I wouldn't say that it takes away your choice, necessarily. It's simply that it becomes a habit which is intensely difficult to break, more so than most habits. For instance, I've been a compulsive over-eater most of my life. While I was a teenager, this wasn't a problem. Once my metabolism started slowing down a bit around age 20, I began putting on the pounds, until eventually I grew from 178 lbs to where I was before starting my recent diet, 255 lbs at 6'1" tall.
Those who have been skinny their whole lives will probably have trouble understanding this, too.
The habit crept up on me. I didn't even realize I had it until I was way up, around 240, and despaired about my size and shape. I found comfort in eating, and would console myself by finding something tasty to eat.
Game addiction, or most other addictions, are also similar. You don't realize you have a problem until you're so deep in the problem you don't see a way out. In Everquest, I became despondent about work stresses, and played to tune them out. It was a good rush, and returning to "reality" was a real let-down. I had friends who cared about me in the on-line world, and in real life my wife was constantly nagging me about my playing time, my children demanded my attention and whined a lot, and work was always breathing down my neck.
Realize, now that I'm out of that mode of living, I don't feel this way about things:) Work is simply a fun challenge, my children love me and need nurturing, and, amazingly, if I put forth a little effort to help around the house once or twice a day, my wife doesn't nag me about it! She loves me, too:) Pretty cool.
So I agree with the above poster, mostly. You *can* make the choice not to participate in the addictive behavior, but it causes extreme stress. However, much as with my weight-loss campaign, the only way to succeed in the long term is to deal with the stress and pain, not give in to the habit that made you miserable yet you want at the same time, and just look forward to a day that ends without you having given in. With food, last night I had the overwhelming compulsion to go get more cookies from the jar; instead, I redirected that to pickles. At 6 calories a pickle, they are pretty much crunchy water:) This is much the same recovery process any addict goes through with any substance or other addictive thing. I just hope people don't discount the pain because "it's just a game." The pain is real, the relationships are real, and those who think "game addiction" is simply due to absence of will or desire to work are deluding themselves, terribly misinformed, or deliberately trolling to get a reaction.
I disagree, NineNine. Addiction takes many forms, and it is not necessarily due to someone having "zero self-esteem", "zero drive", or being "intrinsically lazy". Ask any co-worker where I work as a Sr. UNIX Systems Administrator, and they'll let you know that I work hard as the best of them. I've lost 30 pounds in the last 5 months (30 more to go and I'll be back at college weight!), do hard exercise every day to keep in shape, spend time with my family every evening reading stories, playing, and talking, am respected by my coworkers for the quality of my work, and feel I have well-balanced life at the moment.
I'm also a recovering Everquest addict. The habit is under control now, and I play at most 2-4 hours per week; I had quit cold-turkey for about 6 months to be sure I had the habit under control. However, for a long time I would spend many, many hours in front of the console, typing madly and levelling my characters. I lost a job over it. My wife was extraordinarily unhappy about it. My children hardly saw me.
Don't stereotype people by their addiction(s). They are not intrinsically lazy, nor do they lack self-esteem or drive. That's simply *your* perspective on them. Yes, they allow their habits to grow out of control, and often can't quit until they get a "reality check" to remind them of what's really important in life. That's where I'm at now. I'm just glad I didn't let the addiction get to the point where I felt like I could not get out. However, I was close, and only the shock of getting fired from a job for non-productivity got me off the EQ grind and back into reality.
As other have mentioned in these threads, neither the addict nor the non-addicted have a full, balanced perspective on the issue. I personally think that recovered/recovering addicts generally have the most balanced perspective regarding the addiction. But, then again, that's where I'm at, so probably my perspective is skewed. We have enough hatred and bigotry in this world without people adding to it with such loaded terms. A person is a person, with his or her own needs, desires, and motivations. Only that person truly understands her perspective, and speech loaded with vitriol towards someone due to her habits does nothing to improve the situation.
Re:No offense but these were some boring questions
on
William Shatner Replies
·
· Score: 4, Informative
Number nine was from Wil Wheaton, of "Stand By Me" and "Next Generation" fame. Wil has established himself as somewhat of a geek icon, largely from the Slashdot community, because he's plunged headfirst into GNU/Linux and running his own web site.
I think the reason for number nine was because Wil Wheaton sells memorabilia of some of his artwork (I use that term loosely, but some are really cute). One of his pieces is a picture of a name tag that reads like this:
"Hello, My Name Is William Fucking Shatner"
From what I've read on WWDN (wilwheaton.net), Wil was worried that Bill would be offended by the piece, which is also featured on T-shirts sold by Wil. If I recall correctly, Shatner found it funny.
My personal favorite is "Wil's Got A Posse!". I just smile, because it makes me realize that he's just a dude like you and me, who's excited that people like him for what he is and what he's doing now, rather than how they knew him as a child actor. I guess I'm part of the posse, I visit the site about once a week to catch up on what's new in Wil's life:)
Although I'm not diabetic, I completely agree that the Palm is an amazing platform for tracking any day-to-day statistics you need to analyze. I use the Hacker's Diet tools, available at http://www.fourmilab.ch/ . I can put my weight and whether or not I exercised in each morning very quickly, and it automatically takes care of the analysis parts, letting me know where my weight trend is going and what percent of days of the month I've exercised.
For those people that are naturally skinny, this isn't a big deal. For the obese like me, it's an amazing tool to motivate and keep you on the diet & exercise program. I've dropped from 255 lbs to 232 lbs on my trend chart, and I just don't see how I could sustain the loss without the tracking in my Palm. I guess I could use pen and paper, or Microsoft Excel, but how does that compare to sitting on the bus for a few seconds, dropping in the weight I was at this morning, and instantly checking out the results on my trends to calculate my calorie deficit?
For this, and other special-purpose uses, the Palm and others are an excellent resource. I'm on track to get down below 200 lbs by early next year, and without knowing that I *am* consistely dropping the pounds, despite what the scale may read for the day, it's easy to get discouraged.
Yeah, once I'm toting around a Palm plus the full-sized collapsible keyboard, I often wonder if it wouldn't be easier to use my laptop. The biggest benefits for the Palm + full-sized keyboard combination are weight and battery life. I really like the combination, though, because the Palm has zero bootup time, and setting up the keyboard is really fast. It's a bit tough to edit the SGML documentation I maintain (for Bugzilla) on the Palm, though, and for that I'm still stuck using my laptop if I'm away from my desk. Anybody know any really good XML/SGML Docbook-compatible editing environments for the Palm?
I live by Pluckerbooks. I also use it to immortalize a whole bunch of other content. I have a "dailypluck.sh" cron job that runs through the front pages & some of the discussion at Slashdot, freshmeat, some other GNU/Linux news sites, the CNN front page, and other places. Before I plunge into my novel or technical reading on the bus to work in the morning, I catch up on my news sites to know what's new in the world.
After electronic books, I use my palm mostly for scheduling. I have terrible memory for non-technical stuff, and the fact that it beeps at me to remind me of something is a huge help. I also set a lot of daily tasks with reminders so that I can remember to do them. They may sound simple to others, but a reminder to "spend an hour or two on your business Disaster Recovery Plan" leads to the important but long-term projects being completed in time.
My wife also uses her PDA for ebooks, contacts, and calendars. She's caught on to the trick of using the calendar as the to-do list, and really prefers it to the built-in to-do. She's gone from carrying around an enormous, clunky planner with her everywhere, and often jotting down notes on sticky pads to get lost later, to jotting all the notes down in her PDA as a note attachment to the calendar event, and then liberally using the Palm's search function to locate those notes later. It's much easier to find that note you jotted down in your PDA three months later than it is to find that yellow sticky you stuck on a page that later fluttered out.
Everybody's entitled to my opinion. This week, I've read "20,000 Leagues Under the Sea", "The Hacker's Diet: How to lose weight and hair through stress and poor nutrition", P.T. Barnum's "The Art of Money Getting", and I'm partway into reviewing "Perl Programming" and Mark Twain's "Tom Sawyer Abroad". I did all this while riding the bus (an hour each way for work 5 days a week) and during the odd spare moment here and there.
All this on a little Palm M500 I picked up for $125 at the time. Yes, the screen is small. However, it's perfectly adequate for reading electronic books. My only worry is that I'll wear out the "down" button on the front of my M500.
Restricting bandwidth based on MAC address of the client is a trivial exercise, easily done with a GNU/Linux box or any of a number of proprietary "black box" solutions. It would be a simple matter of data entry to key the MAC addresses into the system at the time it is shipped, or even to hire a data entry person to key in a few hundred thousand over the course of a year. If the MAC is not registered, it gets zero bandwidth.
Even more trivial is global restriction of all addresses behind a pair of GNU/Linux based routers to a certain maximum throughput. Set the bucket to a max of, say, 100kbytes/sec, and don't worry about any individual user consuming more than this. If someone figures out a unique hack to get around your restriction, spend time figuring out how they did it and create a deterrent.
I agree with the above poster. If you are an ISP expecting people to observe bandwidth restrictions, you'd darn well be limiting it at the head-end, and not expecting the client end to be secure. If you don't control both ends of the connection, you can never be certain your connection is secure. Even this weakness exists in public-key cryptography. If the person you are sending an encrypted email to doesn't store her messages securely, they are still subject to easy theft despite all your precautions.
Criminally charging and arresting people for modifying a piece of hardware in their house that they own is stupid. If it's leased, it's a slightly different story, and they should be prosecuted for destroying property owned by someone else. But for the love of Pete, calling it theft of service just because they don't have someone on staff with the brains to implement a real solution to their bandwidth problems is laughable.
However, if poorly implemented, from the "internal view", you have two authoritative DNS servers for your domain which are a pain to administer and easy to let get out of sync. Having a split DNS is simply a bad idea, although many companies do it.
There are two suggestions I recommend to avoid this situation: 1. Just don't do it. Set up a third-level domain, such as "internal.mycompany.com", and have an authoritative internal DNS for that third-level domain. Then your machines would be "dick.internal.mycompany.com" and "jane.internal.mycompany.com", but if your search domain is set up right, you can just type dick or jane and get to the box. 2. Set up your own "top-level domain", which is the same name as the second-level domain you own externally. Say your company is "xyzinc", and you own the domain "xyzinc.com". Simply create your internal systems all with the TLD of "xyzinc", so you can get to them at "dick.xyzinc" and "jane.xyzinc".
Both solutions work well, but I'm sure there are others even more effective. But for heaven's sake, don't put yourself through the management headache of having multiple DNS servers which think they are authoritative for the same domain that you have to keep manually updated with different records on each (though I think you can do some monkeying with zone transfers to make it more automated). Although strong from a security standpoint (in the sense that you don't open up internal records to external scrutiny), it's just simply unnecessary if you plan your domain layout correctly.
On the other hand, the tinydns discretion based on requesting IP address seems a very interesting solution if you really, really want internal machines to have your externally-available domain name suffix. Regardless, it seems this approach would be valuable from a security standpoint, even resolving names within the same web server farm.
Darnit, I have mod points, but I have to contribute to the discussion!
Virtualization of computing resources is going very mainstream these days. You have products such as VMWare, competitors for Sun hardware, and even the staunch favorite, User Mode Linux.
I'm running DNS right now in a UML sandbox. Although chroot is an excellent security policy for services, if you want true isolation from the main system in case of break-in, it's hard to beat a UML. There is even a special image provided at the UML home page which runs DNS, and only DNS. It's very handy, and is designed to run while taking only 16 MB of RAM.
Suffice to say, I'm very impressed. For running critical services which, in the past, have required a chrooted environment (such as DNS), user mode linux is a powerful alternative.
Now, would it have had anything to do with helping stop a DOS attack? Nope, but I'm just following the thread here:)
> This may sound mean to the EQ coders, but I can't imagine what they're possibly doing that could require 500MB of RAM if they were coding properly. I mean, do they trickle-load levels at all? Where on earth could that be going?
It doesn't appear to be much of the code bloat, it is instead the high-polygon player and equipment models, enormous textures, motion-captured animation sequences, etc. I think there probably is a lot of room for improvement on how they handle it to reduce the memory usage. Using the new graphics engine in the days prior to the Luclin release, the game could still run on a machine with 64MB RAM easily. At the release of Luclin, resident memory usage spiked to nearly 800MB before they reduced it.
I would say there is some massive room for improvement. However, it has definitely helped fuel hardware upgrades among the gaming population...
> Always thought that NT was a pretty direct child of VMS
The biggest OS/2 relics in NT are the HPFS filesystem, now extended and called "NTFS", and LAN Manager networking (SMB + CIFS). Windows NT 3.0 and 3.1 included native HPFS support; I'm not sure about recent revisions.
There's a link below to a nice history of Microsoft, IBM, and OS/2. Even casual reading can illuminate to one that there is zero, or nearly zero, relationship to be extracted between OS/2's support for MS Windows executables, and GNU/Linux's support of same via Wine/WineX. VMS is very prominent in the development of Windows NT, and IMHO the NT series (NT 3.0, 3.1, 4.0, 2K, XP) are a nicely-engineered, functional piece of work. However, GNU/Linux is more usable for me on a day-to-day basis at this point...
> You know, you generate more pure textual data per time unit than anyone I've ever seen on Slashdot.:-)
Non-geek translation: I write too much:-) I'm thinking of starting my own blog so that I can expound in more detail, and simply provide a link from Slashdot so I don't exposit more than the underpinnings of the argument in the thread of the discussion, saving the reams of ruminations for another forum...
> I'd be more inclined to say that the Linux underpinnings are lightyears ahead of the Windows underpinnings
Again, I concur. However, I try to avoid such language generally because people come to regard it as flamebait for a Microsoft Windows vs. GNU/Linux debate. That said, many Linux kernel abilities (such as true multithreading) are of comparatively recent date... there's definitely some mutual "chasing the taillights" going on.
Interesting to me, however, is that Microsoft Windows NT development was inaugurated in 1987, with release in 1993. Linus Torvalds unleashed the Linux kernel on August 25, 1991. It seems the reality is that GNU/Linux, far from being an "upstart" operating system, predated Microsoft's second 32-bit operating system (OS/2 was their first) by two years. However, it's tough for me to be positive of the data prior to 1995, when I discovered GNU/Linux and started goofing with it from time to time, finally using it seriously and full-time in 1997.
The usual weasel-words apply: the.01 Linux kernel was not yet ready for prime-time, real threading was not introduced until the 2.2 series (with varying definitions of "real"), yadda yadda yadda. Fundamentally, the operating system that "took over the world" was Microsoft Windows 95. It has now been largely abandoned in favor of Windows XP, just released last year. For much of the last seven years, however, GNU/Linux has remained fundamentally the same: a 32-bit UNIX-like kernel using free GNU tools. No painful migrations, like for MS Windows users in 1995, to a 32-bit shell running on a 16-bit operating system.
My personal feeling is, with tools like WineX, we are bringing more and more advantages to the GNU/Linux table. People I know are converting, or at least running a second box to check it out, which would have been almost unheard of outside of geek circles seven years ago. The consensus-based development model for many free software projects may be a very slow process, but it is one that develops an enormous momentum. Eventually, Windows users that are accustomed to making upgrades every 2-4 years will simply make the ultimate upgrade to a GNU/Linux system.
Wow, cripes, I'm lapsing philosophical this morning. And long-winded, as always.
Yep, as I mentioned in another thread, Microsoft had their fingers deep into the vital organs of OS/2, and when they ripped them out there was little IBM could do to stop the hemmorhaging.
I hadn't realized about MS's development FUD campaign. That would explain the lack of killer apps for OS/2 -- thanks!
It definitely is an entirely different ballgame now with GNU/Linux and Wine rather than IBM's OS/2. My goal was to fix the mistaken impression of many posters that the reason OS/2 failed was because it emulated MS Windows. While it may have been a teeny, tiny factor in the overall picture, the fact is that Windows NT is the illegitimate stepchild of OS/2 and VMS. Microsoft divorced IBM, left him with all the bills, refused to pay child support, and hired professional druggists to attempt to poison the child's prescription.
So I think that Crossover/WineX are a very good thing, ultimately leading people to software freedom. It will be a long road, but it's a better "ecological model" (to use Linus' verbiage) than commercial software, and in the long run it will kill commodity software off.
> What, precisely, is it that you want "integrated"?
At this point, I want just a few, simple things that I take for granted in a business accounting package: * Automated invoicing. This means that I enter the invoice information, and the accounts receivable information is automatically entered into my books. Also, that I can print out the invoice from the accounting package, and record that I printed it. Envelope labelling based on invoice information, and the ability to get the address information when I click on the accounts receivable entry would be a nice bonus. * Check printing * Payroll "understanding". I mean, enter the tax information for the employee one time, and from that point on when you verify gross pay the taxes will already be resolved.
These three things would make a business accounting package usable for me. At this point, GNUCash is a nice personal finance utility, but fairly inadequate for business usage. If it annoys me enough, eventually I may get around to writing some modules for it, but my business has been shelved temporarily (due to the economic climate) while I do a regular, full-time gig to pay the bills:)
>native, though less capable (i.e. no usable alternative shell) UnxUtils
I've used unxutils as well. They are OK. For routine, automated systems administration, though, they are simply maddening! I'm not terribly concerned about the latest software rollout working "fast", so Cygwin's slowness is irrelevant. I just want the software update rolled out across some 300 NT/2K workstations at some point over the course of the day.
However, I've tried to use cygwin to run KDE before (in hopes that I could somehow have an "integrated" platform to work from, using the GNU/Linux interface with which I'm familiar, running atop Win32), and it definitely is incredibly slow for some apps. Wine is lightyears ahead of Cygwin, performance-wise, IMHO. Of course, we're comparing apples and beach balls, here, so it's entirely my subjective opinion.
The model limitation is related to the physical memory of your machine, not the video RAM. For whatever reason, under WineX my 512MB RAM shows up as 504 to Everquest (perhaps due to AGP texturing, I'm not entirely sure), and it only recommends I enable up to 20 models with that limitation on RAM. It seems to eat a great deal more RAM under WineX at the moment than under Win32, that's for sure.
[blockquote] NT has OS/2 roots? [/blockquote]
Sorry to go offtopic, but why not:)
Major portions of the operating system have roots in both Digital's VMS operating system and IBM's OS/2. Microsoft and IBM began work together on OS/2 in April of 1987. By October of 1988, Microsoft had recruited David Cutler, of Digital fame, as the project lead for NT. Up until 1990, the NT "personality", or UI, was wholly OS/2; while the "core" of the operating system was VMS, the API and user interface were very, very much OS/2-ish. Later that year, Microsoft leaked plans to fork the OS/2 API into the "Win32" API. Microsoft was slated to be the developer for OS/2 3.0, but decided it would be more profitable to release a 32-bit operating system that was wholly their own rather than continuing to rely on IBM's OS/2 license.
In a nutshell, OS/2 was killed by Microsoft because they abandoned development on the work to develop their own. Note that OS/2 had a SIX-YEAR lead on Microsoft Windows NT, and still couldn't beat it. They weren't even the first movers in the 32-bit arena. VMS had been released on VAX in 1978. The reason it was so easy for OS/2 to run Windows applications was because the Win32 API, at the time of the Windows NT 3.0 release, was barely a minor fork from the OS/2 API. If I recall correctly, IBM licensed the Windows 3.1 emulation from MS as well, but I could be wrong there. However, they still had a problem running Windows 95 executables due to the whole 32-bit yet running on DOS thing. By the time their emulation was up to snuff, Microsoft already had had enormous market penetration.
Back to the topic at hand, the WineX/Crossover efforts are a completely different ballpark than IBM's emulation. Microsoft Windows is in much the same situation now as IBM was in 1993 after the release of NT: there's a new operating system on the block which is very cool, very sexy, and very incompatible with our system. However, it's gaining market share at an incredible rate, and there's not much we can do about it. The key difference here is that GNU/Linux is not in bed with Microsoft, dependent upon the very company it is competing against for continued development.
OS/2 died because Microsoft had its fingers buried deep into the vital organs the whole time. When they ripped them out, all of IBM's efforts to stop the hemmorhaging failed rather predictably. Windows Emulation (or, if you are a Wine zealot, "implementation") on GNU/Linux is a completely different situation.
It's a compelling argument you make. My office is currently investigating a Microsoft Outlook deployment. I eagerly searched for open-source/free software alternatives that achieve the same integrated functionality, and came up lacking.
It's better to use free software on a non-free operating system than to use no free software at all, but it's worthwhile to examine the possibility that the Cygwin project, among others, are making it easier for people to stay with Microsoft Windows than use GNU/Linux, just as it's worthwhile to wonder whether Wine and WineX are are keeping more people tied to Windows as well.
My personal experience? Administrators using Cygwin become very impressed with the power of the *nix-like environment. They begin using it on a day-to-day basis, and see no reason to stop using Microsoft Windows on their desktops.
Until the day they run up against the huge limitations of using Cygwin, be it the performance penalty, lack of target platform support in many popular software packages, or simply any one of a string of painful interoperability problems. They then install GNU/Linux on a second partition or a second machine, and experience the full power of the operating system, and amazing graphical user interfaces which have been blazing trails for the last 2 years that Microsoft has attempted to follow. In turn, the best innovations of the MS desktop have been integrated into these free projects, as well.
So I'd have to say that because of the enormous minority situation GNU/Linux is in, Cygwin and running free software applications on a proprietary operating system like Microsoft Windows is an excellent "Gateway Drug" for power users, and eventually lesser users. Likewise, being able to run non-free software on a free system eventually leads to the replacement of that non-free piece by those who value software freedom over the convenience or entertainment value of the non-free software.
From my point of view, all roads lead to the ultimate "killer app" of the GNU/Linux world: the General Public License and the open culture it forces on otherwise selfish individuals and corporations. Yep, it's viral, and those who wish to take away software freedom hate it. Heck, a lot of freedom-loving individuals hate it too because it infringes on *their* individual freedom by requiring them to contribute their code back to the common pool. Love it or hate it, it's the reason why people are migrating to GNU/Linux from proprietary systems (of all types) in unheard-of numbers.
This is a difficult question for anyone to answer. You see, Sony Online Entertainment (SoE) puts people in an uncomfortable position: If you admit to having used ShowEQ, you can get your account banned. For those for whom the game is an important part of their life, this would be a horrible blow and they should consider very carefully whether they wish to acknowledge such use even on a semi-anonymous medium such as Slashdot.
That said, it is not illegal to run ShowEQ. It is, however, definitely a violation of Sony's End User License Agreement to which one affirmatively agrees each time one plays the game. If you are comfortable with the potential consequences for violating the "contract" you have entered into with Sony Online Entertainment, or else confident enough in your discretion that you don't believe they will find out you have abrogated your agreement, read on.
Yes, ShowEQ works with EQ under Transgaming WineX. You have to know what you are doing in order for it to work, however. Check out their message boards at http://seq.sourceforge.net for details.
In the interest of full disclosure, and since I really don't give a rat's third left whisker if SoE cancels my account or not, I have used ShowEQ. I am currently using Transgaming WineX. However, I'm not using ShowEQ anymore, least of all alonside EQ under WineX. It just seemed to take a lot of the fun out of the game to have all the monster locations disclosed. The stat-tracking stuff is pretty cool, though, but you can get similar functionality using a non-agreement-infringing log parser. While I wouldn't discourage curious hackers from checking out what ShowEQ can do, I would encourage them also to avoid using it in day-to-day play. It tends to lead to either boredom or conflict.
Umm, that said, if any SoE employee reads this, I'd rather not have the account terminated if it can be avoided:) I just shelled out $35 today, after a five-month hiatus, for a 3-month account because it's supported under Transgaming WineX. However, the remedies for being caught using infringing software are quite clear that one forfeits the subscription and monies paid to SoE and cannot use the service any longer. I can live with that penalty if I decided to use ShowEQ again.
The question, before you plunge into the world of ShowEQ, is:
Can you live with the consequences of getting caught?
I say, check it out, and write up your experiences at seq.sourceforge.net. Hell, they even have an Electronic Frontier Foundation donation account set up so you can get a warm fuzzy supporting software freedom while supporting EULA-infringing software. To top it all off, you don't even need the binary package release of Transgaming's WineX to have easy installation of WineX. Gentoo has winex as part of their portage system, and their are binary builds of the CVS checkout available on the net. Those work perfectly well for playing EQ. The only reason Transgaming has a binary-only release for subscribers is because they have licensese for certain copy protection schemes which require them not to release the source for that particular application of the suite.
I concur. I have been playing Everquest for several hours today, and full-screen (as on Windows by default) it has been rock-solid as far as stability is concerned. The memory usage seems significantly higher than under Windows at the moment, so that I cannot load all the models with 512MB RAM, but otherwise it's quite decent.
Things that aren't perfect regarding Everquest under x86 GNU/Linux at the moment: * The installer does not give you any indication of progress until it is ready for you to change CDs or it is done installing. * I experienced one error installing Shadows of Luclin, regarding some texture expansion.exe being unable to run. However, the game still ran fine. * The patcher takes longer than it does on Windows * The patcher (at least on my install) segfaults instead of launching Everquest. The workaround is to invoke Everquest with "winex eqgame.exe patchme". * Alt-Tab results in some goofiness under KDE. Probably just due to my window manager, but alt-tabbing out, and then clicking or alt-tabbing back requires that I hit ctrl-alt in order to have any keyboard input. Otherwise the game seems to think I always have Alt depressed. Note this doesn't seem to happen when running it fullscreen. * Alt-clicking items to paste them as a clickable link in chat doesn't seem to work; my window manager grabs the alt and tries to drag the window. Yet another problem that doesn't show up in full-screen mode. * Performance at the moment is significantly slower than under Win32. It's not horrible, but it is noticeable. * Luclin player models have some visual glitches. Due to the age of my hardware (Geforce2MX, 933MHz P3, 512MB PC133 RAM) I've rarely played with them turned on anyway, but I did so briefly today and noted that most characters appeared at least partially nude. * If you use more Luclin models than recommended by the Options dialog, it's possible EQ may not load at all. It effectively grinds to a halt if the resident memory usage displayed by "top" comes anywhere near the physical memory limitation of your machine. With all models turned on, by the time I got to the 1-frame-per-second character selection screen, EQ had some 400+ MB resident memory requirement, and a total RAM requirement of 820MB...
I haven't had Microsoft Windows installed on my gaming PC in several months, and renewed my EQ account just to check out the support under WineX. This is a nice boon, although I'm certain my wife won't be as happy as I am that EQ runs under GNU/Linux just fine now. Darnit, my Armagetron, BZFlag, and Chromium binges are going to be curtailed I think.
All in all, I'd say it's a fairly successful initial offering to appease the rabid EQ supporters among the Transgaming subscribers. It's still "unsupported" and has some rough edges, but the glitches that are present are certainly not show-stoppers.
Regarding the earlier comments about how by using Wine, one is treading sand that swallowed up IBM's OS/2: OS/2 failed for a myriad of reasons, not nearly solely due to support for Microsoft Windows programs. Although many believe that was a factor, I do not believe that it had any major impact. There were many things leading to the downfall of OS/2, not the least of which was terrible marketing on the part of IBM, the lack of any "Killer App" for the platform, and the lack of significant bundling deals with OEM's which meant that OS/2 had to be installed by the user (a task then, as now, beyond the interest of the majority of computer users). Indeed, GNU/Linux shares some of the same problems right now, regardless of Win32 exe support.
However, from my point of view, GNU/Linux is in a uniquely superior position compared to OS/2, and the ability to run Win32 packages natively has nothing to do with it. The secret weapon is the GNU General Public License. Business FUD regarding the platform is ineffective in the long run, because the operating system is guaranteed to live on, barring unforeseen litigation. Nobody can fork off their own version of the OS (Read: Microsoft forking Win NT from the OS/2 effort) because the GPL requires those changes to be given back to the community if they are released in binary format.
We're in a completely different battlefield today than the one that sunk the OS/2 Titanic. GNU/Linux's weaknesses are becoming its strengths. I agree that the Wine approach may not be effective at winning over large numbers of MS Windows users to GNU/Linux. However, that's not really the point, IMHO.
See, I wiped out MS Windows because I value my software freedom.
I don't find playing proprietary games under GNU/Linux to be an abrogation of these principles. Games are content-driven (in that new content drives sales), while productivity applications are procedure-driven, and any procedure can be automated and commoditized. Basically, IMHO games are art, and making the concession that the publisher must get back the production cost of the art before the programmatic portions can be released under the GPL seems, to me, to be sound philosophy which acknowledges business needs while supporting software freedom.
Because I value software freedom, and am far more productive under GNU/Linux and KDE than Microsoft's operating system, I replaced Microsoft Windows with the commoditized, free software GNU/Linux. Many others are making this same step. Transgaming's WineX supports these subscribers by providing them with a service unobtainable in an exclusive free-software realm. I value the service, and contribute my $5 a month to their continued success.
I feel the need to comment on the disconnect many Slashdot posters seem to feel. They rail that "Free Software Zealots" seem to change their minds with the topic at hand. The reality is, if you watch those who post, some users feel passionately about a certain topic, and others don't. I've found precious few comments amongst several topics that would cause me to believe the poster had an internal disconnect regarding their point of view on free software.
I have one position I have held for at least the last three years (while using GNU/Linux for years before that for simply practical reasons), and it is quite similar to that which Richard Stallman has claimed many, many times (in different words): The point of the Free Software Foundation is to preserve freedom. In some cases, when there is no adequate free alternative, proprietary software can be used. It is better that a user have a free tool on a proprietary operating system, and thus begin to understand the importance of software freedom, than to have no freedom-supporting software at all. However, if use of the proprietary system can be avoided, it should.
Transgaming provides a valuable commodity: the ability to play popular, but still proprietary, gaming titles on a free operating system. It's a bit the reverse of the FSF's usual line. Games, however, are a form of art, and stand uniquely apart from that which can be easily commoditized into a free system.
As a side note, as far as I have found there is no *good* current free software alternative to a proprietary package such as Quickbooks Pro. I've tried MoneyDance. I've tried GNUCash. If you know of more that fit the bill, please educate me. In the meantime, products such as Transgaming's WineX, and Codeweaver's Crossover Office allow me to use proprietary tools such as this, yet keep my system otherwise as free as I possibly can. There will come a day when the free alternatives do the job "well enough", and that will be the day I convert to a free software financial package. This is neither a contradictory position, nor one that is out of line with the "hard line" of the FSF. It's simply an acknowledgement of reality.
So please, those who've railed against the Slashdot community's collective Jekyll/Hyde syndrome, realize that different people post to different threads. There's a significant contingent of chronic copyright infringers on Slashdot. There is also a significant contingent of those who would never break copyright law, even if they think it is bad. There are those who are free software zealots, and refuse to use non-free software. There are those who only use proprietary software, and snub their nose at the attempts to replace sophisticated proprietary packages.
And there are those in between, that hold both the ideals of the Free Software Foundation quite dear, but also acknowledge that idealism and reality must reach a compromise on our desktops if we are to use these fantastic tools to their ultimate advantage.
That's me. Let's not debate over whether this is a horrible move for the community or not. There are at least 3,000 Transgaming subscribers that believe in WineX, and pay their $5 or more a month to have a voice in the direction of the software. Those of us who do are grateful for the service, and happy to appease our free-software-loving sides because those portions of the WineX suite which can be free software (and not encumbered by licensing) ARE kept free by Transgaming.
Then again, this is Slashdot. Here we debate over whether or not to have a debate:)
In most organizations, the NOC staff, CIO/CTO/whatchacallit, in-house "tool" programmers (those who write programs which are not a profit center), IT managers, printer monkeys, etc. are all considered "support staff" for machines. If you have a 24/7 rotation of a single NOC on-staff, to be able to handle vacations and such on a 4-on, 3-off rotation, you need at least six guys. Preferably seven, in case two people are sick or take vacation near the same time.
I'm not trying to straw-man you, but most people consider the total number of people supporting machines to count toward the mystical "user to admin ratio". For instance, back in a 90-person, 100+ machine organization several years ago, we had two support people. We took care of all the desktop support and kept the servers running at about a 50-to-1 ratio. The "unseen cost" was the fact that the programmers often handled many basic systems administration tasks themselves. At my current job, in a 400+ person organization with, admittedly, much more severe development needs (banking industry, lots of development to integrate vital services, rather than "run the file servers"), we have a total of 12 support people:
1. Chief Information Officer. 2. Manager of Systems Administration. 3. Programming lead. 4. Java & Windows programmer. 5. Java & *nix programmer. 6. UNIX Admin (-- me). 7. Telephony admin (we're mostly a call center, so this is a HUGE job). 8,9,10: Systems Administrators. 11,12: "Night Operators" who swap backup tapes, handle printed reports, act as security, move stuff around we don't want to move during the day, etc.
If you wanted to stretch it, you could say that we have four "sysadmins" for 400 people. Woot, 100-to-1 support ratio! Reality is quite different. I personally think our organization will have some pruning in the near future, and am working my butt off now to bring our system up to speed with automation issues so we don't have to spend all our time putting out fires when the RIF comes. Realistically, though, we have close to a 40-to-1 machine/admin ratio (including servers, workstations, etc.) For many organizations, if you are not 100-to-1 including *all* your support staff and hours spent by other people doing sysadmin tasks, you're not where you need to be and will have a RIF once management figures out where the costs are distributed.
Magnificently automated processes are the stuff that makes high admin/user ratios work. No bones about it. But if you're going to count it, make sure you count the full cost of systems administration, including the people you don't think are sysadmins. They are still support.
Important note: support ratios are, and IMHO will always be, majorly out-of-whack for software development firms. The needs of a software developer are so different from that of a secretary they simply can't be easily compared. About the only choice there is to distribute a great deal of the administration load among those who work on their machines. There are exceptions, and modes of development that minimize this problem (dedicated sandboxes, shared compile farms, etc.), but overall you'll generally have a lower admin/user ratio in software development.
I hate to add the obligatory GNU/Linux reference to your post, but here goes anyway:)
You already see this future in devices such as the Agenda VR3 PDA. Although the company itself has disappeared in the U.S., the user community has applied some interesting technologies to solving the problem of very little storage, very little RAM, and a relatively slow CPU.
The Agenda has 8 MB of working RAM. The data storage for the unit is 16MB of flash RAM. On initial release, the Agenda was very slow and problematical, primarily because it would eat up its 8 MB of RAM with the kernel + X Windows, leaving precious little for applications. The solution was ingenious: combine compressed storage on the flash (to conserve that 16MB of flash storage) with "XIP", or "eXecute In Place" technology in the Linux kernel. XIP prevents one from storing the file to be XIP'd uncompressed, but allows one to actually run the program directly from the storage media, without loading much (or any) of it into RAM. With hard disks, this would be unbearably slow. With flash RAM, while considerably slower than regular RAM, it works pretty well.
I'm actually considering implementing an XIP solution for our servers at work. I'm trying to aggregate about 400 employees on some servers, but the problem is that if each one of them is running off the server, the RAM overhead of KDE + apps for each person is abominable. If I had a common, solid-state-storage (or else a very large RAMdisk), combined with XIP, it may be possible to dramatically reduce the memory overhead of launching hundreds of copies of the same app. XIP it on a RAMdisk!
Of course, it may not work. As a matter of fact, I'm betting that it won't work. But wouldn't it be cool if it did?
Elimination of magnetic storage will probably not be complete during my lifetime (I'm 29 now). However, I consider removing the idea that we store on one media, and run from another, to be beneficial to the industry as a whole. Faster media, programs that execute directly from media rather than intermediately storing in RAM. Woot!
I don't post much, but when I do, I like to be right, or at least have an informed opinion. Thanks for your clarification! I hadn't thought about the fact fraud was happening as well.
Your argument, Squarewav, holds water like a sieve. I apologize in advance for correcting grammar, punctuation, and spelling from your initial posting.
If you recorded the Simpsons, removed the commercials, put in your own ads, and rebroadcast it, is it theft?
The answer is a resounding NO. If you rebroadcast their material, substituting your own ads, you have profited by infringing copyright. Motive in copyright infringment cases is very, very important. If you do not profit from infringement, the violation is not as serious as if you gain profit. Additionally, profitting from copyright infringement leaves you liable for damages equal to or greater than the profit you gained by infringing someone else's copyright.
It's not theft. It is copyright infringement. They are two dramatically different things, although the major software and media companies would have you believe otherwise.
[Let's assume] you owned a furniture store and wanted to sell Lazyboy chairs, but you didn't want to pay for them. So you made your own chair, that looks exactly the same, and then put a lazyboy logo on it. [You then] sold it for next to nothing; is [this] theft?
Again, you are incorrect. If you produce goods similar to, or in many cases as identical to (in any case where patent, trademark, or copyright do not apply), someone else's and place your own trademark upon it, you have done nothing wrong. If, however, you place another's trademark upon it (implying that it was produced by the other manufacturer), you are guilty of trademark infringement. Alternatively, if you use a patented invention and do not pay patents to the patentor, you are liable for patent infringement claims. Similarly, if you duplicate a copyright work, you are not guilty of stealing from the author; you are guilty of infringing upon his right to control copying of his work.
The only reasons one could proffer the arguments above, that I can see, is 1) simply lack of education regarding U.S. law. I am not a lawyer, but I do believe I have a sound understanding of laws where they affect my day-to-day life. There are also 2) those paid to have that viewpoint. Hilary Rosen and others are paid part to promote these views of copyright infringement as theft. To promote an alternative view disagrees with the corporate agenda, and this disagreement would most likely eliminate their sources of income.
The fundamental problem with the thinking comes about because of the nature of what we're dealing with. Information is trivially reproduced, even when spoken. I suggest you study the history of copyright, to fully understand the nature of the laws. Today, we have a society where such information can be reproduced for (effectively) free. It's my personal opinion that Copyright is a doomed concept. However, we have not come up with a suitable reward yet for authorship to promote the science and arts that is not Copyright. Until we do, we will be stuck with this system that so obviously maps so poorly to reality.
There are certainly cases where the line between copyright infringement and theft is very blurred. For instance, if one breaks into a computer system and makes copies of information that were never intended to be made public. One has obviously violated copyright in that case, since U.S. law regards all authorship as copyrighted. Is it theft? In that case, I don't know; just as "breaking and entering" is considered "breaking and entering" (vandalism and trespass, if you prefer), if you don't steal anything for entering, but instead copy important documents, you've not stolen the documents, but made copies in violation of the wishes and reasonable expectation of the holder. The company or individual never intended to release the information to the public for profit (the point of copyright), the information was reasonably expected to remain private, and consent for this action was implicitly denied. In that regard, information violation seems more analogous to rape than theft: one has expressly violated the wishes of the holder of the information, taken nothing from them, but used them in a way inconsistent with their will. Copyright infringement on released goods, however, is similar to using a hooker for her intended purpose, but refusing to pay her. One has no implied contract, the other does. The penalties for rape are spelled out in the law, and include government-sanctioned prison time. The penalties for not paying your prostitute are the same as for not paying any service person: if your bill is not paid, you are sent to a collection agency, which then may take you to court to seek damages. It is (often) not treated the same as theft, since the "goods" (a service) are intangible, you have not deprived anyone of anything except time invested (which has value, but is again intangible and cannot be stolen) and potential profits. In some cases, particularly where the one infringed upon believes the intent was to defraud (once again, fraud law, not theft), they may seek criminal remedies. Most don't, though, because by so doing they are depriving themselves of a potential customer, getting bad press, and preventing the infringer from quickly paying the damages by depriving him/her of income.
Note that the paragraph above is entirely my opinion, and not really part of my initial refutation. I simply think that most software companies and authors would do well to remember that they simply sell their time for money. Their "product" is a service, and our current model of copyright attempts to treat information as a tangible good, which it is not. Those prepared to acknowledge this fact (as Microsoft seems to be doing with their license renewal services) will probably do OK as the economy transforms to take advantage of new realities. Those who insist on treating intangible as tangible will eventually go out of business as realists (the customers) begin to treat it as the intangible, inherently value-less thing it is.
Fundamentally, it's a thorny issue with a whole lot of ramifications. Those who attempt to cast copyright infringement as a black and white case of theft are intentionally misleading you as to what is going on. It is its own legal domain with its own remedies and penalties, entirely aside from traditional laws regarding property theft. There is no such thing as "intellectual property": there are patents, trademarks, and copyrights which give rights to works based upon a well-established but sorely broken legal framework.
Although you are factually correct, you unintentionally mislead users by the fact you omit. This is an Apache exploit which only occurs on GNU/Linux x86 systems. The code, IIRC, just doesn't compile anywhere else. If the one devising the exploit was a more competent programmer, it could have been a cross-platform Apache exploit, but the reality is that right now, you have to be running a specific GNU/Linux-Apache-mod_ssl setup in order to be infected. In that sense, then, it's a "Linux worm", because no other platforms will get nailed by it.
Caveat: Attach "at the moment" to the end of that last sentence. I fully expect someone to grab the source and modify it to run cross-platform within a few weeks.
Slashdot Mirror
I wouldn't say that it takes away your choice, necessarily. It's simply that it becomes a habit which is intensely difficult to break, more so than most habits. For instance, I've been a compulsive over-eater most of my life. While I was a teenager, this wasn't a problem. Once my metabolism started slowing down a bit around age 20, I began putting on the pounds, until eventually I grew from 178 lbs to where I was before starting my recent diet, 255 lbs at 6'1" tall.
:) Work is simply a fun challenge, my children love me and need nurturing, and, amazingly, if I put forth a little effort to help around the house once or twice a day, my wife doesn't nag me about it! She loves me, too :) Pretty cool.
:) This is much the same recovery process any addict goes through with any substance or other addictive thing. I just hope people don't discount the pain because "it's just a game." The pain is real, the relationships are real, and those who think "game addiction" is simply due to absence of will or desire to work are deluding themselves, terribly misinformed, or deliberately trolling to get a reaction.
Those who have been skinny their whole lives will probably have trouble understanding this, too.
The habit crept up on me. I didn't even realize I had it until I was way up, around 240, and despaired about my size and shape. I found comfort in eating, and would console myself by finding something tasty to eat.
Game addiction, or most other addictions, are also similar. You don't realize you have a problem until you're so deep in the problem you don't see a way out. In Everquest, I became despondent about work stresses, and played to tune them out. It was a good rush, and returning to "reality" was a real let-down. I had friends who cared about me in the on-line world, and in real life my wife was constantly nagging me about my playing time, my children demanded my attention and whined a lot, and work was always breathing down my neck.
Realize, now that I'm out of that mode of living, I don't feel this way about things
So I agree with the above poster, mostly. You *can* make the choice not to participate in the addictive behavior, but it causes extreme stress. However, much as with my weight-loss campaign, the only way to succeed in the long term is to deal with the stress and pain, not give in to the habit that made you miserable yet you want at the same time, and just look forward to a day that ends without you having given in. With food, last night I had the overwhelming compulsion to go get more cookies from the jar; instead, I redirected that to pickles. At 6 calories a pickle, they are pretty much crunchy water
I disagree, NineNine. Addiction takes many forms, and it is not necessarily due to someone having "zero self-esteem", "zero drive", or being "intrinsically lazy". Ask any co-worker where I work as a Sr. UNIX Systems Administrator, and they'll let you know that I work hard as the best of them. I've lost 30 pounds in the last 5 months (30 more to go and I'll be back at college weight!), do hard exercise every day to keep in shape, spend time with my family every evening reading stories, playing, and talking, am respected by my coworkers for the quality of my work, and feel I have well-balanced life at the moment.
I'm also a recovering Everquest addict. The habit is under control now, and I play at most 2-4 hours per week; I had quit cold-turkey for about 6 months to be sure I had the habit under control. However, for a long time I would spend many, many hours in front of the console, typing madly and levelling my characters. I lost a job over it. My wife was extraordinarily unhappy about it. My children hardly saw me.
Don't stereotype people by their addiction(s). They are not intrinsically lazy, nor do they lack self-esteem or drive. That's simply *your* perspective on them. Yes, they allow their habits to grow out of control, and often can't quit until they get a "reality check" to remind them of what's really important in life. That's where I'm at now. I'm just glad I didn't let the addiction get to the point where I felt like I could not get out. However, I was close, and only the shock of getting fired from a job for non-productivity got me off the EQ grind and back into reality.
As other have mentioned in these threads, neither the addict nor the non-addicted have a full, balanced perspective on the issue. I personally think that recovered/recovering addicts generally have the most balanced perspective regarding the addiction. But, then again, that's where I'm at, so probably my perspective is skewed. We have enough hatred and bigotry in this world without people adding to it with such loaded terms. A person is a person, with his or her own needs, desires, and motivations. Only that person truly understands her perspective, and speech loaded with vitriol towards someone due to her habits does nothing to improve the situation.
YAUP: Yet Another Useless Phrase
Mechanical engineers build bombs.
Civil Engineers build targets.
Maybe that's the upgrade plan?
Number nine was from Wil Wheaton, of "Stand By Me" and "Next Generation" fame. Wil has established himself as somewhat of a geek icon, largely from the Slashdot community, because he's plunged headfirst into GNU/Linux and running his own web site.
:)
I think the reason for number nine was because Wil Wheaton sells memorabilia of some of his artwork (I use that term loosely, but some are really cute). One of his pieces is a picture of a name tag that reads like this:
"Hello, My Name Is
William Fucking Shatner"
From what I've read on WWDN (wilwheaton.net), Wil was worried that Bill would be offended by the piece, which is also featured on T-shirts sold by Wil. If I recall correctly, Shatner found it funny.
My personal favorite is "Wil's Got A Posse!". I just smile, because it makes me realize that he's just a dude like you and me, who's excited that people like him for what he is and what he's doing now, rather than how they knew him as a child actor. I guess I'm part of the posse, I visit the site about once a week to catch up on what's new in Wil's life
Although I'm not diabetic, I completely agree that the Palm is an amazing platform for tracking any day-to-day statistics you need to analyze. I use the Hacker's Diet tools, available at http://www.fourmilab.ch/ . I can put my weight and whether or not I exercised in each morning very quickly, and it automatically takes care of the analysis parts, letting me know where my weight trend is going and what percent of days of the month I've exercised.
For those people that are naturally skinny, this isn't a big deal. For the obese like me, it's an amazing tool to motivate and keep you on the diet & exercise program. I've dropped from 255 lbs to 232 lbs on my trend chart, and I just don't see how I could sustain the loss without the tracking in my Palm. I guess I could use pen and paper, or Microsoft Excel, but how does that compare to sitting on the bus for a few seconds, dropping in the weight I was at this morning, and instantly checking out the results on my trends to calculate my calorie deficit?
For this, and other special-purpose uses, the Palm and others are an excellent resource. I'm on track to get down below 200 lbs by early next year, and without knowing that I *am* consistely dropping the pounds, despite what the scale may read for the day, it's easy to get discouraged.
Yeah, once I'm toting around a Palm plus the full-sized collapsible keyboard, I often wonder if it wouldn't be easier to use my laptop. The biggest benefits for the Palm + full-sized keyboard combination are weight and battery life. I really like the combination, though, because the Palm has zero bootup time, and setting up the keyboard is really fast. It's a bit tough to edit the SGML documentation I maintain (for Bugzilla) on the Palm, though, and for that I'm still stuck using my laptop if I'm away from my desk. Anybody know any really good XML/SGML Docbook-compatible editing environments for the Palm?
I live by Pluckerbooks. I also use it to immortalize a whole bunch of other content. I have a "dailypluck.sh" cron job that runs through the front pages & some of the discussion at Slashdot, freshmeat, some other GNU/Linux news sites, the CNN front page, and other places. Before I plunge into my novel or technical reading on the bus to work in the morning, I catch up on my news sites to know what's new in the world.
After electronic books, I use my palm mostly for scheduling. I have terrible memory for non-technical stuff, and the fact that it beeps at me to remind me of something is a huge help. I also set a lot of daily tasks with reminders so that I can remember to do them. They may sound simple to others, but a reminder to "spend an hour or two on your business Disaster Recovery Plan" leads to the important but long-term projects being completed in time.
My wife also uses her PDA for ebooks, contacts, and calendars. She's caught on to the trick of using the calendar as the to-do list, and really prefers it to the built-in to-do. She's gone from carrying around an enormous, clunky planner with her everywhere, and often jotting down notes on sticky pads to get lost later, to jotting all the notes down in her PDA as a note attachment to the calendar event, and then liberally using the Palm's search function to locate those notes later. It's much easier to find that note you jotted down in your PDA three months later than it is to find that yellow sticky you stuck on a page that later fluttered out.
Everybody's entitled to my opinion. This week, I've read "20,000 Leagues Under the Sea", "The Hacker's Diet: How to lose weight and hair through stress and poor nutrition", P.T. Barnum's "The Art of Money Getting", and I'm partway into reviewing "Perl Programming" and Mark Twain's "Tom Sawyer Abroad". I did all this while riding the bus (an hour each way for work 5 days a week) and during the odd spare moment here and there.
All this on a little Palm M500 I picked up for $125 at the time. Yes, the screen is small. However, it's perfectly adequate for reading electronic books. My only worry is that I'll wear out the "down" button on the front of my M500.
Restricting bandwidth based on MAC address of the client is a trivial exercise, easily done with a GNU/Linux box or any of a number of proprietary "black box" solutions. It would be a simple matter of data entry to key the MAC addresses into the system at the time it is shipped, or even to hire a data entry person to key in a few hundred thousand over the course of a year. If the MAC is not registered, it gets zero bandwidth.
Even more trivial is global restriction of all addresses behind a pair of GNU/Linux based routers to a certain maximum throughput. Set the bucket to a max of, say, 100kbytes/sec, and don't worry about any individual user consuming more than this. If someone figures out a unique hack to get around your restriction, spend time figuring out how they did it and create a deterrent.
I agree with the above poster. If you are an ISP expecting people to observe bandwidth restrictions, you'd darn well be limiting it at the head-end, and not expecting the client end to be secure. If you don't control both ends of the connection, you can never be certain your connection is secure. Even this weakness exists in public-key cryptography. If the person you are sending an encrypted email to doesn't store her messages securely, they are still subject to easy theft despite all your precautions.
Criminally charging and arresting people for modifying a piece of hardware in their house that they own is stupid. If it's leased, it's a slightly different story, and they should be prosecuted for destroying property owned by someone else. But for the love of Pete, calling it theft of service just because they don't have someone on staff with the brains to implement a real solution to their bandwidth problems is laughable.
However, if poorly implemented, from the "internal view", you have two authoritative DNS servers for your domain which are a pain to administer and easy to let get out of sync. Having a split DNS is simply a bad idea, although many companies do it.
There are two suggestions I recommend to avoid this situation:
1. Just don't do it. Set up a third-level domain, such as "internal.mycompany.com", and have an authoritative internal DNS for that third-level domain. Then your machines would be "dick.internal.mycompany.com" and "jane.internal.mycompany.com", but if your search domain is set up right, you can just type dick or jane and get to the box.
2. Set up your own "top-level domain", which is the same name as the second-level domain you own externally. Say your company is "xyzinc", and you own the domain "xyzinc.com". Simply create your internal systems all with the TLD of "xyzinc", so you can get to them at "dick.xyzinc" and "jane.xyzinc".
Both solutions work well, but I'm sure there are others even more effective. But for heaven's sake, don't put yourself through the management headache of having multiple DNS servers which think they are authoritative for the same domain that you have to keep manually updated with different records on each (though I think you can do some monkeying with zone transfers to make it more automated). Although strong from a security standpoint (in the sense that you don't open up internal records to external scrutiny), it's just simply unnecessary if you plan your domain layout correctly.
On the other hand, the tinydns discretion based on requesting IP address seems a very interesting solution if you really, really want internal machines to have your externally-available domain name suffix. Regardless, it seems this approach would be valuable from a security standpoint, even resolving names within the same web server farm.
Darnit, I have mod points, but I have to contribute to the discussion!
:)
Virtualization of computing resources is going very mainstream these days. You have products such as VMWare, competitors for Sun hardware, and even the staunch favorite, User Mode Linux.
I'm running DNS right now in a UML sandbox. Although chroot is an excellent security policy for services, if you want true isolation from the main system in case of break-in, it's hard to beat a UML. There is even a special image provided at the UML home page which runs DNS, and only DNS. It's very handy, and is designed to run while taking only 16 MB of RAM.
Suffice to say, I'm very impressed. For running critical services which, in the past, have required a chrooted environment (such as DNS), user mode linux is a powerful alternative.
Now, would it have had anything to do with helping stop a DOS attack? Nope, but I'm just following the thread here
> This may sound mean to the EQ coders, but I can't imagine what they're possibly doing that could require 500MB of RAM if they were coding properly. I mean, do they trickle-load levels at all? Where on earth could that be going?
It doesn't appear to be much of the code bloat, it is instead the high-polygon player and equipment models, enormous textures, motion-captured animation sequences, etc. I think there probably is a lot of room for improvement on how they handle it to reduce the memory usage. Using the new graphics engine in the days prior to the Luclin release, the game could still run on a machine with 64MB RAM easily. At the release of Luclin, resident memory usage spiked to nearly 800MB before they reduced it.
I would say there is some massive room for improvement. However, it has definitely helped fuel hardware upgrades among the gaming population...
> Always thought that NT was a pretty direct child of VMS
The biggest OS/2 relics in NT are the HPFS filesystem, now extended and called "NTFS", and LAN Manager networking (SMB + CIFS). Windows NT 3.0 and 3.1 included native HPFS support; I'm not sure about recent revisions.
There's a link below to a nice history of Microsoft, IBM, and OS/2. Even casual reading can illuminate to one that there is zero, or nearly zero, relationship to be extracted between OS/2's support for MS Windows executables, and GNU/Linux's support of same via Wine/WineX. VMS is very prominent in the development of Windows NT, and IMHO the NT series (NT 3.0, 3.1, 4.0, 2K, XP) are a nicely-engineered, functional piece of work. However, GNU/Linux is more usable for me on a day-to-day basis at this point...
http://www.os2bbs.com/os2news/OS2History.html
> You know, you generate more pure textual data per time unit than anyone I've ever seen on Slashdot. :-)
:-) I'm thinking of starting my own blog so that I can expound in more detail, and simply provide a link from Slashdot so I don't exposit more than the underpinnings of the argument in the thread of the discussion, saving the reams of ruminations for another forum...
.01 Linux kernel was not yet ready for prime-time, real threading was not introduced until the 2.2 series (with varying definitions of "real"), yadda yadda yadda. Fundamentally, the operating system that "took over the world" was Microsoft Windows 95. It has now been largely abandoned in favor of Windows XP, just released last year. For much of the last seven years, however, GNU/Linux has remained fundamentally the same: a 32-bit UNIX-like kernel using free GNU tools. No painful migrations, like for MS Windows users in 1995, to a 32-bit shell running on a 16-bit operating system.
Non-geek translation: I write too much
> I'd be more inclined to say that the Linux underpinnings are lightyears ahead of the Windows underpinnings
Again, I concur. However, I try to avoid such language generally because people come to regard it as flamebait for a Microsoft Windows vs. GNU/Linux debate. That said, many Linux kernel abilities (such as true multithreading) are of comparatively recent date... there's definitely some mutual "chasing the taillights" going on.
Interesting to me, however, is that Microsoft Windows NT development was inaugurated in 1987, with release in 1993. Linus Torvalds unleashed the Linux kernel on August 25, 1991. It seems the reality is that GNU/Linux, far from being an "upstart" operating system, predated Microsoft's second 32-bit operating system (OS/2 was their first) by two years. However, it's tough for me to be positive of the data prior to 1995, when I discovered GNU/Linux and started goofing with it from time to time, finally using it seriously and full-time in 1997.
The usual weasel-words apply: the
My personal feeling is, with tools like WineX, we are bringing more and more advantages to the GNU/Linux table. People I know are converting, or at least running a second box to check it out, which would have been almost unheard of outside of geek circles seven years ago. The consensus-based development model for many free software projects may be a very slow process, but it is one that develops an enormous momentum. Eventually, Windows users that are accustomed to making upgrades every 2-4 years will simply make the ultimate upgrade to a GNU/Linux system.
Wow, cripes, I'm lapsing philosophical this morning. And long-winded, as always.
Yep, as I mentioned in another thread, Microsoft had their fingers deep into the vital organs of OS/2, and when they ripped them out there was little IBM could do to stop the hemmorhaging.
I hadn't realized about MS's development FUD campaign. That would explain the lack of killer apps for OS/2 -- thanks!
It definitely is an entirely different ballgame now with GNU/Linux and Wine rather than IBM's OS/2. My goal was to fix the mistaken impression of many posters that the reason OS/2 failed was because it emulated MS Windows. While it may have been a teeny, tiny factor in the overall picture, the fact is that Windows NT is the illegitimate stepchild of OS/2 and VMS. Microsoft divorced IBM, left him with all the bills, refused to pay child support, and hired professional druggists to attempt to poison the child's prescription.
So I think that Crossover/WineX are a very good thing, ultimately leading people to software freedom. It will be a long road, but it's a better "ecological model" (to use Linus' verbiage) than commercial software, and in the long run it will kill commodity software off.
> What, precisely, is it that you want "integrated"?
:)
At this point, I want just a few, simple things that I take for granted in a business accounting package:
* Automated invoicing. This means that I enter the invoice information, and the accounts receivable information is automatically entered into my books. Also, that I can print out the invoice from the accounting package, and record that I printed it. Envelope labelling based on invoice information, and the ability to get the address information when I click on the accounts receivable entry would be a nice bonus.
* Check printing
* Payroll "understanding". I mean, enter the tax information for the employee one time, and from that point on when you verify gross pay the taxes will already be resolved.
These three things would make a business accounting package usable for me. At this point, GNUCash is a nice personal finance utility, but fairly inadequate for business usage. If it annoys me enough, eventually I may get around to writing some modules for it, but my business has been shelved temporarily (due to the economic climate) while I do a regular, full-time gig to pay the bills
>native, though less capable (i.e. no usable alternative shell) UnxUtils
I've used unxutils as well. They are OK. For routine, automated systems administration, though, they are simply maddening! I'm not terribly concerned about the latest software rollout working "fast", so Cygwin's slowness is irrelevant. I just want the software update rolled out across some 300 NT/2K workstations at some point over the course of the day.
However, I've tried to use cygwin to run KDE before (in hopes that I could somehow have an "integrated" platform to work from, using the GNU/Linux interface with which I'm familiar, running atop Win32), and it definitely is incredibly slow for some apps. Wine is lightyears ahead of Cygwin, performance-wise, IMHO. Of course, we're comparing apples and beach balls, here, so it's entirely my subjective opinion.
The model limitation is related to the physical memory of your machine, not the video RAM. For whatever reason, under WineX my 512MB RAM shows up as 504 to Everquest (perhaps due to AGP texturing, I'm not entirely sure), and it only recommends I enable up to 20 models with that limitation on RAM. It seems to eat a great deal more RAM under WineX at the moment than under Win32, that's for sure.
:)
[blockquote]
NT has OS/2 roots?
[/blockquote]
Sorry to go offtopic, but why not
Major portions of the operating system have roots in both Digital's VMS operating system and IBM's OS/2. Microsoft and IBM began work together on OS/2 in April of 1987. By October of 1988, Microsoft had recruited David Cutler, of Digital fame, as the project lead for NT. Up until 1990, the NT "personality", or UI, was wholly OS/2; while the "core" of the operating system was VMS, the API and user interface were very, very much OS/2-ish. Later that year, Microsoft leaked plans to fork the OS/2 API into the "Win32" API. Microsoft was slated to be the developer for OS/2 3.0, but decided it would be more profitable to release a 32-bit operating system that was wholly their own rather than continuing to rely on IBM's OS/2 license.
In a nutshell, OS/2 was killed by Microsoft because they abandoned development on the work to develop their own. Note that OS/2 had a SIX-YEAR lead on Microsoft Windows NT, and still couldn't beat it. They weren't even the first movers in the 32-bit arena. VMS had been released on VAX in 1978. The reason it was so easy for OS/2 to run Windows applications was because the Win32 API, at the time of the Windows NT 3.0 release, was barely a minor fork from the OS/2 API. If I recall correctly, IBM licensed the Windows 3.1 emulation from MS as well, but I could be wrong there. However, they still had a problem running Windows 95 executables due to the whole 32-bit yet running on DOS thing. By the time their emulation was up to snuff, Microsoft already had had enormous market penetration.
Back to the topic at hand, the WineX/Crossover efforts are a completely different ballpark than IBM's emulation. Microsoft Windows is in much the same situation now as IBM was in 1993 after the release of NT: there's a new operating system on the block which is very cool, very sexy, and very incompatible with our system. However, it's gaining market share at an incredible rate, and there's not much we can do about it. The key difference here is that GNU/Linux is not in bed with Microsoft, dependent upon the very company it is competing against for continued development.
OS/2 died because Microsoft had its fingers buried deep into the vital organs the whole time. When they ripped them out, all of IBM's efforts to stop the hemmorhaging failed rather predictably. Windows Emulation (or, if you are a Wine zealot, "implementation") on GNU/Linux is a completely different situation.
Hmm... not exactly "side scrollers", but these are some 2d-ish fun games I play on GNU/Linux quite often:
Chromium. Requires 3D acceleration, however. Nifty particle effects, runs at 50fps even on crappy accelerators if they have decent GNU/Linux support.
Super Methane Brothers: Easy enough for a child to play, but rather addicting.
Plus an assortment of other 3D and 2D games that are quite comprehensive on Mandrake Linux.
It's a compelling argument you make. My office is currently investigating a Microsoft Outlook deployment. I eagerly searched for open-source/free software alternatives that achieve the same integrated functionality, and came up lacking.
It's better to use free software on a non-free operating system than to use no free software at all, but it's worthwhile to examine the possibility that the Cygwin project, among others, are making it easier for people to stay with Microsoft Windows than use GNU/Linux, just as it's worthwhile to wonder whether Wine and WineX are are keeping more people tied to Windows as well.
My personal experience? Administrators using Cygwin become very impressed with the power of the *nix-like environment. They begin using it on a day-to-day basis, and see no reason to stop using Microsoft Windows on their desktops.
Until the day they run up against the huge limitations of using Cygwin, be it the performance penalty, lack of target platform support in many popular software packages, or simply any one of a string of painful interoperability problems. They then install GNU/Linux on a second partition or a second machine, and experience the full power of the operating system, and amazing graphical user interfaces which have been blazing trails for the last 2 years that Microsoft has attempted to follow. In turn, the best innovations of the MS desktop have been integrated into these free projects, as well.
So I'd have to say that because of the enormous minority situation GNU/Linux is in, Cygwin and running free software applications on a proprietary operating system like Microsoft Windows is an excellent "Gateway Drug" for power users, and eventually lesser users. Likewise, being able to run non-free software on a free system eventually leads to the replacement of that non-free piece by those who value software freedom over the convenience or entertainment value of the non-free software.
From my point of view, all roads lead to the ultimate "killer app" of the GNU/Linux world: the General Public License and the open culture it forces on otherwise selfish individuals and corporations. Yep, it's viral, and those who wish to take away software freedom hate it. Heck, a lot of freedom-loving individuals hate it too because it infringes on *their* individual freedom by requiring them to contribute their code back to the common pool. Love it or hate it, it's the reason why people are migrating to GNU/Linux from proprietary systems (of all types) in unheard-of numbers.
This is a difficult question for anyone to answer. You see, Sony Online Entertainment (SoE) puts people in an uncomfortable position: If you admit to having used ShowEQ, you can get your account banned. For those for whom the game is an important part of their life, this would be a horrible blow and they should consider very carefully whether they wish to acknowledge such use even on a semi-anonymous medium such as Slashdot.
:) I just shelled out $35 today, after a five-month hiatus, for a 3-month account because it's supported under Transgaming WineX. However, the remedies for being caught using infringing software are quite clear that one forfeits the subscription and monies paid to SoE and cannot use the service any longer. I can live with that penalty if I decided to use ShowEQ again.
That said, it is not illegal to run ShowEQ. It is, however, definitely a violation of Sony's End User License Agreement to which one affirmatively agrees each time one plays the game. If you are comfortable with the potential consequences for violating the "contract" you have entered into with Sony Online Entertainment, or else confident enough in your discretion that you don't believe they will find out you have abrogated your agreement, read on.
Yes, ShowEQ works with EQ under Transgaming WineX. You have to know what you are doing in order for it to work, however. Check out their message boards at http://seq.sourceforge.net for details.
In the interest of full disclosure, and since I really don't give a rat's third left whisker if SoE cancels my account or not, I have used ShowEQ. I am currently using Transgaming WineX. However, I'm not using ShowEQ anymore, least of all alonside EQ under WineX. It just seemed to take a lot of the fun out of the game to have all the monster locations disclosed. The stat-tracking stuff is pretty cool, though, but you can get similar functionality using a non-agreement-infringing log parser. While I wouldn't discourage curious hackers from checking out what ShowEQ can do, I would encourage them also to avoid using it in day-to-day play. It tends to lead to either boredom or conflict.
Umm, that said, if any SoE employee reads this, I'd rather not have the account terminated if it can be avoided
The question, before you plunge into the world of ShowEQ, is:
Can you live with the consequences of getting caught?
I say, check it out, and write up your experiences at seq.sourceforge.net. Hell, they even have an Electronic Frontier Foundation donation account set up so you can get a warm fuzzy supporting software freedom while supporting EULA-infringing software. To top it all off, you don't even need the binary package release of Transgaming's WineX to have easy installation of WineX. Gentoo has winex as part of their portage system, and their are binary builds of the CVS checkout available on the net. Those work perfectly well for playing EQ. The only reason Transgaming has a binary-only release for subscribers is because they have licensese for certain copy protection schemes which require them not to release the source for that particular application of the suite.
I concur. I have been playing Everquest for several hours today, and full-screen (as on Windows by default) it has been rock-solid as far as stability is concerned. The memory usage seems significantly higher than under Windows at the moment, so that I cannot load all the models with 512MB RAM, but otherwise it's quite decent.
.exe being unable to run. However, the game still ran fine.
:)
Things that aren't perfect regarding Everquest under x86 GNU/Linux at the moment:
* The installer does not give you any indication of progress until it is ready for you to change CDs or it is done installing.
* I experienced one error installing Shadows of Luclin, regarding some texture expansion
* The patcher takes longer than it does on Windows
* The patcher (at least on my install) segfaults instead of launching Everquest. The workaround is to invoke Everquest with "winex eqgame.exe patchme".
* Alt-Tab results in some goofiness under KDE. Probably just due to my window manager, but alt-tabbing out, and then clicking or alt-tabbing back requires that I hit ctrl-alt in order to have any keyboard input. Otherwise the game seems to think I always have Alt depressed. Note this doesn't seem to happen when running it fullscreen.
* Alt-clicking items to paste them as a clickable link in chat doesn't seem to work; my window manager grabs the alt and tries to drag the window. Yet another problem that doesn't show up in full-screen mode.
* Performance at the moment is significantly slower than under Win32. It's not horrible, but it is noticeable.
* Luclin player models have some visual glitches. Due to the age of my hardware (Geforce2MX, 933MHz P3, 512MB PC133 RAM) I've rarely played with them turned on anyway, but I did so briefly today and noted that most characters appeared at least partially nude.
* If you use more Luclin models than recommended by the Options dialog, it's possible EQ may not load at all. It effectively grinds to a halt if the resident memory usage displayed by "top" comes anywhere near the physical memory limitation of your machine. With all models turned on, by the time I got to the 1-frame-per-second character selection screen, EQ had some 400+ MB resident memory requirement, and a total RAM requirement of 820MB...
I haven't had Microsoft Windows installed on my gaming PC in several months, and renewed my EQ account just to check out the support under WineX. This is a nice boon, although I'm certain my wife won't be as happy as I am that EQ runs under GNU/Linux just fine now. Darnit, my Armagetron, BZFlag, and Chromium binges are going to be curtailed I think.
All in all, I'd say it's a fairly successful initial offering to appease the rabid EQ supporters among the Transgaming subscribers. It's still "unsupported" and has some rough edges, but the glitches that are present are certainly not show-stoppers.
Regarding the earlier comments about how by using Wine, one is treading sand that swallowed up IBM's OS/2:
OS/2 failed for a myriad of reasons, not nearly solely due to support for Microsoft Windows programs. Although many believe that was a factor, I do not believe that it had any major impact. There were many things leading to the downfall of OS/2, not the least of which was terrible marketing on the part of IBM, the lack of any "Killer App" for the platform, and the lack of significant bundling deals with OEM's which meant that OS/2 had to be installed by the user (a task then, as now, beyond the interest of the majority of computer users). Indeed, GNU/Linux shares some of the same problems right now, regardless of Win32 exe support.
However, from my point of view, GNU/Linux is in a uniquely superior position compared to OS/2, and the ability to run Win32 packages natively has nothing to do with it. The secret weapon is the GNU General Public License. Business FUD regarding the platform is ineffective in the long run, because the operating system is guaranteed to live on, barring unforeseen litigation. Nobody can fork off their own version of the OS (Read: Microsoft forking Win NT from the OS/2 effort) because the GPL requires those changes to be given back to the community if they are released in binary format.
We're in a completely different battlefield today than the one that sunk the OS/2 Titanic. GNU/Linux's weaknesses are becoming its strengths. I agree that the Wine approach may not be effective at winning over large numbers of MS Windows users to GNU/Linux. However, that's not really the point, IMHO.
See, I wiped out MS Windows because I value my software freedom.
I don't find playing proprietary games under GNU/Linux to be an abrogation of these principles. Games are content-driven (in that new content drives sales), while productivity applications are procedure-driven, and any procedure can be automated and commoditized. Basically, IMHO games are art, and making the concession that the publisher must get back the production cost of the art before the programmatic portions can be released under the GPL seems, to me, to be sound philosophy which acknowledges business needs while supporting software freedom.
Because I value software freedom, and am far more productive under GNU/Linux and KDE than Microsoft's operating system, I replaced Microsoft Windows with the commoditized, free software GNU/Linux. Many others are making this same step. Transgaming's WineX supports these subscribers by providing them with a service unobtainable in an exclusive free-software realm. I value the service, and contribute my $5 a month to their continued success.
I feel the need to comment on the disconnect many Slashdot posters seem to feel. They rail that "Free Software Zealots" seem to change their minds with the topic at hand. The reality is, if you watch those who post, some users feel passionately about a certain topic, and others don't. I've found precious few comments amongst several topics that would cause me to believe the poster had an internal disconnect regarding their point of view on free software.
I have one position I have held for at least the last three years (while using GNU/Linux for years before that for simply practical reasons), and it is quite similar to that which Richard Stallman has claimed many, many times (in different words): The point of the Free Software Foundation is to preserve freedom. In some cases, when there is no adequate free alternative, proprietary software can be used. It is better that a user have a free tool on a proprietary operating system, and thus begin to understand the importance of software freedom, than to have no freedom-supporting software at all. However, if use of the proprietary system can be avoided, it should.
Transgaming provides a valuable commodity: the ability to play popular, but still proprietary, gaming titles on a free operating system. It's a bit the reverse of the FSF's usual line. Games, however, are a form of art, and stand uniquely apart from that which can be easily commoditized into a free system.
As a side note, as far as I have found there is no *good* current free software alternative to a proprietary package such as Quickbooks Pro. I've tried MoneyDance. I've tried GNUCash. If you know of more that fit the bill, please educate me. In the meantime, products such as Transgaming's WineX, and Codeweaver's Crossover Office allow me to use proprietary tools such as this, yet keep my system otherwise as free as I possibly can. There will come a day when the free alternatives do the job "well enough", and that will be the day I convert to a free software financial package. This is neither a contradictory position, nor one that is out of line with the "hard line" of the FSF. It's simply an acknowledgement of reality.
So please, those who've railed against the Slashdot community's collective Jekyll/Hyde syndrome, realize that different people post to different threads. There's a significant contingent of chronic copyright infringers on Slashdot. There is also a significant contingent of those who would never break copyright law, even if they think it is bad. There are those who are free software zealots, and refuse to use non-free software. There are those who only use proprietary software, and snub their nose at the attempts to replace sophisticated proprietary packages.
And there are those in between, that hold both the ideals of the Free Software Foundation quite dear, but also acknowledge that idealism and reality must reach a compromise on our desktops if we are to use these fantastic tools to their ultimate advantage.
That's me. Let's not debate over whether this is a horrible move for the community or not. There are at least 3,000 Transgaming subscribers that believe in WineX, and pay their $5 or more a month to have a voice in the direction of the software. Those of us who do are grateful for the service, and happy to appease our free-software-loving sides because those portions of the WineX suite which can be free software (and not encumbered by licensing) ARE kept free by Transgaming.
Then again, this is Slashdot. Here we debate over whether or not to have a debate
In most organizations, the NOC staff, CIO/CTO/whatchacallit, in-house "tool" programmers (those who write programs which are not a profit center), IT managers, printer monkeys, etc. are all considered "support staff" for machines. If you have a 24/7 rotation of a single NOC on-staff, to be able to handle vacations and such on a 4-on, 3-off rotation, you need at least six guys. Preferably seven, in case two people are sick or take vacation near the same time.
I'm not trying to straw-man you, but most people consider the total number of people supporting machines to count toward the mystical "user to admin ratio". For instance, back in a 90-person, 100+ machine organization several years ago, we had two support people. We took care of all the desktop support and kept the servers running at about a 50-to-1 ratio. The "unseen cost" was the fact that the programmers often handled many basic systems administration tasks themselves. At my current job, in a 400+ person organization with, admittedly, much more severe development needs (banking industry, lots of development to integrate vital services, rather than "run the file servers"), we have a total of 12 support people:
1. Chief Information Officer.
2. Manager of Systems Administration.
3. Programming lead.
4. Java & Windows programmer.
5. Java & *nix programmer.
6. UNIX Admin (-- me).
7. Telephony admin (we're mostly a call center, so this is a HUGE job).
8,9,10: Systems Administrators.
11,12: "Night Operators" who swap backup tapes, handle printed reports, act as security, move stuff around we don't want to move during the day, etc.
If you wanted to stretch it, you could say that we have four "sysadmins" for 400 people. Woot, 100-to-1 support ratio! Reality is quite different. I personally think our organization will have some pruning in the near future, and am working my butt off now to bring our system up to speed with automation issues so we don't have to spend all our time putting out fires when the RIF comes. Realistically, though, we have close to a 40-to-1 machine/admin ratio (including servers, workstations, etc.) For many organizations, if you are not 100-to-1 including *all* your support staff and hours spent by other people doing sysadmin tasks, you're not where you need to be and will have a RIF once management figures out where the costs are distributed.
Magnificently automated processes are the stuff that makes high admin/user ratios work. No bones about it. But if you're going to count it, make sure you count the full cost of systems administration, including the people you don't think are sysadmins. They are still support.
Important note: support ratios are, and IMHO will always be, majorly out-of-whack for software development firms. The needs of a software developer are so different from that of a secretary they simply can't be easily compared. About the only choice there is to distribute a great deal of the administration load among those who work on their machines. There are exceptions, and modes of development that minimize this problem (dedicated sandboxes, shared compile farms, etc.), but overall you'll generally have a lower admin/user ratio in software development.
I hate to add the obligatory GNU/Linux reference to your post, but here goes anyway :)
You already see this future in devices such as the Agenda VR3 PDA. Although the company itself has disappeared in the U.S., the user community has applied some interesting technologies to solving the problem of very little storage, very little RAM, and a relatively slow CPU.
The Agenda has 8 MB of working RAM. The data storage for the unit is 16MB of flash RAM. On initial release, the Agenda was very slow and problematical, primarily because it would eat up its 8 MB of RAM with the kernel + X Windows, leaving precious little for applications. The solution was ingenious: combine compressed storage on the flash (to conserve that 16MB of flash storage) with "XIP", or "eXecute In Place" technology in the Linux kernel. XIP prevents one from storing the file to be XIP'd uncompressed, but allows one to actually run the program directly from the storage media, without loading much (or any) of it into RAM. With hard disks, this would be unbearably slow. With flash RAM, while considerably slower than regular RAM, it works pretty well.
I'm actually considering implementing an XIP solution for our servers at work. I'm trying to aggregate about 400 employees on some servers, but the problem is that if each one of them is running off the server, the RAM overhead of KDE + apps for each person is abominable. If I had a common, solid-state-storage (or else a very large RAMdisk), combined with XIP, it may be possible to dramatically reduce the memory overhead of launching hundreds of copies of the same app. XIP it on a RAMdisk!
Of course, it may not work. As a matter of fact, I'm betting that it won't work. But wouldn't it be cool if it did?
Elimination of magnetic storage will probably not be complete during my lifetime (I'm 29 now). However, I consider removing the idea that we store on one media, and run from another, to be beneficial to the industry as a whole. Faster media, programs that execute directly from media rather than intermediately storing in RAM. Woot!
I don't post much, but when I do, I like to be right, or at least have an informed opinion. Thanks for your clarification! I hadn't thought about the fact fraud was happening as well.
The answer is a resounding NO. If you rebroadcast their material, substituting your own ads, you have profited by infringing copyright. Motive in copyright infringment cases is very, very important. If you do not profit from infringement, the violation is not as serious as if you gain profit. Additionally, profitting from copyright infringement leaves you liable for damages equal to or greater than the profit you gained by infringing someone else's copyright.
It's not theft. It is copyright infringement. They are two dramatically different things, although the major software and media companies would have you believe otherwise.
Again, you are incorrect. If you produce goods similar to, or in many cases as identical to (in any case where patent, trademark, or copyright do not apply), someone else's and place your own trademark upon it, you have done nothing wrong. If, however, you place another's trademark upon it (implying that it was produced by the other manufacturer), you are guilty of trademark infringement. Alternatively, if you use a patented invention and do not pay patents to the patentor, you are liable for patent infringement claims. Similarly, if you duplicate a copyright work, you are not guilty of stealing from the author; you are guilty of infringing upon his right to control copying of his work.
The only reasons one could proffer the arguments above, that I can see, is 1) simply lack of education regarding U.S. law. I am not a lawyer, but I do believe I have a sound understanding of laws where they affect my day-to-day life. There are also 2) those paid to have that viewpoint. Hilary Rosen and others are paid part to promote these views of copyright infringement as theft. To promote an alternative view disagrees with the corporate agenda, and this disagreement would most likely eliminate their sources of income.
The fundamental problem with the thinking comes about because of the nature of what we're dealing with. Information is trivially reproduced, even when spoken. I suggest you study the history of copyright, to fully understand the nature of the laws. Today, we have a society where such information can be reproduced for (effectively) free. It's my personal opinion that Copyright is a doomed concept. However, we have not come up with a suitable reward yet for authorship to promote the science and arts that is not Copyright. Until we do, we will be stuck with this system that so obviously maps so poorly to reality.
There are certainly cases where the line between copyright infringement and theft is very blurred. For instance, if one breaks into a computer system and makes copies of information that were never intended to be made public. One has obviously violated copyright in that case, since U.S. law regards all authorship as copyrighted. Is it theft? In that case, I don't know; just as "breaking and entering" is considered "breaking and entering" (vandalism and trespass, if you prefer), if you don't steal anything for entering, but instead copy important documents, you've not stolen the documents, but made copies in violation of the wishes and reasonable expectation of the holder. The company or individual never intended to release the information to the public for profit (the point of copyright), the information was reasonably expected to remain private, and consent for this action was implicitly denied. In that regard, information violation seems more analogous to rape than theft: one has expressly violated the wishes of the holder of the information, taken nothing from them, but used them in a way inconsistent with their will. Copyright infringement on released goods, however, is similar to using a hooker for her intended purpose, but refusing to pay her. One has no implied contract, the other does. The penalties for rape are spelled out in the law, and include government-sanctioned prison time. The penalties for not paying your prostitute are the same as for not paying any service person: if your bill is not paid, you are sent to a collection agency, which then may take you to court to seek damages. It is (often) not treated the same as theft, since the "goods" (a service) are intangible, you have not deprived anyone of anything except time invested (which has value, but is again intangible and cannot be stolen) and potential profits. In some cases, particularly where the one infringed upon believes the intent was to defraud (once again, fraud law, not theft), they may seek criminal remedies. Most don't, though, because by so doing they are depriving themselves of a potential customer, getting bad press, and preventing the infringer from quickly paying the damages by depriving him/her of income.
Note that the paragraph above is entirely my opinion, and not really part of my initial refutation. I simply think that most software companies and authors would do well to remember that they simply sell their time for money. Their "product" is a service, and our current model of copyright attempts to treat information as a tangible good, which it is not. Those prepared to acknowledge this fact (as Microsoft seems to be doing with their license renewal services) will probably do OK as the economy transforms to take advantage of new realities. Those who insist on treating intangible as tangible will eventually go out of business as realists (the customers) begin to treat it as the intangible, inherently value-less thing it is.
A few links for you to peruse:
Fundamentally, it's a thorny issue with a whole lot of ramifications. Those who attempt to cast copyright infringement as a black and white case of theft are intentionally misleading you as to what is going on. It is its own legal domain with its own remedies and penalties, entirely aside from traditional laws regarding property theft. There is no such thing as "intellectual property": there are patents, trademarks, and copyrights which give rights to works based upon a well-established but sorely broken legal framework.
Although you are factually correct, you unintentionally mislead users by the fact you omit. This is an Apache exploit which only occurs on GNU/Linux x86 systems. The code, IIRC, just doesn't compile anywhere else. If the one devising the exploit was a more competent programmer, it could have been a cross-platform Apache exploit, but the reality is that right now, you have to be running a specific GNU/Linux-Apache-mod_ssl setup in order to be infected. In that sense, then, it's a "Linux worm", because no other platforms will get nailed by it.
Caveat: Attach "at the moment" to the end of that last sentence. I fully expect someone to grab the source and modify it to run cross-platform within a few weeks.