TiVo's do not have a high failure rate. Of the 11 TiVo's owned by friends and family only one has had issues -- the modem blew out due to a power surge. An issue which has been largely resolved in the newer TiVo's, and is completely irrelevant with the S2 TiVo's if you use broadband connections. It can also be worked around by using an external modem.
And neither TiVo's financial reportings nor the TiVo Community boards bear out your alleged "high failure rate".
TiVo's are little more than a PC with some specialty hardware and a lot of specialty software. The hardware is all integrated circuits, a power supply, and a HD. The only mechanical systems there are the fan on the PS and the HD. Neither have particularly high failure rates, and the HD's actually appear to have much lower failure rates than one would expect from consumer grade IDE drives -- there are still tons of S1 TiVo's running with their original HD's out there, dating back to the late 90's. Oh, and if either one of these does fail, it's pretty trivial to replace it and keep on going.
If you're not lucky, the damn thing will go psychotic
What lovely terminology there... care to explain yourself? Right now it's just sounding like more BS.
TiVo's biggest downsides are the cost (esp. for the subscription) and the perceived lack of features (as compared to something like MythTV). Whether or not the cost bugs you is obviously a personal choice. As for MythTV/FreeVo -- completely different markets being served. MythTV/FreeVo is for people who like to hack and twiddle and stuff. TiVo is for people who just want things to work and do a job well.
True, I just brought it back to the kernel specifically since that's what SCO is claiming infringement on.
As for other OSS, obviously it would be up to IBM to decide to prosecute or not. Of course, they better be careful that they're not distributing it already:) (And, that none of those packages they are distributing is utilizing it... as long as the two packages were using the same license, or a compatible license and complying with it, they'd still be SOL).
Now if all these other respondants could get it through their skulls that I never implied that IBM was claiming patents in Caldera Linux...
IBM probably took this long to prepare the suit because they were doing patent research on their own portfolio, comparing it to SCO's claims of their products, and seeing where the two overlapped.
And it's entirely possible that they're not done with the research yet.
On top of that, SCO kept on putting out press releases that needed to be responded to. Eventually IBM must've felt it was time to respond and put out this beheomoth of a brief. I expect it'll be several days or weeks before SCO responds to this -- 45 pages of legalese takes awhile to comprehend fully, and doing an off-the-cuff response could torpedo their entire suit.
As for a "quick sweep" though -- forget it. Unless SCO suddenly decides that it's a good idea to settle, it's still going to be years before this is over.
Right you are... I thought IBM distributed the Linux for zSeries themselves, but it's actually a SuSE distro. And IBM doesn't deliver it either -- quoting from the Linux for the IBM eServer zSeries page: "SuSE Linux Enterprise Server 8 for S/390(R) (delivered directly from SuSE Linux AG)"
This might actually give them legal recourse if someone submitted something that was distributed before they noticed it. A very shrewd legal decision on IBM's part. It also nicely invalidates pretty much everything I said in my previous post regarding this:)
As Mr. Perens says, software patents are eeeeevil!
I wasn't claiming that IBM is attempting to enforce patent infringement on any GPL'd code. In fact, I read that portion quite carefully since I knew they couldn't do so.
And it's inside four different products, not just Unixware.
IBM can't use the patents against the Linux kernel anymore than SCO can -- as IBM points out in its own brief, distribution of the source under the GPL prohibits you from certain legal tactics... like licensing. You are implicitly licensing any IP (copyrights, trade secrets, patents) that you may own that are applicable to the source you distribute. Don't like it? Don't distribute. Also don't use GPL code in a distributed product, since doing so requires distribution of the source.
Of course, one day a patch could come along that IBM doesn't like and it would cease to provide newer kernels including that patch in order to preserve its rights... but that's a battle that would have to be fought when that time comes. Most likely IBM would inform the relevant people of their rights and their intentions, and the code wouldn't be integrated into the kernel. I'm not saying that this is a good thing, or that software patents are a good thing, just pointing out how it would work.
Sure they can legally do this. Can they ethically do this is another question.
You know what? I'll sell you a license for you to continue using Linux without being sued by me. Only $5.
What? What's my basis for this? Well, without it, I just might sue you!
Sure, I have absolutely no code in the kernel or any other basis to sue you, but that doesn't prevent me from selling licenses to fools.
And, right now, the only people who would buy this license are fools. SCO has absolutely zero proof that there's anything they could prosecute over. They have claims, but any damn fool can make claims. Until they show some substantive evidence that they have a case there is absolutely no reason to pay their extortion.
Agreed, but I don't see how this case is considered "frivilous crap". RedHat may be able to show a distinct harm in business and/or name based on SCO's claims. Hell, the Gartner Group study is a key piece of evidence in this, as ironic as that may be. SCO explictly named RedHat in a recent conference call. And they've done all of this without filing a suit or putting forth any substantive proof. I'd be very surprised if a judge dismissed the case as "frivilous crap".
That said, these are some of the hardest legal claims to prove. RedHat would have to divulge some internal sales and profit data, which is why most of these kinds of cases don't ever get filed, and few move past the preliminary stages. If RedHat is willing to divulge that kind of information -- presuming they have it in the first place -- then they'd certainly have a case against SCO. Whether or not it's a winnable case is another matter.
And while I'd be tempted to say that the first two counts would be thrown out offhand (you can't really go to court to disprove something like copyright infringement AFAIK, at least not as the defendant-as-litigator), they are key to the remaining portions of the case. Kinda. We'll see.
Haven't seen the legal complaint, but my guess would be defamation of business and damage to business (or whatever the legal terminology is).
Hopefully this will force SCO to shut up and cut the FUD they've been spewing. At this point if they continued to put out statements like the ones they have the judge could hold them in contempt of court (presuming RedHat requests and gets a gag order against future statements).
Note that this is different from the claims they've lodged against IBM -- SCO has been reasonably quiet about the alleged trade secret infringements. Instead they've been claiming all kinds of copyright infringements without actually having filed a single lawsuit or legal brief regarding it. The copyright claims and the trade secret claims are completely separate.
The level of FUD streaming from SCO, Microsoft, and companies like the Gartner Group has been worrying me. While the Linux kernel may contain some infringing code, I doubt that it would take very long to replace. What caused harm to BSD in the AT&T/Berkeley days was most certainly not the judgement -- which basically handed AT&T its own ass on a platter, even though the BSD folk had to go change a few code sections -- but the FUD that arose during the trial. It kept private corporations from adopting any of the BSD variants, and its effects linger to the current day. The difference is that there was no private, for-profit company that would step up to the plate and force AT&T to shut up (not that, realistically, there were many companies big enough to try that kind of thing). The Linux world has several companies, both in and out of the USA, that sell Linux/Linux-related services as their primary profit method. And they're big enough to take on SCO (RedHat is technically bigger, with a market cap roughly 8x that of SCO and more profit last quarter than SCO has made in it's entire existance). This suit could cut the SCO FUD off at the knees (although it won't prevent companies like the Gartner Group from publishing "reports") and severely reduce the damage that can be done before the cases go to court (it's going to be 5 years folks).
There is, however, a (d) -- to use the concept of jurisprudence (which is valid in the UK as well as the US, since both are based on common law) to decline to convict someone of breaking an unjust law.
Which is why serving on a jury is one of the most important things you can do. No, it's generally not fun, it's a pain in the ass and a disruption from what you want to do, but it's still important. Especially if you can get on a jury that can affect the ruling in a case regarding an unjust law.
I don't believe the UK has juries like we do in the US, so it may not be applicable there, but it certainly is in every US state (except, perhaps, Louisiana)
Even this info is useless, because wsprintf is insecure only if it is used the wrong way
Yes, but the point being that it's pretty damned easy to use it in the wrong way. More importantly, it's very likely that someone else will come behind you to patch the program and end up using it the wrong way. End result? Don't use wsprintf()... at the very least use wsnprintf() (or whatever the hell the equivalent of snprintf() is for wide character sets). I know, snprintf() isn't standard, but it's implemented on virtually every platform nowadays. And there's other ways to avoid using the insecure function anyway.
Besides, everyone uses std::string these days
Well, I do, but I write C++. Anyone writing C is going to have a helluva time using any of the STL! If you do write C, I highly recommend picking up and using a good string library. Look at vsftpd's for example -- they've been very careful to ensure no string buffer issues.
There's still deadlocks, race conditions, and other security holes to worry about, but buffer overruns are by far the most common (and most easily exploitable).
This doesn't stop them from assuming Linux fans will host their games via dedicated servers though
Uh... yeah... right. I'm sure it has absolutely nothing to do with the minor point that a rather large percentage of co-lo/hosted systems are running Linux, and that having a server available for said systems dramatically increases the number of potential dedicated servers.
Making a dedicated server for another OS is a pretty easy thing. Your networking, physics, and other core code is probably fairly platform independant. There may be tweaks, but they're going to be minor unless you vastly deviated from standards.
Making the UI is another matter. DX9 ports poorly (as in - not at all). There is very little financial incentive for doing so, since the user base is insanely small, and the number of users who would buy for platform X but haven't already bought for Windows is even smaller -- and you'd be burned at the stake if you tried to sell the engine twice, instead of just the content.
Bitching about the community support from Valve shows just how inanely biased and out of touch with reality you are.
Oh... and I returned HL when it first came out. I'm not a Valve fan boy, although I am very impressed with the HL2 demos thus far.
Furthermore, a 256 Meg USB2 drive is cheaper than a new burner.
Really? OfficeDepot (or maybe OfficeMax) just had a 40x CD-RW for free. And you can get them for $20-40 without any rebates or any other hassle. The 256MB USB2 Flash cards have come way down in price, but they're not that cheap yet.
Of course, that's not usable in a notebook:)
If your burner is slow then your drive is either outdated or you have bad software... modern -RW's run at 32x and can rewrite a full CD in 3 mins. Flash memory is insanely slow in writing, so you'd probably be about even in overall time.
All of that said, I do think a USB dongle device would be more convienent and easier to deal with than CD's -- smaller, easier to change, etc. -- but I had to correct your statement.
I think $199 is overpriced for an OS, but I did wind up paying $149 for my copy of XP Pro. $49-100 is more inline IMO, but that's just me... the OS should not cost as much as an entire PC (sans monitor).
That said, I suspect many people won't jump through the hoops to get a refund, even if applicable, simply because it's not worth their time. Small claims court isn't time intensive, but even if the entire process only takes 10 hours (including travel time to and from the courthouse, time on the phone with Customer No Service, writing letters, etc.) then you've only recouped $20/hour. No, I'm not paid for my offtime, but I can tell you that I do value it at more than $20/hour in order to do what *I* want. And this doesn't fall into that category. If it does for you, more power to you, but don't expect everyone to think that this is a "good deal".
That said, it's an incomplete story... the author hasn't yet gotten his check. It's doubtful that the company would disobey the court order in this case -- you have that kind of problem with small claims rulings against small businesses and individuals, but large corps aren't likely to disappear overnight and wouldn't want to deal with the contempt of court ruling that would inevitably follow -- but until he receives the check it's still up in the air.
Additionally, the philosophical point here is to screw MS. Well, this case doesn't do that. Maybe if a few thousand cases were lobbied against a single manufacturer, but right now it's the computer maker that's out the money, not MS. And while they've lost money, they haven't lost anywhere close to enough to revamp their licensing agreement. Worse yet, even if several thousand orders were served against a single computer manufacturer the end result is likely to be that that manufacturer goes belly up -- they'll be out a great deal of money upfront, decide to renegotiate their licenses with MS, and then lose competitive pricing due to the new pricing structure. This really is a case of squeezing the middle man.
I don't see anything that's going to change this short of massive governmental oversight into MS contracts... if nothing else a court order stating their contracts have to be public (which would probably be unconstitutional) or equally draconian interference.
No, I don't have a solution. Most complex issues don't have simple answers.
They made the example because formatting the drive is insufficient -- the partition table itself must be encrypted, and that requires repartitioning. Since 99% of all PCs are used in Windows, it's a fairly reasonable thing to talk about how to do it for that OS.
They are probably using the File Allocation Table itself (the FAT in FAT) to store some encryption data or authentication info. That would also tie them to a particular implementation of FAT most likely.
Uh, except that you don't set that up in fdisk. fdisk merely creates the partition table. You can decide what FS is in which partition independant of that.
Any hardware specialists in the house to speculate?
They mention using FDISK from a DOS prompt to set your drive up, so existing installs and non-windows machines need not apply.
How'd you make that leap of intuition?
Yes, existing installs need not apply... that makes sense. They're writing encrypted data to the drive, and mixing encrypted and unencrypted would be a bad idea. But how on earth do you think this is tied to Windows?
The encryption is occurring at a BIOS/hardware level. You can run whatever OS you want and it'd work fine because the entire point was that you could setup the drive in a normal fashion -- you don't need to use any special tools to do it. If you were restricted to a particular OS or to a particular FS then you'd have to use Abit's own tools to do it.
Not such a cool idea IMO, more junk like the tube based audio they put out. Lots of flash with no substance, since if someone wants that data they're going to get it. I seriously doubt they implemented a sturdy enough encryption system to resist any significant governmental cracking... at least not one that can run in real time. It's mostly for the overly paranoid dweebs out there who don't realize that nobody wants to read their data.
Re:Internet Journalism at its Finest
on
Half-Life 2 Delayed?
·
· Score: 3, Insightful
As opposed to the unsubstantiated rumor that it's delayed in the first place? From such a highly regarded news source as Gamespot (which is where the "story" broke)?
NP... I've worked in the industry and had a lot more exposure to it than most people. There are a lot of scummy creditors out there, but there's also a ton of people out there that will happily screw anyone they can. All in all the credit bureau system helps even out the playing field... it's just a matter of ensuring that it does stay level (and, like I've said, I think some of the recent laws push the teeter toward corporations).
One thing I didn't explain well, but should've pointed out, is that the credit bureaus started out not by sharing information on everyone -- just on the people who weren't trustworthy. If you passed a bad check, they wanted to know about it. Didn't give a damn if you paid ontime or not. And to some extent that perseveres to today -- if you want credit data you can pay a certain amount. If you pass back data on your own deadbeats then you get a rather hefty discount. If you pass back information on everyone, you get an additional discount, but it's a pretty small one.
Of course, as financial institutions have discovered, there's different levels of good and bad credit, and it's profitable (and sane) to offer different levels of service to those different levels. If I know that you've always paid your bills ontime, have a solid income, etc. then I may want to offer you the best rate possible to get your business. If you've been dodgy on a few bills in the past, then maybe I'll raise that rate a bit to hedge against the chance that you'll be dodgy with me. And if you have clear credit issues then maybe I shouldn't offer you a loan at all.
I'm not really for credit checks on hiring, insurance, etc. though because except in a few cases (like when you're responsible for company money) there's not a convincing argument for the purpose. I know some studies have shown that people with bad credit can be less reliable, but they've been far from conclusive and you also start verging on a virtual debtors' prison with it. The evidence for insurance and credit checks is even more flimsy.
troustrup's style, which I *hated* from day one of, "int* i" which of course lead to people using, "int* i, j".
Only because those people were idiots, they didn't actually understand what the hell they were declaring, and they were using bad style to perpetuate the previous two issues.
Not bad style as in "int* i", but bad style as in declaring multiple variables on one line. It's a complete non-issue if you declare each variable on it's own line. It also makes the code far more readable, and makes types much more obvious (esp. if you use Stroustrup's style -- since the type of the variable is not "int", it's "int*" -- the two are different and you intermix them at your own peril). I think a lot of people would like to do away with the godawful C syntax of type var1, var2, var3; -- because of reasons you mention. It just leads to obfusication and bad practices. It was retained in C++ only because of backwards compatibility.
BTW, your code is bugged since you assign a value to *j without allocating memory to it. That will crash you on some platforms or in some compile modes. As well it should.
Well, just make sure you buy a brand of speakers that's good and keep the line around for awhile... this pretty much precludes anything from a big box store. You want timbrally matched speakers in all 5 positions -- preferably the same speaker in all 5, but that's not realistic for most people. If you buy from BB/CC/etc then you'll end up with speakers that you can't match when you purchase more down the line, and that'll suck. I bought my first set of speakers (Paradigm Titans) nearly a decade ago and I could still buy new ones/replacements from numerous dealers (online and off) right now.
My setup isn't perfect, but it's pretty damn good, and I have no intention of changing it anytime soon. Paradigm PS-60 mains, LCR450 center, and Titan rears. Subwoofer is a Paradigm PS-1000 -- not the greatest sub, but it kicked major ass when paired with the Titans, was a good price, and does a decent job. Would like to replace it at some point (with something less boomy and less scratched up -- the outside was damaged in a move), but dunno that that'll happen any time soon.
And yes, the subwoofer is important... I started off with a Onkyo receiver, CD player, and the Titans. When I added the PS-1000 it totally changed the system and turned it from pretty good to amazing... and this was while I was in college. Got quite a few compliments on the system... as well as some complaints about the volume level:)
On the other hand, I never agreed to anything regarding the collection and analysis of data about me.
Actually, you did. You just didn't notice it amongst all the other mice type whenever you:
Applied for any kind of loan (mortgage, car, personal, school, etc.)
Applied for a checking account
Applied for a savings account
Applied for a credit card
Did virtually any other financial transaction beyond paying cash.
Why? Because all of those imply trust of one form or another. Most of them are the banking system trusting you -- and how the hell are they supposed to do that? Oh, sure, in the case of the mortgage and car loan they can come for the physical item (foreclosure or reposession), but what about all those unsecured debts? Hell, how do they know that you're not going to write bad checks? The answer is in the credit bureaus. They started off for precisely this reason -- nobody knew who they could trust, which is stifling in a lend-borrow situation. How am I supposed to start up a new business if nobody will loan me the money, because nobody knows if I can be trusted or not?
The history of the credit bureaus starts small, with individual companies agreeing to share info about who's a good risk and who isn't. Eventually the banks started to get in on it because they do a lot more lending than most companies do. The various small networks consolidated and we wind up where we are today.
A mistake in the data could lead to a erroneous and defamatory conclusion, going so far as to limit my job prospects! Fixing the data, takes way too much time and way too much money.
BS. If you think that, you don't know the laws. Pull your credit report. See anything wrong on it? Notify the credit bureau and the company in question. They have 30 days to fix it, and the credit bureau must send you a new copy of your credit report at the end of that time. If they fail to do so, you can sue the bureau and the reporting company... and for enough money that you probably won't need to worry about job prospects in the future. And the judge is going to be on your side, not on the credit bureau's. That's how the law is written. They must have very definitive proof of what's in your report, or they're screwed. The bureaus now take this kind of thing seriously because it is big bucks if they don't fix it.
Yes, stuff can still go wrong on your report, and there's the related issue of identity theft, but at least now the consumer has the power to get it fixed.
Opted out should be the default state.
I agree, and it's one of the things I deeply dislike about the law in question. What I hate even more (oddly enough) is that opting out isn't permanent, or even long term -- you have to opt out every YEAR. That's absurd.
Fine. As long as you understand that they then have the right to say "Certainly sir. And how would you like to pay your outstanding mortgage balance of $235702.46?"
Or to give you whatever money of yours they have, or do whatever's necessary to sever all financial ties with you immediately.
You're not a customer? Then they're not going to have crap for information on you. They may send you solicitations, but that information is acquired from the credit bureau. You can tell Citibank to be put on their do not solicit list, and then your data will get flushed early in the process whenever it gets pulled from the bureaus. Yes, I've worked in this field, doing this exact thing. If you don't want your data to be sold by the bureaus, you can request that from the bureaus as well. There are three major ones (Equifax, Experian, Trans Union) and a few hundred thousand small ones (all of whom feed the big three).
You don't actually expect a company to do business with you if they're not allowed to keep records, right? Might I suggest you do some research into how godawful the banking industry was prior to the introduction of the credit bureaus? Think "Good Ole Boys Network" and you'll have a start on it... but it was considerably worse.
I'm not saying that some additional protections on consumer privacy shouldn't be in place (as a bare minimum everyone should be entitled to viewing their own credit report on demand, for no more than cost of mailing or free online). And I'm also not saying that the pendulum hasn't swung too far in the wrong direction (the law a couple years ago allowing companies unprecedented sharing of consumer information went way too far). But anyone who makes statements like that generally has no clue how the financial system, particularly the credit portion of it, actually works.
Slashdot Mirror
And exactly what do you base this on?
TiVo's do not have a high failure rate. Of the 11 TiVo's owned by friends and family only one has had issues -- the modem blew out due to a power surge. An issue which has been largely resolved in the newer TiVo's, and is completely irrelevant with the S2 TiVo's if you use broadband connections. It can also be worked around by using an external modem.
And neither TiVo's financial reportings nor the TiVo Community boards bear out your alleged "high failure rate".
TiVo's are little more than a PC with some specialty hardware and a lot of specialty software. The hardware is all integrated circuits, a power supply, and a HD. The only mechanical systems there are the fan on the PS and the HD. Neither have particularly high failure rates, and the HD's actually appear to have much lower failure rates than one would expect from consumer grade IDE drives -- there are still tons of S1 TiVo's running with their original HD's out there, dating back to the late 90's. Oh, and if either one of these does fail, it's pretty trivial to replace it and keep on going.
If you're not lucky, the damn thing will go psychotic
What lovely terminology there... care to explain yourself? Right now it's just sounding like more BS.
TiVo's biggest downsides are the cost (esp. for the subscription) and the perceived lack of features (as compared to something like MythTV). Whether or not the cost bugs you is obviously a personal choice. As for MythTV/FreeVo -- completely different markets being served. MythTV/FreeVo is for people who like to hack and twiddle and stuff. TiVo is for people who just want things to work and do a job well.
True, I just brought it back to the kernel specifically since that's what SCO is claiming infringement on.
:) (And, that none of those packages they are distributing is utilizing it... as long as the two packages were using the same license, or a compatible license and complying with it, they'd still be SOL).
As for other OSS, obviously it would be up to IBM to decide to prosecute or not. Of course, they better be careful that they're not distributing it already
Now if all these other respondants could get it through their skulls that I never implied that IBM was claiming patents in Caldera Linux...
IBM probably took this long to prepare the suit because they were doing patent research on their own portfolio, comparing it to SCO's claims of their products, and seeing where the two overlapped.
And it's entirely possible that they're not done with the research yet.
On top of that, SCO kept on putting out press releases that needed to be responded to. Eventually IBM must've felt it was time to respond and put out this beheomoth of a brief. I expect it'll be several days or weeks before SCO responds to this -- 45 pages of legalese takes awhile to comprehend fully, and doing an off-the-cuff response could torpedo their entire suit.
As for a "quick sweep" though -- forget it. Unless SCO suddenly decides that it's a good idea to settle, it's still going to be years before this is over.
Right you are... I thought IBM distributed the Linux for zSeries themselves, but it's actually a SuSE distro. And IBM doesn't deliver it either -- quoting from the Linux for the IBM eServer zSeries page:
:)
"SuSE Linux Enterprise Server 8 for S/390(R) (delivered directly from SuSE Linux AG)"
This might actually give them legal recourse if someone submitted something that was distributed before they noticed it. A very shrewd legal decision on IBM's part. It also nicely invalidates pretty much everything I said in my previous post regarding this
As Mr. Perens says, software patents are eeeeevil!
I wasn't claiming that IBM is attempting to enforce patent infringement on any GPL'd code. In fact, I read that portion quite carefully since I knew they couldn't do so.
And it's inside four different products, not just Unixware.
As another poster said, read the thread.
Well, actually, they can't.
IBM can't use the patents against the Linux kernel anymore than SCO can -- as IBM points out in its own brief, distribution of the source under the GPL prohibits you from certain legal tactics... like licensing. You are implicitly licensing any IP (copyrights, trade secrets, patents) that you may own that are applicable to the source you distribute. Don't like it? Don't distribute. Also don't use GPL code in a distributed product, since doing so requires distribution of the source.
Of course, one day a patch could come along that IBM doesn't like and it would cease to provide newer kernels including that patch in order to preserve its rights... but that's a battle that would have to be fought when that time comes. Most likely IBM would inform the relevant people of their rights and their intentions, and the code wouldn't be integrated into the kernel. I'm not saying that this is a good thing, or that software patents are a good thing, just pointing out how it would work.
I want proof that they can legally do this
Sure they can legally do this. Can they ethically do this is another question.
You know what? I'll sell you a license for you to continue using Linux without being sued by me. Only $5.
What? What's my basis for this? Well, without it, I just might sue you!
Sure, I have absolutely no code in the kernel or any other basis to sue you, but that doesn't prevent me from selling licenses to fools.
And, right now, the only people who would buy this license are fools. SCO has absolutely zero proof that there's anything they could prosecute over. They have claims, but any damn fool can make claims. Until they show some substantive evidence that they have a case there is absolutely no reason to pay their extortion.
Agreed, but I don't see how this case is considered "frivilous crap". RedHat may be able to show a distinct harm in business and/or name based on SCO's claims. Hell, the Gartner Group study is a key piece of evidence in this, as ironic as that may be. SCO explictly named RedHat in a recent conference call. And they've done all of this without filing a suit or putting forth any substantive proof. I'd be very surprised if a judge dismissed the case as "frivilous crap".
That said, these are some of the hardest legal claims to prove. RedHat would have to divulge some internal sales and profit data, which is why most of these kinds of cases don't ever get filed, and few move past the preliminary stages. If RedHat is willing to divulge that kind of information -- presuming they have it in the first place -- then they'd certainly have a case against SCO. Whether or not it's a winnable case is another matter.
And while I'd be tempted to say that the first two counts would be thrown out offhand (you can't really go to court to disprove something like copyright infringement AFAIK, at least not as the defendant-as-litigator), they are key to the remaining portions of the case. Kinda. We'll see.
Haven't seen the legal complaint, but my guess would be defamation of business and damage to business (or whatever the legal terminology is).
Hopefully this will force SCO to shut up and cut the FUD they've been spewing. At this point if they continued to put out statements like the ones they have the judge could hold them in contempt of court (presuming RedHat requests and gets a gag order against future statements).
Note that this is different from the claims they've lodged against IBM -- SCO has been reasonably quiet about the alleged trade secret infringements. Instead they've been claiming all kinds of copyright infringements without actually having filed a single lawsuit or legal brief regarding it. The copyright claims and the trade secret claims are completely separate.
The level of FUD streaming from SCO, Microsoft, and companies like the Gartner Group has been worrying me. While the Linux kernel may contain some infringing code, I doubt that it would take very long to replace. What caused harm to BSD in the AT&T/Berkeley days was most certainly not the judgement -- which basically handed AT&T its own ass on a platter, even though the BSD folk had to go change a few code sections -- but the FUD that arose during the trial. It kept private corporations from adopting any of the BSD variants, and its effects linger to the current day. The difference is that there was no private, for-profit company that would step up to the plate and force AT&T to shut up (not that, realistically, there were many companies big enough to try that kind of thing). The Linux world has several companies, both in and out of the USA, that sell Linux/Linux-related services as their primary profit method. And they're big enough to take on SCO (RedHat is technically bigger, with a market cap roughly 8x that of SCO and more profit last quarter than SCO has made in it's entire existance). This suit could cut the SCO FUD off at the knees (although it won't prevent companies like the Gartner Group from publishing "reports") and severely reduce the damage that can be done before the cases go to court (it's going to be 5 years folks).
You got it precisely.
There is, however, a (d) -- to use the concept of jurisprudence (which is valid in the UK as well as the US, since both are based on common law) to decline to convict someone of breaking an unjust law.
Which is why serving on a jury is one of the most important things you can do. No, it's generally not fun, it's a pain in the ass and a disruption from what you want to do, but it's still important. Especially if you can get on a jury that can affect the ruling in a case regarding an unjust law.
I don't believe the UK has juries like we do in the US, so it may not be applicable there, but it certainly is in every US state (except, perhaps, Louisiana)
I loved it so much that I registered it.
Development ceased immediately.
And, of course, I no longer have my registration code.
Scorched3D showed up as a game in The All Seeing Eye last night as well, which is another Good Thing.
Even this info is useless, because wsprintf is insecure only if it is used the wrong way
Yes, but the point being that it's pretty damned easy to use it in the wrong way. More importantly, it's very likely that someone else will come behind you to patch the program and end up using it the wrong way. End result? Don't use wsprintf()... at the very least use wsnprintf() (or whatever the hell the equivalent of snprintf() is for wide character sets). I know, snprintf() isn't standard, but it's implemented on virtually every platform nowadays. And there's other ways to avoid using the insecure function anyway.
Besides, everyone uses std::string these days
Well, I do, but I write C++. Anyone writing C is going to have a helluva time using any of the STL! If you do write C, I highly recommend picking up and using a good string library. Look at vsftpd's for example -- they've been very careful to ensure no string buffer issues.
There's still deadlocks, race conditions, and other security holes to worry about, but buffer overruns are by far the most common (and most easily exploitable).
This doesn't stop them from assuming Linux fans will host their games via dedicated servers though
Uh... yeah... right. I'm sure it has absolutely nothing to do with the minor point that a rather large percentage of co-lo/hosted systems are running Linux, and that having a server available for said systems dramatically increases the number of potential dedicated servers.
Making a dedicated server for another OS is a pretty easy thing. Your networking, physics, and other core code is probably fairly platform independant. There may be tweaks, but they're going to be minor unless you vastly deviated from standards.
Making the UI is another matter. DX9 ports poorly (as in - not at all). There is very little financial incentive for doing so, since the user base is insanely small, and the number of users who would buy for platform X but haven't already bought for Windows is even smaller -- and you'd be burned at the stake if you tried to sell the engine twice, instead of just the content.
Bitching about the community support from Valve shows just how inanely biased and out of touch with reality you are.
Oh... and I returned HL when it first came out. I'm not a Valve fan boy, although I am very impressed with the HL2 demos thus far.
Furthermore, a 256 Meg USB2 drive is cheaper than a new burner.
:)
Really? OfficeDepot (or maybe OfficeMax) just had a 40x CD-RW for free. And you can get them for $20-40 without any rebates or any other hassle. The 256MB USB2 Flash cards have come way down in price, but they're not that cheap yet.
Of course, that's not usable in a notebook
If your burner is slow then your drive is either outdated or you have bad software... modern -RW's run at 32x and can rewrite a full CD in 3 mins. Flash memory is insanely slow in writing, so you'd probably be about even in overall time.
All of that said, I do think a USB dongle device would be more convienent and easier to deal with than CD's -- smaller, easier to change, etc. -- but I had to correct your statement.
I think $199 is overpriced for an OS, but I did wind up paying $149 for my copy of XP Pro. $49-100 is more inline IMO, but that's just me... the OS should not cost as much as an entire PC (sans monitor).
That said, I suspect many people won't jump through the hoops to get a refund, even if applicable, simply because it's not worth their time. Small claims court isn't time intensive, but even if the entire process only takes 10 hours (including travel time to and from the courthouse, time on the phone with Customer No Service, writing letters, etc.) then you've only recouped $20/hour. No, I'm not paid for my offtime, but I can tell you that I do value it at more than $20/hour in order to do what *I* want. And this doesn't fall into that category. If it does for you, more power to you, but don't expect everyone to think that this is a "good deal".
It's more the principle of the thing.
That said, it's an incomplete story... the author hasn't yet gotten his check. It's doubtful that the company would disobey the court order in this case -- you have that kind of problem with small claims rulings against small businesses and individuals, but large corps aren't likely to disappear overnight and wouldn't want to deal with the contempt of court ruling that would inevitably follow -- but until he receives the check it's still up in the air.
Additionally, the philosophical point here is to screw MS. Well, this case doesn't do that. Maybe if a few thousand cases were lobbied against a single manufacturer, but right now it's the computer maker that's out the money, not MS. And while they've lost money, they haven't lost anywhere close to enough to revamp their licensing agreement. Worse yet, even if several thousand orders were served against a single computer manufacturer the end result is likely to be that that manufacturer goes belly up -- they'll be out a great deal of money upfront, decide to renegotiate their licenses with MS, and then lose competitive pricing due to the new pricing structure. This really is a case of squeezing the middle man.
I don't see anything that's going to change this short of massive governmental oversight into MS contracts... if nothing else a court order stating their contracts have to be public (which would probably be unconstitutional) or equally draconian interference.
No, I don't have a solution. Most complex issues don't have simple answers.
You're deeply overthinking this. Vastly.
They made the example because formatting the drive is insufficient -- the partition table itself must be encrypted, and that requires repartitioning. Since 99% of all PCs are used in Windows, it's a fairly reasonable thing to talk about how to do it for that OS.
They are probably using the File Allocation Table itself (the FAT in FAT) to store some encryption data or authentication info. That would also tie them to a particular implementation of FAT most likely.
Uh, except that you don't set that up in fdisk. fdisk merely creates the partition table. You can decide what FS is in which partition independant of that.
Any hardware specialists in the house to speculate?
Yes.
They mention using FDISK from a DOS prompt to set your drive up, so existing installs and non-windows machines need not apply.
How'd you make that leap of intuition?
Yes, existing installs need not apply... that makes sense. They're writing encrypted data to the drive, and mixing encrypted and unencrypted would be a bad idea. But how on earth do you think this is tied to Windows?
The encryption is occurring at a BIOS/hardware level. You can run whatever OS you want and it'd work fine because the entire point was that you could setup the drive in a normal fashion -- you don't need to use any special tools to do it. If you were restricted to a particular OS or to a particular FS then you'd have to use Abit's own tools to do it.
Not such a cool idea IMO, more junk like the tube based audio they put out. Lots of flash with no substance, since if someone wants that data they're going to get it. I seriously doubt they implemented a sturdy enough encryption system to resist any significant governmental cracking... at least not one that can run in real time. It's mostly for the overly paranoid dweebs out there who don't realize that nobody wants to read their data.
As opposed to the unsubstantiated rumor that it's delayed in the first place? From such a highly regarded news source as Gamespot (which is where the "story" broke)?
Yeah, that's journalistic integrity right there!
NP... I've worked in the industry and had a lot more exposure to it than most people. There are a lot of scummy creditors out there, but there's also a ton of people out there that will happily screw anyone they can. All in all the credit bureau system helps even out the playing field... it's just a matter of ensuring that it does stay level (and, like I've said, I think some of the recent laws push the teeter toward corporations).
One thing I didn't explain well, but should've pointed out, is that the credit bureaus started out not by sharing information on everyone -- just on the people who weren't trustworthy. If you passed a bad check, they wanted to know about it. Didn't give a damn if you paid ontime or not. And to some extent that perseveres to today -- if you want credit data you can pay a certain amount. If you pass back data on your own deadbeats then you get a rather hefty discount. If you pass back information on everyone, you get an additional discount, but it's a pretty small one.
Of course, as financial institutions have discovered, there's different levels of good and bad credit, and it's profitable (and sane) to offer different levels of service to those different levels. If I know that you've always paid your bills ontime, have a solid income, etc. then I may want to offer you the best rate possible to get your business. If you've been dodgy on a few bills in the past, then maybe I'll raise that rate a bit to hedge against the chance that you'll be dodgy with me. And if you have clear credit issues then maybe I shouldn't offer you a loan at all.
I'm not really for credit checks on hiring, insurance, etc. though because except in a few cases (like when you're responsible for company money) there's not a convincing argument for the purpose. I know some studies have shown that people with bad credit can be less reliable, but they've been far from conclusive and you also start verging on a virtual debtors' prison with it. The evidence for insurance and credit checks is even more flimsy.
Excellent speakers... and yeah, you definitely need a sub with them :)
troustrup's style, which I *hated* from day one of, "int* i" which of course lead to people using, "int* i, j".
Only because those people were idiots, they didn't actually understand what the hell they were declaring, and they were using bad style to perpetuate the previous two issues.
Not bad style as in "int* i", but bad style as in declaring multiple variables on one line. It's a complete non-issue if you declare each variable on it's own line. It also makes the code far more readable, and makes types much more obvious (esp. if you use Stroustrup's style -- since the type of the variable is not "int", it's "int*" -- the two are different and you intermix them at your own peril). I think a lot of people would like to do away with the godawful C syntax of type var1, var2, var3; -- because of reasons you mention. It just leads to obfusication and bad practices. It was retained in C++ only because of backwards compatibility.
BTW, your code is bugged since you assign a value to *j without allocating memory to it. That will crash you on some platforms or in some compile modes. As well it should.
Well, just make sure you buy a brand of speakers that's good and keep the line around for awhile... this pretty much precludes anything from a big box store. You want timbrally matched speakers in all 5 positions -- preferably the same speaker in all 5, but that's not realistic for most people. If you buy from BB/CC/etc then you'll end up with speakers that you can't match when you purchase more down the line, and that'll suck. I bought my first set of speakers (Paradigm Titans) nearly a decade ago and I could still buy new ones/replacements from numerous dealers (online and off) right now.
:)
My setup isn't perfect, but it's pretty damn good, and I have no intention of changing it anytime soon. Paradigm PS-60 mains, LCR450 center, and Titan rears. Subwoofer is a Paradigm PS-1000 -- not the greatest sub, but it kicked major ass when paired with the Titans, was a good price, and does a decent job. Would like to replace it at some point (with something less boomy and less scratched up -- the outside was damaged in a move), but dunno that that'll happen any time soon.
And yes, the subwoofer is important... I started off with a Onkyo receiver, CD player, and the Titans. When I added the PS-1000 it totally changed the system and turned it from pretty good to amazing... and this was while I was in college. Got quite a few compliments on the system... as well as some complaints about the volume level
Actually, you did. You just didn't notice it amongst all the other mice type whenever you:
Why? Because all of those imply trust of one form or another. Most of them are the banking system trusting you -- and how the hell are they supposed to do that? Oh, sure, in the case of the mortgage and car loan they can come for the physical item (foreclosure or reposession), but what about all those unsecured debts? Hell, how do they know that you're not going to write bad checks? The answer is in the credit bureaus. They started off for precisely this reason -- nobody knew who they could trust, which is stifling in a lend-borrow situation. How am I supposed to start up a new business if nobody will loan me the money, because nobody knows if I can be trusted or not?
The history of the credit bureaus starts small, with individual companies agreeing to share info about who's a good risk and who isn't. Eventually the banks started to get in on it because they do a lot more lending than most companies do. The various small networks consolidated and we wind up where we are today.
A mistake in the data could lead to a erroneous and defamatory conclusion, going so far as to limit my job prospects! Fixing the data, takes way too much time and way too much money.
BS. If you think that, you don't know the laws. Pull your credit report. See anything wrong on it? Notify the credit bureau and the company in question. They have 30 days to fix it, and the credit bureau must send you a new copy of your credit report at the end of that time. If they fail to do so, you can sue the bureau and the reporting company... and for enough money that you probably won't need to worry about job prospects in the future. And the judge is going to be on your side, not on the credit bureau's. That's how the law is written. They must have very definitive proof of what's in your report, or they're screwed. The bureaus now take this kind of thing seriously because it is big bucks if they don't fix it.
Yes, stuff can still go wrong on your report, and there's the related issue of identity theft, but at least now the consumer has the power to get it fixed.
Opted out should be the default state.
I agree, and it's one of the things I deeply dislike about the law in question. What I hate even more (oddly enough) is that opting out isn't permanent, or even long term -- you have to opt out every YEAR. That's absurd.
Fine. As long as you understand that they then have the right to say "Certainly sir. And how would you like to pay your outstanding mortgage balance of $235702.46?"
Or to give you whatever money of yours they have, or do whatever's necessary to sever all financial ties with you immediately.
You're not a customer? Then they're not going to have crap for information on you. They may send you solicitations, but that information is acquired from the credit bureau. You can tell Citibank to be put on their do not solicit list, and then your data will get flushed early in the process whenever it gets pulled from the bureaus. Yes, I've worked in this field, doing this exact thing. If you don't want your data to be sold by the bureaus, you can request that from the bureaus as well. There are three major ones (Equifax, Experian, Trans Union) and a few hundred thousand small ones (all of whom feed the big three).
You don't actually expect a company to do business with you if they're not allowed to keep records, right? Might I suggest you do some research into how godawful the banking industry was prior to the introduction of the credit bureaus? Think "Good Ole Boys Network" and you'll have a start on it... but it was considerably worse.
I'm not saying that some additional protections on consumer privacy shouldn't be in place (as a bare minimum everyone should be entitled to viewing their own credit report on demand, for no more than cost of mailing or free online). And I'm also not saying that the pendulum hasn't swung too far in the wrong direction (the law a couple years ago allowing companies unprecedented sharing of consumer information went way too far). But anyone who makes statements like that generally has no clue how the financial system, particularly the credit portion of it, actually works.