAM in one big giant computer, that computer would need to be designed with an address space 132 bits wide.
Obviously, you need to make smaller transistors. Being able to make only 16 such computers would be a waste of technology.
:-)
I still remember when the 386 was new and people were saying exactly the same things about the 16 to 32 bit transition that you are saying about 32 to 64 bits just now.
WHAT??!!
You must work around some incredibly small-minded people. I remember when the IBM-PC (8088) came out, and people were wonering why in god's name they would use an architecture that had an intrinsic 64K barrier. There were already so many programs that were breaking the 64K barrier for code not to mention data.
The Apple II, and most other popular computing platforms before the PC came out had already come up with bank-switching as an extermely gunky hack to get around the 64K barrier. Going to 16 bit architectures was seen as an excelent opportunity to get away from kludgy 64K barriers. The 8086 really flubbed that opportunity. All it really did was to formalize bank-switching into the CPU (vaguely reminiscent of the way that the then-declining PDP-11 solved it).
By the time the 80386 came out people were already using kludgy bank-switching to get around the 1Meg (OK: 640K) barrier of the 8086/PC architecture (think EMS/XMS). I can't think that anybody other than the most low-level toy-tinkerer would think that the '386 didn't solve a real and pressing problem. The 80[12]86 were failed attempts to solve the problem, and the conversation I remember about the '386 was more along the lines of "it's about time".
When the '386 came out, it wasn't uncommon to buy machines with more than 640K/1Meg -- not to mention breaking the 65K barrier.
That having been said, the 4GB limit is not as much of an issue today. Although not quit rare, it's still uncommon to see a desktop machine ship with more than 1GB of ram -- more than 4GB is rare for anything other than server boxes and high-end workstations.
I guess that the short form of this rant is that, although there's definitely a real use for 64bit machines, it's nowhere near the pressing need that it was for 32-bit by the time that '386 boxes came out.
Every time something like this happens, it pushes large companies a step closer towards realizing that perhaps software patents are not an entirely Good Thing
Don't hold your breath. Microsoft is probably one of the primary proponents of software patents -- and if the Eolas patent didn't get their attention (actually, I think it gave them the idea) I don't think anything will.
Besides, I'm not even sure if this is a pure software patent. This appears to be at least partially a hardware patent (although all I've read is the (rather sketchy) PR fluff and not the patents themselves.
In any event, Microsoft appears to have induced the company to reveal the details of their idea (under NDA?) and then wilfully infringed. Very different than the normal issues people have with software patents.
If you accept Alacritch's press reports at face value, this is not a patent minefiled problem. Microsoft went to the company and got details of their technology (probably under NDA). They then cut off communication with the company and implemented their idea without any sort of agreement.
This is an example of the kinds of minefield a small startup faces when they try to deal with Microsoft. It would appear that MS simply intended to walk all over this company and suck them dry on court costs. It's actually nice to see Microsoft get their ass nailed to the wall on this.
From the sounds of things, this is also (at least partially) a hardwar patent and not a pure software patent issue.
From the looks of things, this appears to be (at least partially) a hardware patent with software aspects to it. (which is quite different than a pure software patent).
In any event, even if the work wasn't patented, it would have been subject to trade secret complaints (if you accept Alacritech's version of the story)
So Microsoft is stealing someone else's Intellectual Property... Why aren't I shocked? I'm expecting that they figured that they'd get away with 4-5 years of slogging it thru a court before they had to face the fact that they were wilfully infringing a patent (by which time the company would have probably been defunct).
I think that companies are learning to expect this from Microsoft.
I would expect that urandom is probably better than gutmann's RNG, since it depends on the entropy pool that feeds/dev/random. As for I/O errors, I think that you can tell DD to skip over read errors (probably at the block level as you worry about), but the man page doesn't make reference to write errors. You might have to skip manually.
I remember having to so something like trying to determine the extent of errors on a drive that had a bad patch. The best I can think to do would be to write a PERL script to read the errors (if any) from dd, and respond accordingly with reruns of dd
One problem with this is that many ISPs are days (or even weeks) behind on responding to complaints. I have a script which automates the process of generating SPAM and virus complaints. In the cases where I've actually gotten a real-live response, it's almost invariably been days after my complaint. (It's only the smallest ISPs that seem to have a fast response time.) In the menatime, these machines have been spewing spam and viruses across the 'net.
If Telestra is like any other large ISP I've seen, I figure that the first thing they should do is hire (or allocate) a good gaggle of AUP investigators so that their intelligence on this problem is reasonably real-time.
They could also write some scripts to log and categorize the DNS queries that they're getting from their customers. It should be fairly easy to automatically identify the worst offenders. You could then send notes to their owners, and if there's no reasonable response, pull the plug. Over the last few years, I think that I've written scripts to do pretty much everything but the last step, so I know it's doable. (that last step should almost always be manual).
You don't have to read the drive partition to properly kill it. listening to the drive partition can actually cause you to miss critical data, if a partition has been removed ("it was the really critical data, so I removed the partition before I gave it to you"). In fact, you should be trashing the entire drive including the partition data.
In Linux, the command would be 'shred/dev/hda' The shred program will do multiple passes of random data (default 25)/proc/partitions should give you a list of the drives available and their sub-partitions. If you're shredding the entire drive (/dev/hda) you can safely ignore the partitons (/dev/hda[0-9]+).
Knoppix boots off of a CD-ROM, so you don't have to worry about accidently erasing the OS -- and, as long as the machine you're
wiping can boot off of a CD, you can even shred the drive in place
before you take it out to ship it (just make sure you shred the right drive).
If you want an 'appliance' then you can build it with
A boot CD-ROM
an external USB drive enclosure which does IDE->USB.
(to 'load', you just unplug the enclosure, load a new drive, plugi it in. The system should recognize it. Syslog output should give you all the info you need).
a SCSI controller with adaptors for the various drive plugs. If you start with an ultra-wide drive, you should be able to get adapters that will allow you to go all the way back to old SCSI-1.
an SATA controller/plug
I don't know if there are any native-firewire drive. Chances are you should be able to take the raw (ide/scsi) drive out of the enclosure and use one of the other connectors to munch it.
Spare cables... If you do this often enough, the cables will start to wear out. As they do replace the cable in the system (avoid plugging/unplugging the end that connects to the board as much as possible to avoid permanent impairment).
You may have to build your own kernel to recognize RLL drives and controllers, or older, but Linux should recognize it. You will probably need a mobo that accepts ISA cards (P3 or earlier CPUs, generally).
For something that even a custom-build Linux kernel can't write to, I think you'd probably have to go to early '80s technology (or really esoteric hardware that's more recent).
How is it illegal to (truthfully) tell people: "This ISP is harboring and supporting a spammer."? That's all that MAPS does.
If people conclude, from that information, that they don't want to accept email comming from that ISP, don't blame MAPS. It's an issue between you, your ISP, and the ones that don't like ISPs who support spammers.
Usually RBLs only block a netblock if an ISP keeps giving a spammer new sub-blocks as their current ones are blacklisted. This is a sign that the ISP is giving active suppport to the spammer, and also leaves the RBL with little choice other than listing the entire ISP's block as tainted. Once that's done, it really doesn't work to unblock any sub-block that claims to be innocent... You really can't tell if that sub-block has been actually (or will be subsequently) assigned to the spammer as they play wack-a-mole, and figuring out if that's the case would be far to time-consuming.
In any case, your contract is with your ISP. In fulfilling your contract, its your ISP that's responsible for being a good netizen and not pissing off millions of people and many other ISP by doing things like harboring spammers. If you should be suing anybody, I'd say it's your ISP you should be suing (unless you can provide some proof that your ISP is innocent of 'harboring a spammer' charges).
Your defamation claims are based on an assertion that MAPS is telling people that your IP is spamming. No such thing. MAPS makes it very clear that banning entire netblocks may (and probably will) also hit some non-spamming customers of that ISP. People who use MAPS, do so knowing the nature of such blocks. Their information isn't about your IP, it's about your ISP.
BTW: If you can make this defamation suit stick against MAPS, you'll make far more money launching a class-action suit against credit bureaus.
How about DVD-CSS? or Microsoft's patents on using XML the way that it was designed to be used? or the details on Microsoft's horridly convoluted NTFS? or......
In this case, the side effects of the shotgun are productive.
Those ISPs that harbour spammers do so because the spammers offer them extra money for so-called 'bullet proof' serving. The only way to discourage such ISPs from doing this is to hit them in the same place. Having customers leave when your netblock is blackholed strikes the same pocketbook that the spammers try to pad.
If your ISP represents to you that they don't support spamming, but then get blackholed for providing services to known spammers, then you might have a reason to sue your ISP for misrepresentation and constructive breach of contract (such blackholeing is a predictable result of harbouring spammers).
If you go to an ISP that doesn't make you agree to an AUP that bans spamming then you're just asking to get your ass nailed by a blackhole shotgun.
Many liberals are all for more privacy. I get the sense that conservatives are all for more privacy for big business, but they could care less about the rights of the average citizen.
Have you considered that you might be gay?
I have sometimes considered that I might be a lesbian with the wrong plumbing, but that's an entirely different story.
There are a number of ways to reap the fruits of your innovation. One is to throw it into a public pool where others can add to your work and return it back to you. The GPL ensures that the changes that others add will come back to me. Licenses like Sun's CDDL are good for Sun, but if I contribute my code, then Sun can take it and make a proprietary version of it. This may be good for Sun, but it's not so good for me and the rest of the wider community.
BSD licenses are similar, except that they allow anybody, to close off the source and produce a proprietary version -- including someone who has added little real value to the code other than embrace-and-extend incompatibilities and the marketing clout to make those incompatible changes 'standard'. AT&T tried to do that with the original BSD code. Their only mistake was that they removed most of the BSD copyright notices from the code they used.
When they block a netblock they're very clear that they're doing it because the controlling ISP is harboring a spammer and acknowledge that individual IPs within the block may, or may not be 'innocent'. There's no untruth in that statement to constitute defamation.
They're not in the mail delivery business. They're in the business of helping people keep spam out of their mailboxes. They're not in competition with you. There's nothing unfair about telling people that your ISP harbours spammers (if that's a true statement).
I don't think that they're doing any interference with your contracts. It's the ISPs who insist on using their list that are doing the blocking, and they have the right to refuse to accept emails from whomever they (don't) like.
Given that a number of jurisdictions (including the federal govt.) have passed laws with the intent of stemming the flow of spam, I'd expect that your suit would get laughed out of court. In some states, you might even get hit with an anti-SLAPP countersuit.
A real-life analogy:
Some time ago, I was running a Tribes(1) game server. At one point, I had someone come in who made a hobby of crashing the server. After trying a number of solutions, I finally concluded that the only solution was to block his packets, so I build a BSD firewall, and blocked his IP. After he switched his IP a few times (he claimed to be friends with the ISP's owner) I finally started blocking entire netblocks In the end, I blocked two/14s and a/22 (8 class Bs and 4 class Cs) belonging to his ISP. I also blocked a/8 belonging to AOL when he got an account there.
Now, did I consider the probability that I was probably blocking out a few innocent players? yep!
It didn't matter, though. My server stayed up, and that was more important to me, in the moment, than pandering to a handfull of innocent players. Blocking over 1/2million IP's is, statistically well under 0.1% of the internet and less than 1% of the US IP pool. Bluntly, a drop in the bucket.
In this case, MAPS has probably concluded that this ISP is doing something similar for the spammer (providing alternative IP addresses whenever one gets blocked). When that happens blocking the IP's as they show up is essentially an endless game of whack-a-mole with the spammr popping up, sending spam till they get blocked and then moving to a new location. At that point the only effective action is to start flagging entire netblocks (as I did). Extreme, but effective.
The poster goes to pains to point out that a massive 180,210 IPs (that is such a strange number. Where did it come from?) have been blocks, but goes to equal pain to avoid identifying either the ISP or the specific netblock(s) which were blocked.
If we go thru the history if the ISP and netblock in question, we may find that an infamous spammer has been using it for the last 6 months with no attempt by the ISP to resolv the problem despite many warnings from MAPS and other anti-spam organizations -- or we may find that MAPS went on a wildcat strike.
Given the very vague real data about this dispute, I'd be inclined to tell the complainant that he's probably the customer of a hardened spam provider, and he may be best to find another provider (as unpleasant as the move will be). If we get more than generic information, I may be able to giver more than a generic suggestion.
Usually Usenet death penalties are a last resort. MAPS may seem like they're assholes, but my guess is that they're finding themselves dealing with some assoles of their own (i.e. the offending ISP). In the moment, they can't tell the difference between you, and the offending spammer(s) who triggered this showdown. (( I'll presume, for the sake of argument, that you're not a spammer yourself )).
They're not willing to deal with you because their beef is with the ISP, and that's the only place where the problem can be resolved. They're iconveniencing you because it's probably one of the few tools left that they have to push your ISP to stop inconveniencing the entire internet.
Sun is one of the most innovative tech companies that has ever existed and its success stems from that,...
Sun started out as a BSD company and benefited greatly from the bazzar development process. BSD is a congolomeration of the contributions from thousands of UNIX users from hundreds of universities (kinda like open source). By the time AT&T sued BSD, the vast bulk of 'proprietary' UNIX was actually contributions from the users who had access to the source code and contributed their IP back to the community.
Some of the reasons why the GPL and BSD got the support they did was that people were upset seeing the work that they put into UNIX closed off, and made proprietary. It can be really a pisser to find someone charging you to use a program that's mostly written by you and contributed to the community.
Although AT&T unix was technically a proprietary product, a legal accident meant that, for most universities it was, virtually speaking, kinda open source. Unix benefitted greatly from this accident and by the 1990s the vast majority of the code came from the pseudo-open sourc community (as AT&T found, much to their chagrin, when they tried to sue Berkeley for misusing 'their' code).
If you think that proprietary licenses promote innovation, then take a look at what has happened to UNIX, Linux and BSD since AT&T closed off the license. Linux has come from nowhere to be at the forefront of the innovative wave. BSD isn't far behind, and SCO, who now claims ownership of the UNIX base is reduced to launching a pitiful lawsuit against IBM to bolster their stock while their software product languishes in a proprietary backwater.
Even SUN has been force back into the Open Source world (Bill Joy was one of the prime movers in the early days of the original BSD). IBM -- once the darling of software monopolists -- who tried to do a proprietary rewrite of UNIX (AIX) has embraced Linux, and even Microsoft has been forced to pay at least lip service to Open Source with their shared source set of initiatives.
It's a pattern that tells me it's not an accident.
I think that it's improper to call Larry a capital asshole. It seems to me that he really did try to straddle the line between proprietary and open source, and he did it in a way that failed. Hopefully this failure will be a learning opportunity for both the Open Source Community and Larry.
This excercise hasn't been a complete loss for either Bitmover Corp. or for the Open Source community. Both have gotten something out of it, but now they're going separate ways.
Also note that BitMover is attempting to make the split as amicable as possible. He could have shut down support and distribution of the free version as of yesterday. Instead he seems to be committing to providing one last (critical) major update, and then close down development of the free version, as well as providing a few month's warning. If he was being an asshole, he would have waited until the Kernel was a week away from the 65K change limit and then dropped support with no warning.
This is something like breaking up with a girlfriend. You can do it in a respectful way, or you can do it with yelling screaming and personal items thrown out in the street. Larry seems to be doing the former. Calling him a capital bastard is pushing things in the other.
Most of my ex-girlfrinds I can still show up at the door at 9pm and be invited in for some (herbal) tea and a nice chat. I really can't quite wrap my mind around people who can't visit any of their exs' without a court order. It's just so disrespectful of the quality time and experiences that came out of the relationship (presuming that the relationship wasn't just a 'gimme' fight). Yes, does take some work to do an amicable breakup, but here's lots of value to being able to have a sane conversation with your ex. Don't knock it until you've tried it.
"But Schwartz said that some people he's spoken to dislike it because it precludes them from using open-source software as a foundation for proprietary projects."
That's the primary intent of the GPL. That's like complaining that water gets you wet. The intent of the GPL is that companies like SUN can't take my code, make minor changes, and claim proprietary ownership of the result (by only distributing the object code, and deckarubg the source code a trade secret)
Sun's CDL contains some wilfull holes that might allow SUN to later shut down others' uses of the code that they release. That's something that (AFAIK) the have declined to fix so far.
The countries don't have any obligation, except to the extent that they're extending GPL code. Even then it's the developer that only needs to allow the IP that they're adding to GPL code that they then wilfully distribute to be contributed to the world at large.
Slashdot Mirror
That they've managed to get a patent for this 'idea'?
A nice idea until Murphy's law strikes.
Not an issue. There are a couple of solutioons:
-
3 way mirror
... Break off the third mirror, back it up, then re=attach it before the next backup.
- Primary is raid-5. Secondary is a simple concatonation (I presume Windows can handle this sort of setup.) Once again, break off secondary to do backup.
- Filesystem (such as Veritas) which allows snapshot filesystems. Snapshot systems keep track of old data when new data is written to the filesystem. This allows you to have a static snapshot of the filesystem (at least, until your snapshot reserve gets filled).
- Bulletproof server (oh hold on -- that's SPAM, not RAID).
Note that 3-way mirrors are an especially good idea for read-mostly filesystems, since extra mirrors will increase read bandwidth -- but decrease write bandwidth (unless done with hardware support).Among the grottiness of the 80x86 (x but due to the design of the MMU only 48-bits worth will be available (currently its about 40 bitsbut due to the design of the MMU only 48-bits worth will be available (currently its about 40 bits)
Not a big deal. Building a single-image system with 1Terrabyte (40 bits) of ram would be rather like building a 2GB memory module for the original Mac-II [I calculated that you could camouflage the system as a desk. The entire top of the 'MemDesk(tm)' would be filled with memory modules. One pedestal of the desk would be the cooling unit and the other would be a 16Kilowatt power supply.])
A number of Astronauts have said that they'd happily take on the risks associated with a Hubble repair mission -- enough to fill the mision crew (as I understand it).
They know the risks far better than I do, and they'll be living the risks. If it's good enough for them, it's good enough for me.
Life is not without risks. We take on risk everytime we do amything from driving the car to going to sleep in our beds. The question is: "Is this worth the risk?". The answer to that question is unique to each individual. I would never do some of the things that Evil Kenevil did, and I'm sure that there are some things that I've done that he would baulk at (probably with good reason).
Obviously, you need to make smaller transistors. Being able to make only 16 such computers would be a waste of technology.
:-)
WHAT??!!
You must work around some incredibly small-minded people. I remember when the IBM-PC (8088) came out, and people were wonering why in god's name they would use an architecture that had an intrinsic 64K barrier. There were already so many programs that were breaking the 64K barrier for code not to mention data.
The Apple II, and most other popular computing platforms before the PC came out had already come up with bank-switching as an extermely gunky hack to get around the 64K barrier. Going to 16 bit architectures was seen as an excelent opportunity to get away from kludgy 64K barriers. The 8086 really flubbed that opportunity. All it really did was to formalize bank-switching into the CPU (vaguely reminiscent of the way that the then-declining PDP-11 solved it).
By the time the 80386 came out people were already using kludgy bank-switching to get around the 1Meg (OK: 640K) barrier of the 8086/PC architecture (think EMS/XMS). I can't think that anybody other than the most low-level toy-tinkerer would think that the '386 didn't solve a real and pressing problem. The 80[12]86 were failed attempts to solve the problem, and the conversation I remember about the '386 was more along the lines of "it's about time".
When the '386 came out, it wasn't uncommon to buy machines with more than 640K/1Meg -- not to mention breaking the 65K barrier.
That having been said, the 4GB limit is not as much of an issue today. Although not quit rare, it's still uncommon to see a desktop machine ship with more than 1GB of ram -- more than 4GB is rare for anything other than server boxes and high-end workstations.
I guess that the short form of this rant is that, although there's definitely a real use for 64bit machines, it's nowhere near the pressing need that it was for 32-bit by the time that '386 boxes came out.
Don't hold your breath. Microsoft is probably one of the primary proponents of software patents -- and if the Eolas patent didn't get their attention (actually, I think it gave them the idea) I don't think anything will.
Besides, I'm not even sure if this is a pure software patent. This appears to be at least partially a hardware patent (although all I've read is the (rather sketchy) PR fluff and not the patents themselves.
In any event, Microsoft appears to have induced the company to reveal the details of their idea (under NDA?) and then wilfully infringed. Very different than the normal issues people have with software patents.
This is an example of the kinds of minefield a small startup faces when they try to deal with Microsoft. It would appear that MS simply intended to walk all over this company and suck them dry on court costs. It's actually nice to see Microsoft get their ass nailed to the wall on this.
From the sounds of things, this is also (at least partially) a hardwar patent and not a pure software patent issue.
In any event, even if the work wasn't patented, it would have been subject to trade secret complaints (if you accept Alacritech's version of the story)
So Microsoft is stealing someone else's Intellectual Property... Why aren't I shocked? I'm expecting that they figured that they'd get away with 4-5 years of slogging it thru a court before they had to face the fact that they were wilfully infringing a patent (by which time the company would have probably been defunct).
I think that companies are learning to expect this from Microsoft.
I would expect that urandom is probably better than gutmann's RNG, since it depends on the entropy pool that feeds /dev/random. As for I/O errors, I think that you can tell DD to skip over read errors (probably at the block level as you worry about), but the man page doesn't make reference to write errors. You might have to skip manually.
I remember having to so something like trying to determine the extent of errors on a drive that had a bad patch. The best I can think to do would be to write a PERL script to read the errors (if any) from dd, and respond accordingly with reruns of dd
If Telestra is like any other large ISP I've seen, I figure that the first thing they should do is hire (or allocate) a good gaggle of AUP investigators so that their intelligence on this problem is reasonably real-time.
They could also write some scripts to log and categorize the DNS queries that they're getting from their customers. It should be fairly easy to automatically identify the worst offenders. You could then send notes to their owners, and if there's no reasonable response, pull the plug. Over the last few years, I think that I've written scripts to do pretty much everything but the last step, so I know it's doable. (that last step should almost always be manual).
In Linux, the command would be 'shred /dev/hda' The shred program will do multiple passes of random data (default 25) /proc/partitions should give you a list of the drives available and their sub-partitions. If you're shredding the entire drive (/dev/hda) you can safely ignore the partitons (/dev/hda[0-9]+).
Knoppix boots off of a CD-ROM, so you don't have to worry about accidently erasing the OS -- and, as long as the machine you're wiping can boot off of a CD, you can even shred the drive in place before you take it out to ship it (just make sure you shred the right drive).
If you want an 'appliance' then you can build it with
- A boot CD-ROM
- an external USB drive enclosure which does IDE->USB.
- a SCSI controller with adaptors for the various drive plugs. If you start with an ultra-wide drive, you should be able to get adapters that will allow you to go all the way back to old SCSI-1.
- an SATA controller/plug
- I don't know if there are any native-firewire drive. Chances are you should be able to take the raw (ide/scsi) drive out of the enclosure and use one of the other connectors to munch it.
- Spare cables... If you do this often enough, the cables will start to wear out. As they do replace the cable in the system (avoid plugging/unplugging the end that connects to the board as much as possible to avoid permanent impairment).
- You may have to build your own kernel to recognize RLL drives and controllers, or older, but Linux should recognize it. You will probably need a mobo that accepts ISA cards (P3 or earlier CPUs, generally).
For something that even a custom-build Linux kernel can't write to, I think you'd probably have to go to early '80s technology (or really esoteric hardware that's more recent).(to 'load', you just unplug the enclosure, load a new drive, plugi it in. The system should recognize it. Syslog output should give you all the info you need).
If people conclude, from that information, that they don't want to accept email comming from that ISP, don't blame MAPS. It's an issue between you, your ISP, and the ones that don't like ISPs who support spammers.
Usually RBLs only block a netblock if an ISP keeps giving a spammer new sub-blocks as their current ones are blacklisted. This is a sign that the ISP is giving active suppport to the spammer, and also leaves the RBL with little choice other than listing the entire ISP's block as tainted. Once that's done, it really doesn't work to unblock any sub-block that claims to be innocent... You really can't tell if that sub-block has been actually (or will be subsequently) assigned to the spammer as they play wack-a-mole, and figuring out if that's the case would be far to time-consuming.
In any case, your contract is with your ISP. In fulfilling your contract, its your ISP that's responsible for being a good netizen and not pissing off millions of people and many other ISP by doing things like harboring spammers. If you should be suing anybody, I'd say it's your ISP you should be suing (unless you can provide some proof that your ISP is innocent of 'harboring a spammer' charges).
Your defamation claims are based on an assertion that MAPS is telling people that your IP is spamming. No such thing. MAPS makes it very clear that banning entire netblocks may (and probably will) also hit some non-spamming customers of that ISP. People who use MAPS, do so knowing the nature of such blocks. Their information isn't about your IP, it's about your ISP.
BTW: If you can make this defamation suit stick against MAPS, you'll make far more money launching a class-action suit against credit bureaus.
How about DVD-CSS? or Microsoft's patents on using XML the way that it was designed to be used? or the details on Microsoft's horridly convoluted NTFS? or ......
Those ISPs that harbour spammers do so because the spammers offer them extra money for so-called 'bullet proof' serving. The only way to discourage such ISPs from doing this is to hit them in the same place. Having customers leave when your netblock is blackholed strikes the same pocketbook that the spammers try to pad.
If your ISP represents to you that they don't support spamming, but then get blackholed for providing services to known spammers, then you might have a reason to sue your ISP for misrepresentation and constructive breach of contract (such blackholeing is a predictable result of harbouring spammers).
If you go to an ISP that doesn't make you agree to an AUP that bans spamming then you're just asking to get your ass nailed by a blackhole shotgun.
Many liberals are all for more privacy.
I get the sense that conservatives are all for more privacy for big business, but they could care less about the rights of the average citizen.
Have you considered that you might be gay?
I have sometimes considered that I might be a lesbian with the wrong plumbing, but that's an entirely different story.
BSD licenses are similar, except that they allow anybody, to close off the source and produce a proprietary version -- including someone who has added little real value to the code other than embrace-and-extend incompatibilities and the marketing clout to make those incompatible changes 'standard'. AT&T tried to do that with the original BSD code. Their only mistake was that they removed most of the BSD copyright notices from the code they used.
They're not in the mail delivery business. They're in the business of helping people keep spam out of their mailboxes. They're not in competition with you. There's nothing unfair about telling people that your ISP harbours spammers (if that's a true statement).
I don't think that they're doing any interference with your contracts. It's the ISPs who insist on using their list that are doing the blocking, and they have the right to refuse to accept emails from whomever they (don't) like.
Given that a number of jurisdictions (including the federal govt.) have passed laws with the intent of stemming the flow of spam, I'd expect that your suit would get laughed out of court. In some states, you might even get hit with an anti-SLAPP countersuit.
A real-life analogy: /14s and a /22 (8 class Bs and 4 class Cs) belonging to his ISP. I also blocked a /8 belonging to AOL when he got an account there.
Some time ago, I was running a Tribes(1) game server. At one point, I had someone come in who made a hobby of crashing the server. After trying a number of solutions, I finally concluded that the only solution was to block his packets, so I build a BSD firewall, and blocked his IP. After he switched his IP a few times (he claimed to be friends with the ISP's owner) I finally started blocking entire netblocks In the end, I blocked two
Now, did I consider the probability that I was probably blocking out a few innocent players? yep!
It didn't matter, though. My server stayed up, and that was more important to me, in the moment, than pandering to a handfull of innocent players. Blocking over 1/2million IP's is, statistically well under 0.1% of the internet and less than 1% of the US IP pool. Bluntly, a drop in the bucket.
In this case, MAPS has probably concluded that this ISP is doing something similar for the spammer (providing alternative IP addresses whenever one gets blocked). When that happens blocking the IP's as they show up is essentially an endless game of whack-a-mole with the spammr popping up, sending spam till they get blocked and then moving to a new location. At that point the only effective action is to start flagging entire netblocks (as I did). Extreme, but effective.
If we go thru the history if the ISP and netblock in question, we may find that an infamous spammer has been using it for the last 6 months with no attempt by the ISP to resolv the problem despite many warnings from MAPS and other anti-spam organizations -- or we may find that MAPS went on a wildcat strike.
Given the very vague real data about this dispute, I'd be inclined to tell the complainant that he's probably the customer of a hardened spam provider, and he may be best to find another provider (as unpleasant as the move will be). If we get more than generic information, I may be able to giver more than a generic suggestion.
Usually Usenet death penalties are a last resort. MAPS may seem like they're assholes, but my guess is that they're finding themselves dealing with some assoles of their own (i.e. the offending ISP). In the moment, they can't tell the difference between you, and the offending spammer(s) who triggered this showdown. (( I'll presume, for the sake of argument, that you're not a spammer yourself )).
They're not willing to deal with you because their beef is with the ISP, and that's the only place where the problem can be resolved. They're iconveniencing you because it's probably one of the few tools left that they have to push your ISP to stop inconveniencing the entire internet.
Sun started out as a BSD company and benefited greatly from the bazzar development process. BSD is a congolomeration of the contributions from thousands of UNIX users from hundreds of universities (kinda like open source). By the time AT&T sued BSD, the vast bulk of 'proprietary' UNIX was actually contributions from the users who had access to the source code and contributed their IP back to the community.
Some of the reasons why the GPL and BSD got the support they did was that people were upset seeing the work that they put into UNIX closed off, and made proprietary. It can be really a pisser to find someone charging you to use a program that's mostly written by you and contributed to the community.
Although AT&T unix was technically a proprietary product, a legal accident meant that, for most universities it was, virtually speaking, kinda open source. Unix benefitted greatly from this accident and by the 1990s the vast majority of the code came from the pseudo-open sourc community (as AT&T found, much to their chagrin, when they tried to sue Berkeley for misusing 'their' code).
If you think that proprietary licenses promote innovation, then take a look at what has happened to UNIX, Linux and BSD since AT&T closed off the license. Linux has come from nowhere to be at the forefront of the innovative wave. BSD isn't far behind, and SCO, who now claims ownership of the UNIX base is reduced to launching a pitiful lawsuit against IBM to bolster their stock while their software product languishes in a proprietary backwater.
Even SUN has been force back into the Open Source world (Bill Joy was one of the prime movers in the early days of the original BSD). IBM -- once the darling of software monopolists -- who tried to do a proprietary rewrite of UNIX (AIX) has embraced Linux, and even Microsoft has been forced to pay at least lip service to Open Source with their shared source set of initiatives.
It's a pattern that tells me it's not an accident.
This excercise hasn't been a complete loss for either Bitmover Corp. or for the Open Source community. Both have gotten something out of it, but now they're going separate ways.
Also note that BitMover is attempting to make the split as amicable as possible. He could have shut down support and distribution of the free version as of yesterday. Instead he seems to be committing to providing one last (critical) major update, and then close down development of the free version, as well as providing a few month's warning. If he was being an asshole, he would have waited until the Kernel was a week away from the 65K change limit and then dropped support with no warning.
This is something like breaking up with a girlfriend. You can do it in a respectful way, or you can do it with yelling screaming and personal items thrown out in the street. Larry seems to be doing the former. Calling him a capital bastard is pushing things in the other.
Most of my ex-girlfrinds I can still show up at the door at 9pm and be invited in for some (herbal) tea and a nice chat. I really can't quite wrap my mind around people who can't visit any of their exs' without a court order. It's just so disrespectful of the quality time and experiences that came out of the relationship (presuming that the relationship wasn't just a 'gimme' fight). Yes, does take some work to do an amicable breakup, but here's lots of value to being able to have a sane conversation with your ex. Don't knock it until you've tried it.
That's the primary intent of the GPL. That's like complaining that water gets you wet. The intent of the GPL is that companies like SUN can't take my code, make minor changes, and claim proprietary ownership of the result (by only distributing the object code, and deckarubg the source code a trade secret)
Sun's CDL contains some wilfull holes that might allow SUN to later shut down others' uses of the code that they release. That's something that (AFAIK) the have declined to fix so far.
The countries don't have any obligation, except to the extent that they're extending GPL code. Even then it's the developer that only needs to allow the IP that they're adding to GPL code that they then wilfully distribute to be contributed to the world at large.