Which puts it between at least 10, possibly 11 years now.
Not sure if that's authoratative, but it's at least a date I found via a quick google search.
It's been obsoleted by the PS2 for what a little over 4 years? (It came out during the Christmas shopping season of 2000 if I remember correctly).
However, it's not like a lot of titles are being released. It's not like the blockbuster games are being dual ported.
I'd see fewer problems with this, if the X-Box 2 was going to be backwards compatible with the X-Box. However, from everything I've read, it's nearly a technical impossibility to do that if what has been publically guessed about the X-Box is true (I'm not sure if Microsoft has officially said anything besides that ATI will be making the video cards). I'm not paying that much attention. I believe it's supposedly going to have a PPC chip of some time (possibly a Cell, which is PPC + an array of vector processing chips if I understand it correctly).
If they released backwards compat consoles ever 2-3 years, I wouldn't care. However, releasing non-compatible ones every 2-4 years is just insane from a consumers perspective. However, as Microsoft is just gettings it's feet wet, I could see why they are doing this (fix thier previous mistakes, and get a head of the game on the hardware cycle is probably a good idea from a business perspective, especially if they can finance the losses).
If you really want this to work under Linux, give the information you have to the Ghostscript guys (that used to be GNU Ghostscript, but I believe they've broken away, Aladdin didn't like it for some reason).
They know printers. They know lots about printers, and printer languages. My guess is that they'll be thrilled to get an opportunity to hack another printer working. I know when I bought a printer that has "PS Support", it had a postscript driver in software that talked a propriatary protocol to the printer. They would have gladly written the output driver for it, but they didn't know how it worked.
Maybe if you know how it works, you'll be able to get them to do something with it.
At least in the U.S., I know a couple of dispatch people (the train equivilent of "Air Traffic Controllers" for planes). One always told me that a train dispatcher made more life of death decisions per hour then then an ATC person did. He joked that guys used to retire from doing train work to being an ATC for the low pressure, low stress atmosphere. I'm not sure either of those statements is true, but I'll bet that it is true, that keeping trains running smoothing isn't as trivial as you make it out to be.
In the end, well, trains don't always travel the same speed, they don't always travel at the speed they are told to. Sometimes they break down. Sometimes the switches aren't thrown properly (so there really are two dimensions, possibly more), sometimes a train runs away. Some times a train is on a section of track its not supposed to be. Trains aren't a trivial problem (we actually had to write a simulation of this in college in a RTS class, you had to do the computations and throw the switches at the right time, or you had yourself a fairly serious collision).
With trains at least, by the time anyone can visually tell you this, all you can really do is jump off and save yourself. It's literally a million pound weapon of death, by the time anyone can see the problem it's over. Having a GPS system on the train would enable you to spot all sorts of upcoming problems with out having to communicate with anyone onboard.
I'm fairly sure I don't have to. He didn't say that the company wouldn't generate any revenue. He said that if Microsoft price stopped rising. Look at the
5 year stock price. It sure looks like it stopped a while back.
Considering it fell 50%, 4 years ago, and well, the fact that they don't issue options like they used to. That they haven't had a metoric rise, yes the last 4 years in fact prove that they can manage to run a company without a rise stock price. They are still a part of a major indexes, which some people feels gives them a solid buyer for any new stock they issue as part of the employee compensation plan (so they can issue stock to employees, the employees then dump the stock on the institutional investor, there is always a buyer, so the price is artificially bouyed because there is artificial demand due to investment strategies in index funds). There might be some truth that if they are removed from the index funds that things might work out a bit differently for them. However, that's not what he proposed. He stated specifically, that "the stock price stop rising".
If they shut down Microsoft R&D, I'm fairly sure the cash management is structured so that the company can run for several years (8-12 quarters). They say that's one of Bill Gate's obsessions. When money was tight way back at the beginning, it was always his goal to have the company finances such that they had at least a years worth of operating expenses on hand. If it's pure juvinile fantasy, pull the balance sheet. It shouldn't take but 2 numbers cited, plus a single division to figure out how many quarters they can actually operate. It's not like the math is particularly difficult:
Cash on hand / quarterly expenses = number of quarters that can be operated without revenue.
That's the number I believe you are discussing. Which I still think is longer they you claim.
The number I was responding to was this different, and I'm fairly sure that the recent past is a disproves the statement if it had been said 3 years ago. Over the last three years, I believe they have posted record profits, and that the last several quarters have gone extremely well due to X-Box and Halo related revenue.
You realize that this is a company that books something on the order of 1 billion dollars a month in profit right? I know it used to be 1 billion dollars a quarter, but not too long ago, I'm fairly sure they posted 12Billion dollars in a single fiscial year.
I'm not sure what their operating costs are (I believe it's something just insanely low once you remove their research costs). Yeah, I'm fairly confident they can stick around once the pyramid scheme collapses.
By the by, I've been told the the pyramid collapsed a while back. Anytime a company re-prices options, it's a fairly sure sign, that a) they are using stocks as a primary form of compensation, and b) that the jig is up, and no one will ever get wildly rich of options again. They repriced a ton of options after the bubble burst and the stock price plummeted 50% from the all time high. I knew several people who were recruited to work there pre-bubble days. Both of them turned down the jobs, as they were about 2/3rds the going rate. However, you got enough options to turn you into a millionare in 10 years assuming they stock price continued it's incredible price rise. Relatively high risk, high reward.
I've lost the link, but there was a pretty good economic analysis a while back that showed, essentially that new investors in Microsoft were paying Microsofts wages and a lot of their taxes by buying up all the stock that got dumped into the market as converted options. The interesting part was that they ended up showing that Microsoft would lose a billion dollars a year if they paid the money out that employees got as options. The thing about that is, that I don't think Microsoft will make nearly as many of their employees millionaires as they used to. So I don't believe they'll lose all that money. I think they are right, that Microsoft played the market to the hilt. The market thought it was getting in on a good deal, when really Microsoft was using the markets capital to compensate employees.
They are still doing it to an extent by giving away actual stock. However, my guess is that is probably a great deal less lucrative then options were back in the day.
The other explaination I've heard put forth, is that Microsoft want's to be able to compete with VMWare. Essentially, virtualizing very big machines is big business. There is plenty of money to be made there. The SA where I work used to do exactly that on server farms. You'd just setup a farm of machines, and install VMWare on all of them. You'd create an instance of each independent server you want running. So rather then purchasing 4 machines, you'd purchase one really big bad machine, and create one instance for each of the 4 machines you wanted.
You save space (which is important in CoLo pricing). You use resources more efficiently. When one one of the other instances is relatively idle, a busy instance can use up those CPU cycles (not true with separate hardware).
I'm told, that if it doesn't exist, it will at some point in the future, that you can migrate instances from VMWare install to VMWare install while it's running. Thus you can migrate running instances from Machine A to Machine B. Down hardware A, do whatever upgrades are needed under the hood, migrate the instances back.
It's a relatively novel way to compete with LPAR's (or whatever the Mainframe speak is for that).
It's very, very impressive technology from what I've been told. Using it in conjunction with racks full of blade servers I'm told is the way to run an incredibly high density setup in a very flexible way and reliable way. I'd really like to try it at some point, if only because we have machines that sit 95% idle. I'd like to combine 3-4 of them onto a single machine (especially for services where they are simple to implement redundant backups: LDAP, SMTP, DNS, DHCP). Just setup a second machine doing pretty much exactly the same instances in the backup mode. It'd free up a ton of hardware that right now we are tying up so that the failure of one machine doesn't affect the rest of the services.
Don't take this the wrong way, but "I don't think that word means what you think it means". Or whatever the obligatory quote from "The Princess Bride" is.
You can have an innovative stone block bridge design. You know, the same things the Romans built 2000 years ago. You can still innovate on them.
innovation
n.
The act of introducing something new.
Something newly introduced.
Innovation inside of the VM subsystem of the kernel happens. It's esoteric and 99% of all people don't care, know or see it. However, if you are "introducing something new" into the VM subsystem, then it surely fits the definition of "innovation".
I have no idea what Rendevous is, or what it does. But I can give you an example of C library that is innovative. Readline written by GNU. It is innovative. To the best of my knowledge they were the first group to introduce such a beast, and to the best of my knowledge it is still fairly unique. It's a straightforward library that you can link into any interactive program where a person might edit a single line of text. It automatically gives you keymappings, history, and all kinds of other goop. So any application I use, that uses it has pretty much the same interface as far as I'm concerned (gdb and bash are the two applications I know I use it with all the time, I'm sure there are others). I like to configure it so I can use vi style commands into it, to speed commandline editting.
Innovation isn't strictly limited to a particular level of implementation. Innovation could happen in the processing of toliet waste water. It's still innovation. So I'm curious about how you feel an innovation can't happen in a C++ library. I mean, mozilla is nothing but a bunch of interconnected C++ libraries. I'm fairly sure there's lots of innovation in there somewhere. It's got to be contained in the C++ libraries somewhere.
I've seen this sort of thing before. Not out of Dell, but out of HP. HP sold two series of machines. Don't ask me the line of computer, this was approximately 5-6 years ago. They sold two computers that we're fundamentally the same. One came pre-done with Win98, the other with NT4.0 or W2K. We picked up the two machines with Win98 with the intention of installing NT4 on them (the price difference was bigger the buying retail NT licenses). At the time, HP was installing plenty of equipment they made/designed that only had Win98 drivers. I searched high and low for the drivers for the video and sound drivers for the onboard stuff for NT4. It just didn't exist. You couldn't get any sound, or anything want that wasn't 800 by 600 at 8 bit color under NT4.
Now, I know why the machines we're so much cheaper. It was a learning experince. We ended up throwing an old sound card, and a cheap recycled video card into the machine. It worked fine after that.
I've always heard the theory, that if at all possibly, you should try and name your company with so it starts with an "A". As it more likely that a customer just looking in the "Yellow Pages" will call you. My mother worked at "Assist Business Service" for a long time. Listed as "ABS" in the phone book. A lot of times, the reason they got picked was "Well, you were the first one in the phone book".
While you still have to be able to deliver after customers contact you, reducing the cost of aquiring customers is a serious boon to you. Getting to the front couple of pages of Google, is probably an incredibly cost effective way of doing that.
Keeping people on your site, normally involves having what they are looking for. However, you need to get them to visit your site first. The IT industry is littered with the casualities of better technology that failed because the lesser technology has better marketing.
Are you just being silly now or what? There are two things to consider. First, you could use the GPL'ed code will act as a filter. Take in said crappy file, output good working file. More then likely, you could take the GNUcash code and write some "plumbing" or "glue" code to get this accomplished. Where 99% of the real work is done by the GPL'ed code already written code. You just write the 1% to give you access to the interface the way you want it. Thus you wouldn't integrate all of that into your application, you'd just filter your files thru it before passing them to your code written under the license of your choice.
Second. The GPL isn't a patent license, it's a copyright license. US Patents protect the concept of how something is done. Copyright protects a particular expression of a concept.
You can take a look at a GPL'ed implementation, and then re-implement your own from scratch using the same concepts. If you want to be really good about it, have someone else look at the GPL'ed implementation. They write specification for how it works, and then you write an implementation from that specification. Thank you Compaq for setting that legal precedence!
Have you considered ballons? It's a much smaller effort, cleaning them up is relatively straight forward. The volume of the ballons is much more compact in pre-joke form (and probably cheaper). If he makes the mistake of poping the ballons, he'll be finding leftover ballon parts a year from now.
In order to deliver in the cover of darkness, rent several vans. Remember the logistical problem of this is to fill the volume of a building. You do realize that say a moving van full ballons will probably fill an 18ft by 10 ft by 10 ft space. So if it is a one story building, that is say 1000sq with 10 foot ceilings, you'll need something on the order of 5-6 moving vans full of (popcorn/ballons). To fill it ceiling to floor right for the entire building.
The bonus to this, is if you can find a rental place (in America, it's relatively simple to find, but I don't know about the UK (where I believe you live from the link I was reading) to rent a helim thing to inflate roughly half of them (ceilings) and then an air compressor to inflate the other half (floor).
With an air compressor (a $90USD), or a vacuum you can use as a blower, if making a lot of noise is acceptable, you can just stand outside the house and fill the ballons. The bigger the balloons, the fewer you need. It might be advantagous to find several just huge ballons (think 4-6 feet in diameter) to just fill large volumes to be inflated on site, while pre-inflating the smaller filler ballons to pack the place fairly tight. to make very difficult to move around in. The other problem with this plan, is that anything that is fragile and at a below the height you plan on putting the ballons/popcorn to might get knocked over during the cleanup.
This idea, isn't an original of mine. Some of my older sisters friends did this to my sister for her birthday. They filled her room to about 4 foot high with ballons in her room.
I've built up my "RedHat-fu", to the point, that it's the one distro I know how to to completely configure a machine from start to finish to be setup exactly the way I want via network boot. It might be possible with Suse, but I've never run that.
I purchase RedHat licenses for everything that is in the DMZ, or runs software that requires RedHat Enterprise Linux for support (think Oracle Databases).
Then I use Whitebox Linux for everything else. It's pretty much exactly the same as RedHat (you can pick another RHEL rebuild if you want, CentOS and Whitebox Linux are my two favorites). Whitebox can have problems from time to time, because it's a one man show. CentOS looks nice, but it sounds like the mailing lists are used less, and the web boards more for discussion and help (I've never participated, but that's the a complaint I've seen on WhiteBox lists about CentOS). I like e-mail lists for help/support. Call me silly. While web boards are nice for random discussions, I'd much rather review e-mail for technical support (both on the giving and receiving end).
I use that for the desktop. Other then, it's a bit RAM hungry, it's fine for a desktop for most people (the lack of a good MP3 player might bother most, but I play oggs, so I'm good with it). You need more then 128MB of RAM to run OpenOffice on it at a reasonable speed. (I was running a PIII-500 w/ 384MB of RAM and it was acceptable, with a new P4 w/ 128MB of RAM it was unbearably slow running Mozilla and OpenOffice at the same time. I put a 1GB of RAM in and now it's wonderful).
In the end, it means I can run almost exactly the same OS at home that I do at work. It's industrial strength, and all of the expertise I build up using it, is going towards one of the two distro's that all major software vendors support. I don't know of any Suse "rebuilds", otherwise I might recommend those.
Can you point out the text of that document? I've found the Press Release, but I can't seem to locate the legal text to review.
Here is the original press release. However, it doesn't seem it implicate what you are saying. I think they are releasing it to anyone who uses the CDDL (at least according the the PR). I'd also be interested in their legal definition of "OpenSolaris developer process". My guess is, either the "OpenSolaris developer process" will end up being so tightly defined that no one can release a version of Solaris without Sun's permission, or it'll be so open ended such that there's a hole that I can end up implementing the patents in any CDDL licensed project. However, without the legal text, I can't be sure. Maybe they are more clever then I think and came up with a definition that can find a middle ground between those two.
I've got no problems with them licensing the patents such that only CDDL projects can use them. While it will create the great divide in terms of mixing and matching source code, it's their right. However, if what they are doing is saying, you can only implement them in such a way that we'll end up with control over you via patent law instead of copyright law, that seems completely underhanded. It also seems to be completely counter to the image they are attempting to spin. I never thought of Sun as that underhanded (they are a coporation, and generally have interests that are somewhat contradictory to mine, but never thought of them as doing anything that qualifies as "evil genious"). Oh well.
Yes, and no. Keith Packard was the guy who started a lot of the "problems". He was talking about changing the fundamental governance, with the goofy word that meant, "free" in french. I've forgotten the name of the project (I can't spell in french, so I'm screwed even if I remembered). Keith wanted to change the font system (Which he did, so XFree86 no longer uses a font system designed in the early 1980's that a mere mortal might actually be able to use). He wanted to write extensions to do various things (alpha blending/transparency if I remember correctly). He wanted to check in patches that everyone was applying for 6 months because it was the only way to get decent performance from various cards. He wanted to do things to make it easier to write and maintain GTK and Qt (no one ever writes Xlib calls anymore besides those two projects, pretty much every one else uses those, or one of a various other sets of widgets).
I'm fairly confident that XFree86 would have ended up forked without the change in license. Too many people were too unhappy with XFree86 for a long time. There was a fork waiting in the wings, pre the license change. If you read Slashdot, you would have read about it. Heck, they posted notes from conference calls discussing how to change XFree86. People kept trying to get Keithp to come back to the project, and technically speaking, I don't think he forked it pre-license change. However, he was trying to setup how to run the project before bothering with the Fork, so it wouldn't die upon starting.
The problem with XFree86 wasn't the license (that just made it bloody obvious to everyone around that XFree86 no longer fit their definition of "Free", thus they really wanted to use the Fork) that the change had to happen. Meanwhile, they just happen to get to setup a sane way to run a project that is happy go lucky with how OpenSource projects are run best.
The fundamental problems are the high-handedness, and well, pretty much non-responsive nature of the people handling the XFree86 project. Fundamentally, you'll find that XFree86 is really disconnected from the interests of the people using the project.
Read up on fundamentally what I'm talking about. I'm not talking out my ass honest. Here's a pretty decent link to the a Wiki Entry
There is a reason to fork, Xfree86 was a reason to fork. But just because someone will not be granted CVS commit access is certianly not a reason.
You do realized that those two sentences are in fact contradictory. That in the end, it was not being given CVS access (actually, having CVS access revoked), that is why most of the X.org started.
It was because the XFree86 people had someone willing to do the work, but they wouldn't let him commit patches because they didn't agree with his design changes. He was one of the most active, and progressive people working on XFree86 at the time.
Finally a mention that 98% of all projects on sourceforges aren't forks (if each project has 60 forks, then only 2% are original projects). 98% of them, might be a duplication of effort, but strictly speaking, not a fork.
Linus should run some form of shared access, where he has lots of people who can commit code directly the the Linux source tree without Linus having to see it.
Presumable a more "FreeBSD" or Debian like system. FreeBSD is a lot closer to a meritocracy then Linux's "Benevolent dictator" system. FreeBSD has a core group of whom several (5-10) people can get access to the actual source repository. Supposedly FreeBSD is fairly elitiest and tight knit (think XFree86, they have roughly the same governance model as FreeBSD, but XFree86 sounds like a lot more of an old boys club them FreeBSD is).
Debian has a system where they are fairly democratic, and have a process where by you can initiate referendums to vote on a change you feel is important enough (generally never done over source code, but has been done over which version of the Linux kernel to ship, and what types of stuff has to be stripped from the Linux kernel before it meets Debian's definition of "Free").
Linus is a dictator of the stock Linux kernel. However, there are so many forks out there of different trees, where lots of people have access to those trees that it's relatively silly to discuss. The other interesting aspect, is I get the distinct impression that in lots of areas of the kernel, Linus does implicitly let people just randomly apply patches. If you are one of the people he trusts working on an area he feels you know best about, he just applies your patches with minimal if any review. You don't get access to his primary sources to do the patch yourself, but you get a relatively unfettered access to the areas you know about. Which is sorta nice, as well, you don't see the kinds of spats that spawned OpenBSD (CVS revision wars, where people undo others work because they disagree, and they have access). When there is a single arbitor of what gets access, it never seems like there are people of two minds in control of the source.
It's like the age old argument, that a Monarchy is the best form of government assuming you have a good and fair king. It's also the worst kind of government if you have a despot. Unfortunatly hereditary monarchies generate a lot more bad kings then good ones.
Nope. You've clearly never used a KVM, at least not one as tempermental as mine (the Belkin Pro Series). I've had pretty good luck with Belkin equipment. I know there are better ones out on the market, but generally they are extremely expensive. KVM's already cause enough video problems, even with really good cables.
Plus, most USB mice can be converted to PS/2, however, I haven't had luck the other way around Converting from PS/2 to USB, which is what I'd have to do to connect my KVM to a USB only MoBo.
The real irony here, is that if the guy said he wanted a miniITX (or full sized) that has serial only capable BIOS, I'd be all over it. I really want a MoBo that can run completely headless and hook it up to a serial terminal. Then I wouldn't have to worry about 90% of the goofiness of USB, PS/2 or KVM's. Then I could have one just sweet DVI/USB/no legacy fancy smancy machine. When I want access to any of the 5 or 6 server style machines, I just ssh into the machine. If I managed to break the machine, then ssh to the terminal server and fix the network. Then I wouldn't need a KVM. The only time I ever switch on my KVM from my primary desktop is to solve a problem that can't be solved over the network, because I broke the network configuration.
While I can do this via Linux and putting the console on a serial port, if I ever need the BIOS, I still have to hook up a keyboard, monitor, and moused directly to it. Which is a hassle. I've always been amused that AMI/Phoenix/BiosStar don't integreate serial only into their primary bios configuration.
Don't take this the wrong way, but uhhh, start identifing say, "normal" sized motherboards that don't have the legacy connectors?
I've never gone looking for them, but I've ever seen one in all of the years I've been looking for MoBo's. The last time I saw one without PS/2 was probably a 386. I think Dell was coming out with a line of machines that did that (in the end, I think they dropped the floppy, but the rest were on board).
We have some IBM servers at work that have a very odd connector that you have to by a breakout cable to turn into PS/2 keyboard and mice, and regular video. That's really just a space saving (it's also a KVM built into each machine, as you can chain them with a non-breakout cable, at the end of the chain you attach a breakout cable).
So while I see your beef with the mini-ATX machines, it's not like what you are looking for is terribly common for a x86 motherboard at all. I wouldn't expect mini-ATX to lead in that category.
Personally, I'd hate to end up with one of those, if only because it wouldn't work with my KVM, and I can't find a USB/DVI KVM (I think I've seen a 2 way USB KVM, but not a 4 or 8 like I have at home and work).
Curious. I wish they would have shipped it, or made it available thru some type of option.
I got pretty good at StarCraft: BroodWar. I could play 2 on 1 against the computer and win nearly every time. Playing three on one was pretty tough. However, a buddy and I could play 2 on 6 with regularity.
It wasn't until the last set of patches, that we couldn't beat several of our favorite maps on 2 on 6. Big Game Hunter was a great map to play, if you did it properly, you could play 2 on 6 as protos, which normally we couldn't do. Playing 2 on 6 on just about any other map was easy if you were patient and played as Terran and the primary base was at all defensible. The last set of patches made the AI very good about early pressure, and often pressure. It also improved it's ability to wave as one huge group, rather then having them with two or three of them in your base at one time. Two or three opponents in your base in the early game is doable. Five or Six was just a complete impossibility for us to deal with.
Especially if the computer wouldn't cheat (just give it extra money, or use skewed stats not available to the player). I don't mind of the computer has "infinite mice", although, anything that simulated the limitation of the number of commands that could be issued would be cool.
The other interesting thing, is that the AI got better with a faster machine. At various points, just upgrading machines without changing the game version would make things more challenging. Finally, on Big Game Hunter we pretty much proved to ourselves that Blizzard wasn't having the AI cheat in terms of picking the proper counter units. The trick to beating Big Game hunter as Protos was to build as many "Carriers" as you could. Put down as many of the Photon cannons as you could. Put down two sets of Upgrade buildings so you can do all of the upgrades in parallel. Defend your main base with proton cannons and other foot troops. When each of you has a dozen fully upgraded carriers, each of you hits a different opponent (Build shield rechargers on the edge of your base towards the side you are going to attack). You should be able to crush your opponent with relative ease. Both teams then pick a third opponent to attack crush them. Then it's just time to clean up. You should have pretty much destroyed everyones forces, go clean up the bases. However, if you used a Carrier to defend your base before you had a dozen of them, the computer would build the perfect counter unit (normally those really cheap small zerg fliers). As long as you hide the carriers in the back of your base, you'd just crush the computer. If not, they had a tendency to build too little anti-air units until it was too late. This basic strategy was figured out while we played it 3 on 5, until everyone got tired of me just essentially doing nothing for the first half of the game while they defended, followed by me beating 4 of the 5 opponents. The last patch we applied (either the 1.09 patch or the one after it if there was one), the computer waved so badly that couldn't keep enough proton cannons on the ground to defend your base.
The other thing I really wish that games makers allowed was the ability to script AI so you could essentially build your own AI scripts, so that the AI was exentisible without writting a DLL.
Hmmm, it's fairly obvious that anything that can create anonymity, can be used for bad intents things.
It's a natural consequence. I really like that I can encrypt my e-mail. It kinda sucks that organized crime (or the kneejerk options terrorists) can use that same technique to hide from justice.
I'd like it if I could travel without giving my name to the airlines. At the very least, deducing the any number of airline hi-jackings would be difficult.
I'd love it if I didn't have to show id in order to enter into my building. However, it would be much easier to rob.
I'd love it if I didn't have to put plates on my car, but that would making getting away with any number of crimes much easier(child abductions, and bank robberies come to mind).
I like that I don't have to show my passport during intrastate travel in the US, but if we had more checkpoints where people had to show identification, we'd probably catch more petty criminals, and keep people from escaping justice.
It would just as worrisome that criminals can get away with beating someone senseless by blurring their picture (in this case the obvious solution is to not equip security cameras with such features). Just because the criminals in his example happen to be police officers shows his inherient distrust of police officers (there are good cops and bad cops, just like most other professions, pardon the pun).
Every police officer I've ever met or dealt with has been cordial, professional, and seemingly a decent human being. Then again, I'm not a young black or latino living in L.A. I understand they have a different experience the I do with police.
Anything that provides anonymity to anyone inheriently has risks. At least in this case it sure sounds fair (everyone can use it, as opposed to something only to be issued to police officers). It provides a way for someone to do something that without fear of retribution (which sometimes is good, and sometimes is bad: whistle blowers good! Criminals bad!). So yes, sometimes the most obvious is the bad stuff. Personally, I've never had a problem with someone coming up and taking my picture on a public street. I'm not that interesting that someone feels compelled to take my picture.
Well, yes and no. I can go track down the documentation I've read about OpenBSD. That they find security flaws isn't shocking. They are quite good at that.
I've never followed the Apache debate further then to realize it exists (I snicker at the fact that you act like I don't know that OpenBSD doesn't ship with the stock Apache). However, there's a difference between a full on audit, and having patches (I got the impression OpenBSD examines critical portions of Apache, but not the complete line by line audit they do of their kernel). I have always been under the impression that some of the larger non-kernel pieces get looked at, but don't get the full on audit that the kernel does (I might just be dated on my knowledge of that, I'm fairly sure that was true at one point in time). I've followed a lot of the people in the Apache projects while I was active in the Subversion project. They are fairly serious about their security, and they are pretty knowledgeable. I couldn't sort out who was right in that argument.
However, most all of the fixes get pushed back to the Apache base source tree. I'd venture to guess that if it really was an exploitable problem, we'd have seen an actual honest to goodness exploit for it. There are still plenty of 1.3 Apache servers around, it's not like a blackhat or web-defacer wouldn't use it. So I have a tendencey to say that OpenBSD is being a bit over the top, but that's their perogative (and their well mandated stance, I'm sure the Apache people have a reasonable reason not to apply the patch. I've had personal interaction with a number of them, all highly intelligent, rational people). If there we real security problems, the Apache people would have no option but to fix the supposed problems. I'd always been under the impression that the OpenBSD guys wrote a completely non-portable fix (it's not like they don't do that on a fairly regular basis, look at OpenSSH, by default the source tree doesn't work on any other OS, you have to go hunt down the "ports" page to get the version that works on other Operating Systems).
Naturally I can't find the quote in context, but essentially it boils down to: "OpenBSD can only audit so much code, and we pick and choose what we audit". Now, it might be that I'm out of the loop enough on OpenBSD that they secure more code then I believe. I'm fairly sure they don't bother auditting their ports tree.
If you took a stock FreeBSD, OpenBSD and Linux install, and installed the minimal amounts of userspace that OpenBSD does, there's a negligable difference in the security differences. OpenBSD's biggest security advantage is that they are "secure by default". Not installing software you don't need. Not installing software that isn't well written is a very good security stance. It's really too bad RedHat doesn't do that more often, but my guess is that there's still a negligble difference in terms of practical risk. (All that said OpenBSD does the best job of any major Operating system I know of shipping secure defaults with a minimalistic, but still usable operating system).
There are a few things that OpenBSD does that are very security conscious but have nothing to do with the code audits. The stack protection. The NoExecute stuff. The crypto built in. They are very leading edge in that. However, from a practical perspective my guess is that there are few attacks on Apache/named/dhcp/ftpd/OpenSSH and whatnot that work on Linux that fail to work on OpenBSD, because OpenBSD modified Apache in a way the Apache project wouldn't accept patches back.
Oh, and give me the link to why pf is vastly superior to IP tables (call it academic interest. I know how to configure iptables and fail to see the inherient failings of it that pf overcomes).
Touche. If DHCP is auditted (I should have thought about that ISC would have it auditted along with BIND), that's great. However, my guess is that OpenBSD didn't do special audits, that didn't get pushed upstream (I know they have some for Apache that they can't pushed upstream, which I alluded to in my post).
Thus OpenBSD isn't inheriently more secure then Linux, NetBSD, or FreeBSD due to the particular audit you are discussing.
Even the OpenBSD guys are pretty clear on the fact that they have limited time and scope for auditting, and that their claim of no remote root exploits in X years (or 1 in 2 * X years or whatever it is they are up to now), doesn't mean an OpenBSD box can't be broken into. If all you are using it for is a straight up packet router, that's wonderful, nearly any OS should be capable of doing that incredibly securely. If you are using it for something that offers remote services. Those remote services are the crux of the problem. Those services more then likely haven't received a full on, all out audit by OpenBSD and Co. That was the main thrust of my point. I mean, OpenBSD finds bugs in auditted code on a fairly regular basis, so auditting isn't the end all be all that the post I was responding too seemed to believe it might be.
And if you believe what you just said makes any difference, you are being just as naive.
OpenBSD, as in the kernel itself, is fairly well auditted (I'm not sure about the coverage, but they do examine things closely from what I remember).
However, Apache isn't auditted. DHCP isn't auditted. The FTP server, I'm fairly sure isn't auditted. Nothing they don't actually write themselves. If you install an OpenBSD machine on the internet and actually turn on services, you'll have just as many security problems as anyone running Linux. OpenSSH has it's fair share of security problems (written by pretty much the same people who wrote OpenBSD). Although with priveledge separation it should have even fewer problems that are actually exploitable to become root.
While Apache does have some security patches applied to it that the stock Apache doesn't, that doesn't make it "Auditted". That means a handful of exploits have been found. I believe the Apache people just don't agree that there is a problem (I'm not clever enough to see who has the more reasonable point of view).
As soon as you start actually using OpenBSD to do anything that allows remote services, you are pretty much into the areas where you could have security problems just like anyone else.
OpenBSD does have some nifty patches to help mitigate certain types of attacks (The memory protection schemes that implement NoExecute on the stack, and some other ways you can mark a page in the VM system as no-execute or no-write).
However, that doesn't mean "OpenBSD is auditted and therefore secure". I'm absolutely confident that if I had shipped Linux for the last 8 years with as little configured to run out of the box as OpenBSD does, Linux could claim no remote root exploits too (the same is probably true of FreeBSD, NetBSD, OS X, Windows, and any other number of Operating systems). (Okay, the Windows claim might be a streach, because I believe there are certain ports that a very difficult to close, but the rest I'm fairly sure are true). The lack of any open ports at all makes it fairly trivial to not have any remote attacks. All you have to do is ensure that your network stacks don't do something stupid with a packet they are routing. Not terribly difficult. The fact that it ships with no services configured is very good. While it probably has a more secure kernel, most exploits out there in the world involve exploiting a user process that is running as root. As which point, you can own an OpenBSD machine as quickly and as easily as a FreeBSD, Linux, or NetBSD machine.
This isn't a court of law. Hearsay is a perfectly legitimate form of information (I've been told by people who have direct experince as live long SA's with a dozen years experience on a half dozen platforms). Just like, I can advise you that it's really stupid to build a wall without mortor. I've never laid a brick in my life. My Dad told me that. So while it's hearsay for me to tell you that a wall without mortor is a really bad idea, that doesn't make it any less accurate as a statement. It's not uncommon to have people re-tell information they have been told from people who had direct evidence. "Mr. Witness, did Mr. MurderSuspect tell you he committed the murders?", is a perfectly legitimate form of evidence against a murder suspect. My boss makes IT decisions based on hearsay all the time. I take review information from authoratitive sources, collect and collate that information. He takes information from someone who has no direct experience and bases decisions on it.
I do have plenty of experience with Solaris directly. I've used QNX, Solaris, Linux, FreeBSD as my primary development platform for various projects where I setup and configured the machines. I've run both x86 and SPARC Solaris. I've even made several bug fixes to code on AIX machines (never admin'ed them directly, but I did compile code on them).
The most memorable fix, is that on Solaris and AIX, how they handle casees in a "select" call is different. One of them sets the bit in the exceptfdset if you call mark a fd for reading that has already reached and read EOF. The other one doesn't. I believe it was AIX that set the exceptfdset bits, and Solaris that didn't.
Information from third parties is plenty valuable. I've personally read the various bug reports about the horrors of AIX w/ Oracle. I've been at the techincal when they tell you "Do X on all of your machines to make Oracle Run faster. Don't do X on a 4.2 or 4.3 version of AIX as it will corrupt your database." (In most instances, that would be Async I/O).
I've never used AIX, as I've been told by numerous independent sources, that Solaris and Linux are far "saner" in terms of Admin and being an SA for. I don't have any direct personal experince, because everyone tells me it's a bad experince.
I've read information about AIX when it comes to systems administration (specifically the instructions to installing various pieces of software that have specialized sections, or O'Reilly books that mention specific differences between AIX and Linux).
I never had a drop of Alchol in my life, but I can still tell you the health consequences it has on a person. Not everything has to be experienced first hand to be valuable as information.
Never smoked a cigeratte in my life either. I know lots of badness about what happens to people who do that.
In both cases, my Dad did those. They sure looked like a bad experience. So I've avoided them. In a similar set of experiences, everyone I know has advised me that while AIX can be made to work, you need a really good reason before they would pick it over Solaris, Linux or FreeBSD (generally the only reason, is integrate with other IBM hardware or software, or you inherited an existing AIX machine that is configured and running). In those cases, it made sense to go with AIX.
So while you can say it's "hearsay", a lot of wisdom and experience is contained in "hearsay". You need to be careful about who you trust, but the people who imparted this wisdom on me I do trust.
I'm an idiot. No one else need apply. People talk about Jan 1984, Mac. I think Superbowl Commercial. My fault. It's fairly rare to see the actually commerical itself. I know that Apple was pretty serious about never, ever using it more then once.
Slashdot Mirror
Here they claim it's been out since 1994.
Which puts it between at least 10, possibly 11 years now.
Not sure if that's authoratative, but it's at least a date I found via a quick google search.
It's been obsoleted by the PS2 for what a little over 4 years? (It came out during the Christmas shopping season of 2000 if I remember correctly).
However, it's not like a lot of titles are being released. It's not like the blockbuster games are being dual ported.
I'd see fewer problems with this, if the X-Box 2 was going to be backwards compatible with the X-Box. However, from everything I've read, it's nearly a technical impossibility to do that if what has been publically guessed about the X-Box is true (I'm not sure if Microsoft has officially said anything besides that ATI will be making the video cards). I'm not paying that much attention. I believe it's supposedly going to have a PPC chip of some time (possibly a Cell, which is PPC + an array of vector processing chips if I understand it correctly).
If they released backwards compat consoles ever 2-3 years, I wouldn't care. However, releasing non-compatible ones every 2-4 years is just insane from a consumers perspective. However, as Microsoft is just gettings it's feet wet, I could see why they are doing this (fix thier previous mistakes, and get a head of the game on the hardware cycle is probably a good idea from a business perspective, especially if they can finance the losses).
Kirby
They know printers. They know lots about printers, and printer languages. My guess is that they'll be thrilled to get an opportunity to hack another printer working. I know when I bought a printer that has "PS Support", it had a postscript driver in software that talked a propriatary protocol to the printer. They would have gladly written the output driver for it, but they didn't know how it worked.
Maybe if you know how it works, you'll be able to get them to do something with it.
Kirby
In the end, well, trains don't always travel the same speed, they don't always travel at the speed they are told to. Sometimes they break down. Sometimes the switches aren't thrown properly (so there really are two dimensions, possibly more), sometimes a train runs away. Some times a train is on a section of track its not supposed to be. Trains aren't a trivial problem (we actually had to write a simulation of this in college in a RTS class, you had to do the computations and throw the switches at the right time, or you had yourself a fairly serious collision).
With trains at least, by the time anyone can visually tell you this, all you can really do is jump off and save yourself. It's literally a million pound weapon of death, by the time anyone can see the problem it's over. Having a GPS system on the train would enable you to spot all sorts of upcoming problems with out having to communicate with anyone onboard.
Kirby
Considering it fell 50%, 4 years ago, and well, the fact that they don't issue options like they used to. That they haven't had a metoric rise, yes the last 4 years in fact prove that they can manage to run a company without a rise stock price. They are still a part of a major indexes, which some people feels gives them a solid buyer for any new stock they issue as part of the employee compensation plan (so they can issue stock to employees, the employees then dump the stock on the institutional investor, there is always a buyer, so the price is artificially bouyed because there is artificial demand due to investment strategies in index funds). There might be some truth that if they are removed from the index funds that things might work out a bit differently for them. However, that's not what he proposed. He stated specifically, that "the stock price stop rising".
If they shut down Microsoft R&D, I'm fairly sure the cash management is structured so that the company can run for several years (8-12 quarters). They say that's one of Bill Gate's obsessions. When money was tight way back at the beginning, it was always his goal to have the company finances such that they had at least a years worth of operating expenses on hand. If it's pure juvinile fantasy, pull the balance sheet. It shouldn't take but 2 numbers cited, plus a single division to figure out how many quarters they can actually operate. It's not like the math is particularly difficult:
Cash on hand / quarterly expenses = number of quarters that can be operated without revenue.
That's the number I believe you are discussing. Which I still think is longer they you claim.
The number I was responding to was this different, and I'm fairly sure that the recent past is a disproves the statement if it had been said 3 years ago. Over the last three years, I believe they have posted record profits, and that the last several quarters have gone extremely well due to X-Box and Halo related revenue.
Kirby
I'm not sure what their operating costs are (I believe it's something just insanely low once you remove their research costs). Yeah, I'm fairly confident they can stick around once the pyramid scheme collapses.
By the by, I've been told the the pyramid collapsed a while back. Anytime a company re-prices options, it's a fairly sure sign, that a) they are using stocks as a primary form of compensation, and b) that the jig is up, and no one will ever get wildly rich of options again. They repriced a ton of options after the bubble burst and the stock price plummeted 50% from the all time high. I knew several people who were recruited to work there pre-bubble days. Both of them turned down the jobs, as they were about 2/3rds the going rate. However, you got enough options to turn you into a millionare in 10 years assuming they stock price continued it's incredible price rise. Relatively high risk, high reward.
I've lost the link, but there was a pretty good economic analysis a while back that showed, essentially that new investors in Microsoft were paying Microsofts wages and a lot of their taxes by buying up all the stock that got dumped into the market as converted options. The interesting part was that they ended up showing that Microsoft would lose a billion dollars a year if they paid the money out that employees got as options. The thing about that is, that I don't think Microsoft will make nearly as many of their employees millionaires as they used to. So I don't believe they'll lose all that money. I think they are right, that Microsoft played the market to the hilt. The market thought it was getting in on a good deal, when really Microsoft was using the markets capital to compensate employees.
They are still doing it to an extent by giving away actual stock. However, my guess is that is probably a great deal less lucrative then options were back in the day.
Kirby
You save space (which is important in CoLo pricing). You use resources more efficiently. When one one of the other instances is relatively idle, a busy instance can use up those CPU cycles (not true with separate hardware).
I'm told, that if it doesn't exist, it will at some point in the future, that you can migrate instances from VMWare install to VMWare install while it's running. Thus you can migrate running instances from Machine A to Machine B. Down hardware A, do whatever upgrades are needed under the hood, migrate the instances back.
It's a relatively novel way to compete with LPAR's (or whatever the Mainframe speak is for that).
It's very, very impressive technology from what I've been told. Using it in conjunction with racks full of blade servers I'm told is the way to run an incredibly high density setup in a very flexible way and reliable way. I'd really like to try it at some point, if only because we have machines that sit 95% idle. I'd like to combine 3-4 of them onto a single machine (especially for services where they are simple to implement redundant backups: LDAP, SMTP, DNS, DHCP). Just setup a second machine doing pretty much exactly the same instances in the backup mode. It'd free up a ton of hardware that right now we are tying up so that the failure of one machine doesn't affect the rest of the services.
Kirby
You can have an innovative stone block bridge design. You know, the same things the Romans built 2000 years ago. You can still innovate on them.
innovation
n.
Innovation inside of the VM subsystem of the kernel happens. It's esoteric and 99% of all people don't care, know or see it. However, if you are "introducing something new" into the VM subsystem, then it surely fits the definition of "innovation".
I have no idea what Rendevous is, or what it does. But I can give you an example of C library that is innovative. Readline written by GNU. It is innovative. To the best of my knowledge they were the first group to introduce such a beast, and to the best of my knowledge it is still fairly unique. It's a straightforward library that you can link into any interactive program where a person might edit a single line of text. It automatically gives you keymappings, history, and all kinds of other goop. So any application I use, that uses it has pretty much the same interface as far as I'm concerned (gdb and bash are the two applications I know I use it with all the time, I'm sure there are others). I like to configure it so I can use vi style commands into it, to speed commandline editting.
Innovation isn't strictly limited to a particular level of implementation. Innovation could happen in the processing of toliet waste water. It's still innovation. So I'm curious about how you feel an innovation can't happen in a C++ library. I mean, mozilla is nothing but a bunch of interconnected C++ libraries. I'm fairly sure there's lots of innovation in there somewhere. It's got to be contained in the C++ libraries somewhere.
Kirby
Now, I know why the machines we're so much cheaper. It was a learning experince. We ended up throwing an old sound card, and a cheap recycled video card into the machine. It worked fine after that.
Kirby
While you still have to be able to deliver after customers contact you, reducing the cost of aquiring customers is a serious boon to you. Getting to the front couple of pages of Google, is probably an incredibly cost effective way of doing that.
Keeping people on your site, normally involves having what they are looking for. However, you need to get them to visit your site first. The IT industry is littered with the casualities of better technology that failed because the lesser technology has better marketing.
Kirby
Second. The GPL isn't a patent license, it's a copyright license. US Patents protect the concept of how something is done. Copyright protects a particular expression of a concept.
You can take a look at a GPL'ed implementation, and then re-implement your own from scratch using the same concepts. If you want to be really good about it, have someone else look at the GPL'ed implementation. They write specification for how it works, and then you write an implementation from that specification. Thank you Compaq for setting that legal precedence!
Kirby
In order to deliver in the cover of darkness, rent several vans. Remember the logistical problem of this is to fill the volume of a building. You do realize that say a moving van full ballons will probably fill an 18ft by 10 ft by 10 ft space. So if it is a one story building, that is say 1000sq with 10 foot ceilings, you'll need something on the order of 5-6 moving vans full of (popcorn/ballons). To fill it ceiling to floor right for the entire building.
The bonus to this, is if you can find a rental place (in America, it's relatively simple to find, but I don't know about the UK (where I believe you live from the link I was reading) to rent a helim thing to inflate roughly half of them (ceilings) and then an air compressor to inflate the other half (floor).
With an air compressor (a $90USD), or a vacuum you can use as a blower, if making a lot of noise is acceptable, you can just stand outside the house and fill the ballons. The bigger the balloons, the fewer you need. It might be advantagous to find several just huge ballons (think 4-6 feet in diameter) to just fill large volumes to be inflated on site, while pre-inflating the smaller filler ballons to pack the place fairly tight. to make very difficult to move around in. The other problem with this plan, is that anything that is fragile and at a below the height you plan on putting the ballons/popcorn to might get knocked over during the cleanup.
This idea, isn't an original of mine. Some of my older sisters friends did this to my sister for her birthday. They filled her room to about 4 foot high with ballons in her room.
Kirby
I purchase RedHat licenses for everything that is in the DMZ, or runs software that requires RedHat Enterprise Linux for support (think Oracle Databases).
Then I use Whitebox Linux for everything else. It's pretty much exactly the same as RedHat (you can pick another RHEL rebuild if you want, CentOS and Whitebox Linux are my two favorites). Whitebox can have problems from time to time, because it's a one man show. CentOS looks nice, but it sounds like the mailing lists are used less, and the web boards more for discussion and help (I've never participated, but that's the a complaint I've seen on WhiteBox lists about CentOS). I like e-mail lists for help/support. Call me silly. While web boards are nice for random discussions, I'd much rather review e-mail for technical support (both on the giving and receiving end).
I use that for the desktop. Other then, it's a bit RAM hungry, it's fine for a desktop for most people (the lack of a good MP3 player might bother most, but I play oggs, so I'm good with it). You need more then 128MB of RAM to run OpenOffice on it at a reasonable speed. (I was running a PIII-500 w/ 384MB of RAM and it was acceptable, with a new P4 w/ 128MB of RAM it was unbearably slow running Mozilla and OpenOffice at the same time. I put a 1GB of RAM in and now it's wonderful).
In the end, it means I can run almost exactly the same OS at home that I do at work. It's industrial strength, and all of the expertise I build up using it, is going towards one of the two distro's that all major software vendors support. I don't know of any Suse "rebuilds", otherwise I might recommend those.
Kirby
Here is the original press release. However, it doesn't seem it implicate what you are saying. I think they are releasing it to anyone who uses the CDDL (at least according the the PR). I'd also be interested in their legal definition of "OpenSolaris developer process". My guess is, either the "OpenSolaris developer process" will end up being so tightly defined that no one can release a version of Solaris without Sun's permission, or it'll be so open ended such that there's a hole that I can end up implementing the patents in any CDDL licensed project. However, without the legal text, I can't be sure. Maybe they are more clever then I think and came up with a definition that can find a middle ground between those two.
I've got no problems with them licensing the patents such that only CDDL projects can use them. While it will create the great divide in terms of mixing and matching source code, it's their right. However, if what they are doing is saying, you can only implement them in such a way that we'll end up with control over you via patent law instead of copyright law, that seems completely underhanded. It also seems to be completely counter to the image they are attempting to spin. I never thought of Sun as that underhanded (they are a coporation, and generally have interests that are somewhat contradictory to mine, but never thought of them as doing anything that qualifies as "evil genious"). Oh well.
Kirby
I'm fairly confident that XFree86 would have ended up forked without the change in license. Too many people were too unhappy with XFree86 for a long time. There was a fork waiting in the wings, pre the license change. If you read Slashdot, you would have read about it. Heck, they posted notes from conference calls discussing how to change XFree86. People kept trying to get Keithp to come back to the project, and technically speaking, I don't think he forked it pre-license change. However, he was trying to setup how to run the project before bothering with the Fork, so it wouldn't die upon starting.
The problem with XFree86 wasn't the license (that just made it bloody obvious to everyone around that XFree86 no longer fit their definition of "Free", thus they really wanted to use the Fork) that the change had to happen. Meanwhile, they just happen to get to setup a sane way to run a project that is happy go lucky with how OpenSource projects are run best.
The fundamental problems are the high-handedness, and well, pretty much non-responsive nature of the people handling the XFree86 project. Fundamentally, you'll find that XFree86 is really disconnected from the interests of the people using the project.
Read up on fundamentally what I'm talking about. I'm not talking out my ass honest. Here's a pretty decent link to the a Wiki Entry
Kirby
It was because the XFree86 people had someone willing to do the work, but they wouldn't let him commit patches because they didn't agree with his design changes. He was one of the most active, and progressive people working on XFree86 at the time.
Finally a mention that 98% of all projects on sourceforges aren't forks (if each project has 60 forks, then only 2% are original projects). 98% of them, might be a duplication of effort, but strictly speaking, not a fork.
Kirby
Linus should run some form of shared access, where he has lots of people who can commit code directly the the Linux source tree without Linus having to see it.
Presumable a more "FreeBSD" or Debian like system. FreeBSD is a lot closer to a meritocracy then Linux's "Benevolent dictator" system. FreeBSD has a core group of whom several (5-10) people can get access to the actual source repository. Supposedly FreeBSD is fairly elitiest and tight knit (think XFree86, they have roughly the same governance model as FreeBSD, but XFree86 sounds like a lot more of an old boys club them FreeBSD is).
Debian has a system where they are fairly democratic, and have a process where by you can initiate referendums to vote on a change you feel is important enough (generally never done over source code, but has been done over which version of the Linux kernel to ship, and what types of stuff has to be stripped from the Linux kernel before it meets Debian's definition of "Free").
Linus is a dictator of the stock Linux kernel. However, there are so many forks out there of different trees, where lots of people have access to those trees that it's relatively silly to discuss. The other interesting aspect, is I get the distinct impression that in lots of areas of the kernel, Linus does implicitly let people just randomly apply patches. If you are one of the people he trusts working on an area he feels you know best about, he just applies your patches with minimal if any review. You don't get access to his primary sources to do the patch yourself, but you get a relatively unfettered access to the areas you know about. Which is sorta nice, as well, you don't see the kinds of spats that spawned OpenBSD (CVS revision wars, where people undo others work because they disagree, and they have access). When there is a single arbitor of what gets access, it never seems like there are people of two minds in control of the source.
It's like the age old argument, that a Monarchy is the best form of government assuming you have a good and fair king. It's also the worst kind of government if you have a despot. Unfortunatly hereditary monarchies generate a lot more bad kings then good ones.
Kirby
Plus, most USB mice can be converted to PS/2, however, I haven't had luck the other way around Converting from PS/2 to USB, which is what I'd have to do to connect my KVM to a USB only MoBo.
The real irony here, is that if the guy said he wanted a miniITX (or full sized) that has serial only capable BIOS, I'd be all over it. I really want a MoBo that can run completely headless and hook it up to a serial terminal. Then I wouldn't have to worry about 90% of the goofiness of USB, PS/2 or KVM's. Then I could have one just sweet DVI/USB/no legacy fancy smancy machine. When I want access to any of the 5 or 6 server style machines, I just ssh into the machine. If I managed to break the machine, then ssh to the terminal server and fix the network. Then I wouldn't need a KVM. The only time I ever switch on my KVM from my primary desktop is to solve a problem that can't be solved over the network, because I broke the network configuration.
While I can do this via Linux and putting the console on a serial port, if I ever need the BIOS, I still have to hook up a keyboard, monitor, and moused directly to it. Which is a hassle. I've always been amused that AMI/Phoenix/BiosStar don't integreate serial only into their primary bios configuration.
Kirby
I've never gone looking for them, but I've ever seen one in all of the years I've been looking for MoBo's. The last time I saw one without PS/2 was probably a 386. I think Dell was coming out with a line of machines that did that (in the end, I think they dropped the floppy, but the rest were on board).
We have some IBM servers at work that have a very odd connector that you have to by a breakout cable to turn into PS/2 keyboard and mice, and regular video. That's really just a space saving (it's also a KVM built into each machine, as you can chain them with a non-breakout cable, at the end of the chain you attach a breakout cable).
So while I see your beef with the mini-ATX machines, it's not like what you are looking for is terribly common for a x86 motherboard at all. I wouldn't expect mini-ATX to lead in that category.
Personally, I'd hate to end up with one of those, if only because it wouldn't work with my KVM, and I can't find a USB/DVI KVM (I think I've seen a 2 way USB KVM, but not a 4 or 8 like I have at home and work).
Kirby
I got pretty good at StarCraft: BroodWar. I could play 2 on 1 against the computer and win nearly every time. Playing three on one was pretty tough. However, a buddy and I could play 2 on 6 with regularity.
It wasn't until the last set of patches, that we couldn't beat several of our favorite maps on 2 on 6. Big Game Hunter was a great map to play, if you did it properly, you could play 2 on 6 as protos, which normally we couldn't do. Playing 2 on 6 on just about any other map was easy if you were patient and played as Terran and the primary base was at all defensible. The last set of patches made the AI very good about early pressure, and often pressure. It also improved it's ability to wave as one huge group, rather then having them with two or three of them in your base at one time. Two or three opponents in your base in the early game is doable. Five or Six was just a complete impossibility for us to deal with.
Especially if the computer wouldn't cheat (just give it extra money, or use skewed stats not available to the player). I don't mind of the computer has "infinite mice", although, anything that simulated the limitation of the number of commands that could be issued would be cool.
The other interesting thing, is that the AI got better with a faster machine. At various points, just upgrading machines without changing the game version would make things more challenging. Finally, on Big Game Hunter we pretty much proved to ourselves that Blizzard wasn't having the AI cheat in terms of picking the proper counter units. The trick to beating Big Game hunter as Protos was to build as many "Carriers" as you could. Put down as many of the Photon cannons as you could. Put down two sets of Upgrade buildings so you can do all of the upgrades in parallel. Defend your main base with proton cannons and other foot troops. When each of you has a dozen fully upgraded carriers, each of you hits a different opponent (Build shield rechargers on the edge of your base towards the side you are going to attack). You should be able to crush your opponent with relative ease. Both teams then pick a third opponent to attack crush them. Then it's just time to clean up. You should have pretty much destroyed everyones forces, go clean up the bases. However, if you used a Carrier to defend your base before you had a dozen of them, the computer would build the perfect counter unit (normally those really cheap small zerg fliers). As long as you hide the carriers in the back of your base, you'd just crush the computer. If not, they had a tendency to build too little anti-air units until it was too late. This basic strategy was figured out while we played it 3 on 5, until everyone got tired of me just essentially doing nothing for the first half of the game while they defended, followed by me beating 4 of the 5 opponents. The last patch we applied (either the 1.09 patch or the one after it if there was one), the computer waved so badly that couldn't keep enough proton cannons on the ground to defend your base.
The other thing I really wish that games makers allowed was the ability to script AI so you could essentially build your own AI scripts, so that the AI was exentisible without writting a DLL.
Kirby
It's a natural consequence. I really like that I can encrypt my e-mail. It kinda sucks that organized crime (or the kneejerk options terrorists) can use that same technique to hide from justice.
I'd like it if I could travel without giving my name to the airlines. At the very least, deducing the any number of airline hi-jackings would be difficult.
I'd love it if I didn't have to show id in order to enter into my building. However, it would be much easier to rob.
I'd love it if I didn't have to put plates on my car, but that would making getting away with any number of crimes much easier(child abductions, and bank robberies come to mind).
I like that I don't have to show my passport during intrastate travel in the US, but if we had more checkpoints where people had to show identification, we'd probably catch more petty criminals, and keep people from escaping justice.
It would just as worrisome that criminals can get away with beating someone senseless by blurring their picture (in this case the obvious solution is to not equip security cameras with such features). Just because the criminals in his example happen to be police officers shows his inherient distrust of police officers (there are good cops and bad cops, just like most other professions, pardon the pun).
Every police officer I've ever met or dealt with has been cordial, professional, and seemingly a decent human being. Then again, I'm not a young black or latino living in L.A. I understand they have a different experience the I do with police.
Anything that provides anonymity to anyone inheriently has risks. At least in this case it sure sounds fair (everyone can use it, as opposed to something only to be issued to police officers). It provides a way for someone to do something that without fear of retribution (which sometimes is good, and sometimes is bad: whistle blowers good! Criminals bad!). So yes, sometimes the most obvious is the bad stuff. Personally, I've never had a problem with someone coming up and taking my picture on a public street. I'm not that interesting that someone feels compelled to take my picture.
Kirby
I've never followed the Apache debate further then to realize it exists (I snicker at the fact that you act like I don't know that OpenBSD doesn't ship with the stock Apache). However, there's a difference between a full on audit, and having patches (I got the impression OpenBSD examines critical portions of Apache, but not the complete line by line audit they do of their kernel). I have always been under the impression that some of the larger non-kernel pieces get looked at, but don't get the full on audit that the kernel does (I might just be dated on my knowledge of that, I'm fairly sure that was true at one point in time). I've followed a lot of the people in the Apache projects while I was active in the Subversion project. They are fairly serious about their security, and they are pretty knowledgeable. I couldn't sort out who was right in that argument.
However, most all of the fixes get pushed back to the Apache base source tree. I'd venture to guess that if it really was an exploitable problem, we'd have seen an actual honest to goodness exploit for it. There are still plenty of 1.3 Apache servers around, it's not like a blackhat or web-defacer wouldn't use it. So I have a tendencey to say that OpenBSD is being a bit over the top, but that's their perogative (and their well mandated stance, I'm sure the Apache people have a reasonable reason not to apply the patch. I've had personal interaction with a number of them, all highly intelligent, rational people). If there we real security problems, the Apache people would have no option but to fix the supposed problems. I'd always been under the impression that the OpenBSD guys wrote a completely non-portable fix (it's not like they don't do that on a fairly regular basis, look at OpenSSH, by default the source tree doesn't work on any other OS, you have to go hunt down the "ports" page to get the version that works on other Operating Systems).
Naturally I can't find the quote in context, but essentially it boils down to: "OpenBSD can only audit so much code, and we pick and choose what we audit". Now, it might be that I'm out of the loop enough on OpenBSD that they secure more code then I believe. I'm fairly sure they don't bother auditting their ports tree.
If you took a stock FreeBSD, OpenBSD and Linux install, and installed the minimal amounts of userspace that OpenBSD does, there's a negligable difference in the security differences. OpenBSD's biggest security advantage is that they are "secure by default". Not installing software you don't need. Not installing software that isn't well written is a very good security stance. It's really too bad RedHat doesn't do that more often, but my guess is that there's still a negligble difference in terms of practical risk. (All that said OpenBSD does the best job of any major Operating system I know of shipping secure defaults with a minimalistic, but still usable operating system).
There are a few things that OpenBSD does that are very security conscious but have nothing to do with the code audits. The stack protection. The NoExecute stuff. The crypto built in. They are very leading edge in that. However, from a practical perspective my guess is that there are few attacks on Apache/named/dhcp/ftpd/OpenSSH and whatnot that work on Linux that fail to work on OpenBSD, because OpenBSD modified Apache in a way the Apache project wouldn't accept patches back.
Oh, and give me the link to why pf is vastly superior to IP tables (call it academic interest. I know how to configure iptables and fail to see the inherient failings of it that pf overcomes).
Kirby
Thus OpenBSD isn't inheriently more secure then Linux, NetBSD, or FreeBSD due to the particular audit you are discussing.
Even the OpenBSD guys are pretty clear on the fact that they have limited time and scope for auditting, and that their claim of no remote root exploits in X years (or 1 in 2 * X years or whatever it is they are up to now), doesn't mean an OpenBSD box can't be broken into. If all you are using it for is a straight up packet router, that's wonderful, nearly any OS should be capable of doing that incredibly securely. If you are using it for something that offers remote services. Those remote services are the crux of the problem. Those services more then likely haven't received a full on, all out audit by OpenBSD and Co. That was the main thrust of my point. I mean, OpenBSD finds bugs in auditted code on a fairly regular basis, so auditting isn't the end all be all that the post I was responding too seemed to believe it might be.
Kirby
OpenBSD, as in the kernel itself, is fairly well auditted (I'm not sure about the coverage, but they do examine things closely from what I remember).
However, Apache isn't auditted. DHCP isn't auditted. The FTP server, I'm fairly sure isn't auditted. Nothing they don't actually write themselves. If you install an OpenBSD machine on the internet and actually turn on services, you'll have just as many security problems as anyone running Linux. OpenSSH has it's fair share of security problems (written by pretty much the same people who wrote OpenBSD). Although with priveledge separation it should have even fewer problems that are actually exploitable to become root.
While Apache does have some security patches applied to it that the stock Apache doesn't, that doesn't make it "Auditted". That means a handful of exploits have been found. I believe the Apache people just don't agree that there is a problem (I'm not clever enough to see who has the more reasonable point of view).
As soon as you start actually using OpenBSD to do anything that allows remote services, you are pretty much into the areas where you could have security problems just like anyone else.
OpenBSD does have some nifty patches to help mitigate certain types of attacks (The memory protection schemes that implement NoExecute on the stack, and some other ways you can mark a page in the VM system as no-execute or no-write).
However, that doesn't mean "OpenBSD is auditted and therefore secure". I'm absolutely confident that if I had shipped Linux for the last 8 years with as little configured to run out of the box as OpenBSD does, Linux could claim no remote root exploits too (the same is probably true of FreeBSD, NetBSD, OS X, Windows, and any other number of Operating systems). (Okay, the Windows claim might be a streach, because I believe there are certain ports that a very difficult to close, but the rest I'm fairly sure are true). The lack of any open ports at all makes it fairly trivial to not have any remote attacks. All you have to do is ensure that your network stacks don't do something stupid with a packet they are routing. Not terribly difficult. The fact that it ships with no services configured is very good. While it probably has a more secure kernel, most exploits out there in the world involve exploiting a user process that is running as root. As which point, you can own an OpenBSD machine as quickly and as easily as a FreeBSD, Linux, or NetBSD machine.
Kirby
I do have plenty of experience with Solaris directly. I've used QNX, Solaris, Linux, FreeBSD as my primary development platform for various projects where I setup and configured the machines. I've run both x86 and SPARC Solaris. I've even made several bug fixes to code on AIX machines (never admin'ed them directly, but I did compile code on them).
The most memorable fix, is that on Solaris and AIX, how they handle casees in a "select" call is different. One of them sets the bit in the exceptfdset if you call mark a fd for reading that has already reached and read EOF. The other one doesn't. I believe it was AIX that set the exceptfdset bits, and Solaris that didn't.
Information from third parties is plenty valuable. I've personally read the various bug reports about the horrors of AIX w/ Oracle. I've been at the techincal when they tell you "Do X on all of your machines to make Oracle Run faster. Don't do X on a 4.2 or 4.3 version of AIX as it will corrupt your database." (In most instances, that would be Async I/O).
I've never used AIX, as I've been told by numerous independent sources, that Solaris and Linux are far "saner" in terms of Admin and being an SA for. I don't have any direct personal experince, because everyone tells me it's a bad experince. I've read information about AIX when it comes to systems administration (specifically the instructions to installing various pieces of software that have specialized sections, or O'Reilly books that mention specific differences between AIX and Linux).
I never had a drop of Alchol in my life, but I can still tell you the health consequences it has on a person. Not everything has to be experienced first hand to be valuable as information.
Never smoked a cigeratte in my life either. I know lots of badness about what happens to people who do that.
In both cases, my Dad did those. They sure looked like a bad experience. So I've avoided them. In a similar set of experiences, everyone I know has advised me that while AIX can be made to work, you need a really good reason before they would pick it over Solaris, Linux or FreeBSD (generally the only reason, is integrate with other IBM hardware or software, or you inherited an existing AIX machine that is configured and running). In those cases, it made sense to go with AIX.
So while you can say it's "hearsay", a lot of wisdom and experience is contained in "hearsay". You need to be careful about who you trust, but the people who imparted this wisdom on me I do trust.
Kirby
Kirby