AOL has spent its karma. They're on a strong, downward spiral. This in spite of the fact that one of the largest, coolest things to come out was largely funded by them. (Mozilla) Here's what they need to do to get back on track:
1) Fire everybody in marketing. Re-hire a whole new marketing arm from the likes of Earthlink.
2) Retrain their call support centers. Make it possible to quit AOL.
3) Make a new corporate motto "don't be evil". Follow it. (Google seems to be weakening its resolve)
4) Hire a bunch of highly qualified engineers, let them play, and let them decide what to sell.
If they do the above, they have a ghost of a chance. If they don't, they're fodder for a buy out within another year or two. They might do OK if they merge with the likes of Earthlink - but not as equals. Only where Earthlink takes control and they do it the right way.
Why is it that we keep hearing about this kind of advancement "to be available in five to ten years", and yet the storage capacity of batteries has been stagnated for at least that long?
Your Sig:
Wake up - the future is arriving faster than you think.
Isn't that a bit, eh, contradictory?
BTW: batteries are DEFINITELY improving. Remember rechargables a la 1980, these dumb nicad thingies that never really worked all that well? Compare that to today's NiMH batteries that outperform many alkalines, and are also rechargeable.
I'm pretty sure the idea is that passive cooling is preferable to active cooling - less noise, less power consumption and no moving part to fail.
And, in my experience, any video card that uses active cooling becomes a "passive cooling" system in a year or two of use, anyway. Fan bearings go bad, the fan quits, and I don't even notice until I'm doing something unrelated. To keep things cooler, I usually remove the fan and use the card with just the heatsink.
It's happened a number of times over the years, including my Voodoo II, my Riva TNT2, my ATI Radeon 8500, and my ATI Radeon 97xx. Yes, the fan is DEAD, with no apparent loss of stability or performance.
So why is that fan there, again? Somehow, they think it's important, and even when caked with 1.5" of dust, with a dead fan, the card just keeps working.
I would hate to watch the show with someone who couldn't stand up to Jamie, Adam does this well and thats why the show works so well. If Jamie was allowed complete control everytime, it would be boring.
I agree. Jamie is a great engineer with an intelligent and accepting, if dour personna, while Adam makes the great engineering more tasteful by interjecting lots of humor.
But I love the show! As a homeschool father, I heavily restrict TV and video games for the fall and winter seasons, but for a few shows. Mythbusters is among my favorite - such a spirit of experimentation and discovery!
I seem to recall that 200 MHz and an 8 Gig hard drive was top of the line, some time during the 90's. Such computers seemed to handle word processing, web browsing, email, etc. just fine. What would those components cost now? Not very much because of miniaturization. A $125 computer should be no big deal right?
Don't forget the concept of the minimum cost of production.
It doesn't actually cost (much) more to produce a 300 GB HDD than an 8 GB HDD. You have the same basic amount of aluminum, wire, circuit boards, chips, solder joints, IDE connectors,etc in either case. It occupies the same amount of space, and still needs to be assembled, checked, packed, shipped, and accounted for.
All these things do not change regardless of the HDD capacity. At a certain point, these costs overtake the profit involved in manufacturing, so that it no longer even makes sense to manufacture the drive any longer. That today is somewheres around $75-100 per drive. (retail) This cost is dropping, and has continued to drop, but it's still there.
This is why a $300 computer today blows the doors off a computer built 5 years ago for $1000, but the 5 year old computer cannot be produced today for its relative value. (say, $125)
What really friggin annoyed me is that I have a shiny, new cell phone, with a universal plug on the bottom. The charger plugs into that plug, and so does the USB cable. So I can't use the USB interface at the same time as I charge the phone with the AC adapter.
BUT IT WON'T CHARGE THE PHONE WITH A USB CABLE!
What kind of retarded corporate brain damage is that? It would be SO NICE if I could plug my phone into my laptop and charge the phone while I'm plugged in at a starbucks.
Friggen Motorola... Otherwise a decent, tri-mode phone The Motorola V325
What then would be the next big thing to top that?
Good question. Who forsaw the Internet?
By definition, the "next big thing" can't be predicted. It's that new thing, that revolutionary way of doing things that could not be predicted but changed the world anyway. How many hunter/gatherers predicted farming? Who could have predicted that the breakup of Bell Telephones in the 1970s would revolutionize interpersonal communications with the Internet?
The next big thing is coming. It probably is already in its infancy. And it's headed to a store near you.
Except that you can - sure people can build pirate radio equipment, but they can also rob houses. (just as illegal) People choose not to for various reasons - legallity being one of them.
When will people learn you can't control, regulate or do much of anything with the internet?
They never will, because it's not true.
What's that you say?
At its heart, the Internet is simply a form of communication. All other forms of communication are regulated, why wouldn't the Internet? The fact that it's new doesn't mean that it's un-regulatable so much as the powers that be haven't regulated it... yet.
Give it time. And then the "next big thing" will come along, and the Internet will be no more interesting than a ham radio today.
Your stupid straw man that you can't tell if it's Gnome or KDE is just that, a stupid straw man.
Ok.
Write a program for END USERS - not tech weenies administering a server. I'm talking about accountants, teachers, lawyers, doctors, and the like. Sell it. Write an installer for it, that works with RedHat, Novell, SUSE, Ubuntu, and Debian - recent versions sold in the past 5 years. And since Linux is what we're talking about, it can't be web-based.
Look, they could issue (for $100? or how much it costs...
Ok. 300 Million people in the USA. Times $100. That's $30 BILLION dollars. So much for cheap.
to people devices which are able to sign with a private key a short string of digits (16? 20?) that they dictate to you over the phone. You dictate back the 20 digits of the signature.
Ever enter a WEP key? It's 26 letters long. I have to retype one at LEAST 2 or 3 times TWICE in order to get it to work, when I have the key printed right in front of me. Do you REALLY think that's going to work reliably over the phone?
No complication, no computer needed.
Eh, let's see. We're going to relay a 20-character random text key twice over the phone, in and out of a $100 computing device. How is this either one of "No complication" or "no computer needed" !?!?!? What is that $100 thingy if not a limited-function computer?
What happens if you lose your $100 thingy?
Ultimately secure? Not.
Meaning, it isn't even a particularly good assurance of what you're after.
But anything like this would be VASTLY better than relying on the same 9-digit fixed number (the SSN) that appears in cleartext on every kind of document, and of which there are hundreds of copies lying around in offices around the country, from banks to insurance companies to medical offices to schools to universities to... you get the idea.
The problem is that you are trying to solve a social problem with a technical solution. You can't do that. No amount of technology usage would eliminate crime. Your solutions is simply too complicated and expensive to work well. Furthermore, it doesn't fail gracefully. Somebody gets your $100 thingie, and they suddenly can do whatever they want with your bank accounts and whatnot.
I STRONGLY recommend that you read some of Bruce Schnier's work. He started out like you - thoroughly convinced that the proper use of encryption could solve all of society's security ills, through his best-selling book "Applied Cryptography".
But then, the real world showed him how he was simply wrong. He was smart enough to swallow his pride and learn his lessons, and he's subsequently become one of the worlds leading experts on system security. Some of his best works include "Secrets and Lies", and his most recent: "Beyond Fear".
Give it a chance. You could make a 6-figure career by applying his principles!
If I am supporting application software on a GNU/Linux system, presumably because I am under contract to do so, I will specify the supported system(s). "We support RedHat AS 3.0 and above" for example.
When's the last time Roxio was "under a contract" to support their application software on ANY system for their Easy CD-Creator product? Never? I didn't think so.
Also, I don't believe users install an operating system without knowing which one it is. If they do, all bets are off.
You may do well to remember this little tidbit:
"Tis far better to keep your mouth shut and appear that you may not know what you are talking about, than to open it and remove all doubt".
The point is just that. All bets are off. Supporting desktop software on Linux is next to impossible.
And if I am supporting application software and the user can't tell me what the operating system is, I am well within my rights to say "I can't help you unless you tell me exactly what operating system you're running".
I'll just be glad that you aren't running my company! If you said this to your users, you'd go out of business rather quickly!
If you assume that the people who use your product are the most incompetent, utterly, incoherently, mind-numbingly ignorant beings to ever grace a keyboard, and you realize that the users of your software are actually quite a bit dumber than that, you might begin to get an approximation of just how bad reality is.
They'll ask for a manual until they get one. But they won't read it. They'll ask for help files, and you'll provide them. And they won't look at them. They'll ask you to warn them before they do something stupid. And they'll click OK despite the siren, flashing icons, and red background on the warning message, without even reading it. They'll even deny that the message was there in the first place.
Worse still, you'll have people who refuse to read these messages, behaving as though they were truly illiterate, even though they are teachers teaching a college English class!
You just don't have a clue. That doesn't make you bad, just clueless.
Linux is a great server operating system. It does amazing in the embedded space, as well. It's rock solid, quite secure, and is a dream come true if you are a programmer. But it is NOT a consumer operating system.
Unix begat OSX, so the potential exists. But it ain't there yet!
And all the while, better tools for identifications are widely available. I could identify myself to my bank simply by sending them a PGP-signed email: all that this requires of me is to click on the "sign it" button in Thunderbird - and I get incredibly better security than monkeying around with SSNs.
Yes, and no. You get better security, as long as your system isn't trojaned, wormed, or compromised. (And no, running Linux or OSX doesn't make you immune to these problems, though it helps) And so long as a multitude of other factors are considered. Such as:
1) Does your private key reflect sufficient randomness?
2) Does the 1-way function used to generate your private key have a "back door" making for trivial penetration?
3) Is your private key sufficiently private?
4) Is your bank USING PGP to authenticate?
5) Is THEIR private key really private? (If not, there's room for a man-in-the-middle attack)
But, even if those issues didn't exist, this solution simply doesn't scale well. What about people who don't have computers? What about people who can barely turn them on? What about people who are illiterate? What about people who don't speak english? How do you make sure that this works when the power is out?
And, if you think phishing is a problem now, boy, just wait until word gets around that private keys are such a big deal!
Our legislation, and public services, are late some 20 years regarding identity management. The scandal is that they are not brought up to date faster, not that some people are selling email footers that we send around for free.
A great sound bite. Unfortunately, it's just not true. You haven't presented a solution that works well, is cheap, widely understandable, fails gracefully, and is in the reach of the average (non-techie) Joe.
What solution presents all of these?
Certainly not your PGP "Web of Trust".
I'm a techie-type, who tends towards paranoia in security, and I've never set it up. It simply offers no real value. Hardly anybody else uses it, and if they did, they wouldn't care about the signed email. Realistically, nobody's going to say "Yes, I knowed it was you, because the Email was SIGNED!!!".
If somebody spoofs an email, it's pretty easy to look in the headers to identify that it wasn't me, and nobody has ever spoofed me to my detriment. Nor do I know anybody who's been so spoofed to their detriment, either. I've seen SPAM go out with forged from: addresses, but nobody believes that the penis pillz offer actually came from that person. Additionally, even if you encrypt an email so that only the recipient can see it, that recipient is then free to forward your message (without encryption) to whomever they like. So, your email is still a matter of public record. The rules of the game are simple: don't send an email that would be a problem if forwarded.
So what was this PGP thingy supposed to do for me, again?
If a user doesn't know what distro they're using, then they wouldn't know what version of Windows they were using either.
But, on either Windows or Macintosh, it's EASY to tell them how to find out what flavor they're running. That is not true for *nix. And so long as this is the case, the "Year of the Linux Desktop" will never happen.
There's no such consistency when using Linux. Are they running KDE 2/3/x? Are they running Gnome 1.x/2.x? Stuff moves around constantly - there's the "normal" Gnome, the "RedHat" Gnome, the "Suse" Gnome, etc.
And there's no promise the user is even using either of the 2 big ones. My example above assumes the user is running IceWM.
Yes, I've swallowed the Blue OSS pill - and I don't regret it. But I defy you to try to support end-user software on the Linux platform.
I used to buy a lot of their stuff. Not any longer. And I'm not alone.
But you're in a very small room. Most people don't know or care about stuff like this. It measures somewhere between a traffic fine for accidentally running a red light and being late for a video rental.
And how many people do you represent? Do you buy for a corporation? Large group? Somehow, I doubt it.
I'm an OSS kinda guy (I write this on my Fedora Core system, using Mozilla) and love it, and have even made sure that our software works on Windows, Mac, and Linux - but none of our customers have *EVER* used our Linux software. A small (but meaningful) percentage of our users are on Macs.
Truth is, much as we who are interested in this stuff might like otherwise, this stuff just doesn't matter to most people - and to those whom it does, Microsoft really is cheaper.
Ever try to support desktop software? Yes, it's getting worse on Windows, but it's still not too bad, compared to supporting some XYZ linux flavor:
Q. What Operating System are you using? A. Linux Q. Ok, what UI are you using? A. What? Q. I mean, what Window Manager? A. What's that? Q. When you click on the start button, what do you see? A. There is no "Start" button... Q. Is there a button where you click on to run a program? A. Yeah. Q. When you click on it, what does it say? A. Enter Command Q. That's it, "Enter Command"? A. Yes. Q. So how do you do stuff? A. What kind of stuff? Q. You know, look at a website. A. Oh, a website! I use Firefox! Q. Good, how do you find FireFox? A. It's on my desktop! Q. So are you using Gnome or KDE? A. I don't know what you're talking about. Q. (deep sigh) A. So, you're looking at a screen, right? Q. Yes. A. And there's a task bar on it, right? Q. No.......
See where this is going? Linux is not for end users. It probably could be - but it just isn't there now. Ubuntu just might be getting there. Macintosh OSX is there. But for end users, only through some very controlled interface, and in some limited capacity.
Now, I was talking with my father-in-law the other day, and he indicated that he would *never* use Linux. I laughed, and told me that he did, every day. And not only that, but he raved to me about it!
With a look of surprise, he asked me how/where - and I pointed to his Dish DVR. (which is Linux-based, all the way down to an ext2/3 filesystem)
See, an airplane is NOT A SOLID OBJECT. It is designed to be very lightweight - more like a kite than a car.
If you were to punch your average plane pretty much anywhere, you'd leave a good-sized dent. There are only a few parts of a plane really suitable to stand on.
On a Cessna 172, you can get inside the plane, step on the foot-steps that are part of the tire, and there's a special place to step up on the fuselage in order to do a visual fuel check.
But if you were to stand on just about anything else, you'd probably cause expensive structural damage. So, if such a lightweight craft were to hit a carbon nanotube line, the plane would most likely be sliced right in half, like so much cheese or warm butter.
The whole point of carbon nanotubes is that they are many times stronger than a steel cable!
The barebones machine in this article is worthless, overhyped garbage. However, I think there is a real market for a silent-as-a-good-DVD-player HTPC that "just works" out of the box and which is unencumbered by DRM. I'd like to see someone make such a system, using the same hardware and disk image on each system...since there's no way I'm going to go through the headache of trying to get a Myth box to work.
It's a free market - why don't YOU do that? If you are right, and there's really a need for this, it'll sell well. Really, it just starts with BUILDING the darn thing - get the right parts together, test them out, make sure you have a source, and then begin promoting it. Start by selling them on Ebay. Get a decent website. Update your/. sig. Etc.
If, as you say, there is a "real market" for these things, you'll become rich indeed.
Isn't having too much swap potentially harmful, though?
Sure - in much the same way that having a red Pacer is harmful.
Yes, something could go wrong. Your dog could go rabid, bite you, and you could die. That's probably many times more likely than the scenario you paint. In short, if you like red Pacers, go for it - it sure won't hurt you any more than any other Pacer.
The only people who should care about how much VM to allocate already know how to allocate it. This is something that should be automated by your distro installer, and should be something that you aren't informed of anymore than the value of SHMMAX.
Yes, there are times where SHMMAX matters. And if you are asking about it, you should already know how to get the right answer, otherwise you are just seriously out of your league.
My strategy generally is to use a file for swap rather than a partition, even in linux.
What I find curious is that you have a strategy. On what relevant experience do you base this strategy? 1 GB of disk space costs less than $0.50. Set up 3 GB of VM if it makes you feel good. The latte you drink while you set it up costs more than the extra disk space!
So go for it!!! Who cares what you do? Heck, give yourself 10x the RAM and see if it actually makes any difference!!! (it won't)
This is sort of like asking: "Which goes faster: the yellow Pacer or the red Pacer?"!
Re:This is what I want in a future OS
on
A New Kind of OS
·
· Score: 1
I think you misunderstand what the future most likely holds.
See, you think that you'll be "held hostage" by some outside company that holds you at ransom for your data, and if you don't pay up, they string you up and cut you off.
I view the future a bit differently: more like a VM.
Imagine a future where you could (securely) get to "your" computer, from any location on the whole planet.
Imagine a future where you never have to worry about hardware failures, or backups.
Imagine a future where performance is automatic - if your webserver gets slashdotted, your VM merely takes more time on a hosted cluster and you pay usage. It's "your" VM, so if you don't like your hosting provider, you switch to another at any time.
Why is this unreasonable? It already is in many contexts...
I've solved this in the past by making sure that their ownership of my ideas is restricted to items produced in the course of (as opposed to during the term of) working for them.
Wow. That sucks!
I've solved this in the past by making sure that none of my ideas are ever owned by anybody but me. I produce software and routinely borrow code and ideas from past projects. It's called "experience" and it's a big part of what I bring to the table. In exchange, I don't do work unless I get ownership of the resulting code. Then, I grant whatever license is appropriate for the use of the resulting code.
This has worked out amazingly well for me. When positioned as a mutual "sharing of ideas", I've had little/no trouble getting people that I work with to agree to these terms, as either a company owner / consultant or an indie consultant.
But then, I'm not going for a job, never have, and I don't work with big companies, though my software company looks to be set to break $1 million in annual sales in the next year or so, with me as CTO.
I personally would not hire an employee on these terms, so YMMV. Decide what you want and go for it. If you want to be a consultant, make sure that you have both good technical and salesmanship skills.
Maybe it's best because it doesn't force you to run super vulnerable buggy OS, or maybe it's best because it doesn't cost a arm and a leg and you have to sell your soul to MS, that is if your key actually works.
None of these reasons imply he actually DOES ANYTHING with the software. A $150,000 truck means nothing to somebody who makes his living hauling goods. The more expensive truck gets the job done, and provides a reliable income.
But the diletante is content with the pickup. Yeah, it hauls stuff, like a couch from the LA-Z boy dealership, but it doesn't compare with the moving van used by professionals.
Gnumeric is a pickup truck. Worse, a mini-pickup. Excel/OO-Calc is more like the moving van.
If what he wants is a spreadsheet, Gnumeric [gnome.org] is the best spreadsheet application I've seen anywhere, and that's including Excel.
Oh, come on, for god's sakes. This is just absurd. What, have you *NEVER* done any decent amount of spreadsheeting?
Why am I asking - OF COURSE YOU HAVEN'T. Gnumeric is a decent spreadsheet, I suppose - when compared to spreadsheets circa 1991. Rows, colums, cells, formulas, all that jazz. Nifty!
Sorry, but suggesting that Gnumeric is *better* that Excel just shows an astonishing tendency to pop off at the mouth whilst having the barest hint of a clue of what the words coming out actually mean.
Sorry, but this is just ridiculous. Had you said OO Calc, I might have bought it. But Gnumeric?!?!?!
It's like saying the wordpad is SOOOO much better than MSWord and OO Writer together! It's like saying that AbiWord is more feature-packed than OO Writer.
Other than freezing an old hard drive to free up bearings to get data off before you throw it out, it's not worth the aggravation.
Wow. When I had to deal with stiction, I did something very different!
1) Unmount the drive. 2) Plug in the drive with power/signal so it's plugged in but not mounted anymore. 3) Turn on the power to the computer. 4) Immediately after power on, twist the drive in a planar spin (like it was a pizza crust you were spinning) so that it frees up and starts spinning. 5) You can tell when the drive is spinning up, because the gyroscopic effect starts to kick in.
Slashdot Mirror
AOL has spent its karma. They're on a strong, downward spiral. This in spite of the fact that one of the largest, coolest things to come out was largely funded by them. (Mozilla) Here's what they need to do to get back on track:
1) Fire everybody in marketing. Re-hire a whole new marketing arm from the likes of Earthlink.
2) Retrain their call support centers. Make it possible to quit AOL.
3) Make a new corporate motto "don't be evil". Follow it. (Google seems to be weakening its resolve)
4) Hire a bunch of highly qualified engineers, let them play, and let them decide what to sell.
If they do the above, they have a ghost of a chance. If they don't, they're fodder for a buy out within another year or two. They might do OK if they merge with the likes of Earthlink - but not as equals. Only where Earthlink takes control and they do it the right way.
Your Post:
Why is it that we keep hearing about this kind of advancement "to be available in five to ten years", and yet the storage capacity of batteries has been stagnated for at least that long?
Your Sig:
Wake up - the future is arriving faster than you think.
Isn't that a bit, eh, contradictory?
BTW: batteries are DEFINITELY improving. Remember rechargables a la 1980, these dumb nicad thingies that never really worked all that well? Compare that to today's NiMH batteries that outperform many alkalines, and are also rechargeable.
I'd call that a heck of an improvement...
How does home-schooling one's children equate to hating them?
I'm pretty sure the idea is that passive cooling is preferable to active cooling - less noise, less power consumption and no moving part to fail.
And, in my experience, any video card that uses active cooling becomes a "passive cooling" system in a year or two of use, anyway. Fan bearings go bad, the fan quits, and I don't even notice until I'm doing something unrelated. To keep things cooler, I usually remove the fan and use the card with just the heatsink.
It's happened a number of times over the years, including my Voodoo II, my Riva TNT2, my ATI Radeon 8500, and my ATI Radeon 97xx. Yes, the fan is DEAD, with no apparent loss of stability or performance.
So why is that fan there, again? Somehow, they think it's important, and even when caked with 1.5" of dust, with a dead fan, the card just keeps working.
I would hate to watch the show with someone who couldn't stand up to Jamie, Adam does this well and thats why the show works so well. If Jamie was allowed complete control everytime, it would be boring.
I agree. Jamie is a great engineer with an intelligent and accepting, if dour personna, while Adam makes the great engineering more tasteful by interjecting lots of humor.
But I love the show! As a homeschool father, I heavily restrict TV and video games for the fall and winter seasons, but for a few shows. Mythbusters is among my favorite - such a spirit of experimentation and discovery!
It's reality TV that doesn't suck.
I seem to recall that 200 MHz and an 8 Gig hard drive was top of the line, some time during the 90's. Such computers seemed to handle word processing, web browsing, email, etc. just fine. What would those components cost now? Not very much because of miniaturization. A $125 computer should be no big deal right?
Don't forget the concept of the minimum cost of production.
It doesn't actually cost (much) more to produce a 300 GB HDD than an 8 GB HDD. You have the same basic amount of aluminum, wire, circuit boards, chips, solder joints, IDE connectors,etc in either case. It occupies the same amount of space, and still needs to be assembled, checked, packed, shipped, and accounted for.
All these things do not change regardless of the HDD capacity. At a certain point, these costs overtake the profit involved in manufacturing, so that it no longer even makes sense to manufacture the drive any longer. That today is somewheres around $75-100 per drive. (retail) This cost is dropping, and has continued to drop, but it's still there.
This is why a $300 computer today blows the doors off a computer built 5 years ago for $1000, but the 5 year old computer cannot be produced today for its relative value. (say, $125)
Say goodbye to karma. Oh well.
What really friggin annoyed me is that I have a shiny, new cell phone, with a universal plug on the bottom. The charger plugs into that plug, and so does the USB cable. So I can't use the USB interface at the same time as I charge the phone with the AC adapter.
BUT IT WON'T CHARGE THE PHONE WITH A USB CABLE!
What kind of retarded corporate brain damage is that? It would be SO NICE if I could plug my phone into my laptop and charge the phone while I'm plugged in at a starbucks.
Friggen Motorola... Otherwise a decent, tri-mode phone The Motorola V325
What then would be the next big thing to top that?
Good question. Who forsaw the Internet?
By definition, the "next big thing" can't be predicted. It's that new thing, that revolutionary way of doing things that could not be predicted but changed the world anyway. How many hunter/gatherers predicted farming? Who could have predicted that the breakup of Bell Telephones in the 1970s would revolutionize interpersonal communications with the Internet?
The next big thing is coming. It probably is already in its infancy. And it's headed to a store near you.
What's that you say?
You can't regulate the airwaves?
Except that you can - sure people can build pirate radio equipment, but they can also rob houses. (just as illegal) People choose not to for various reasons - legallity being one of them.
When will people learn you can't control, regulate or do much of anything with the internet?
They never will, because it's not true.
What's that you say?
At its heart, the Internet is simply a form of communication. All other forms of communication are regulated, why wouldn't the Internet? The fact that it's new doesn't mean that it's un-regulatable so much as the powers that be haven't regulated it... yet.
Give it time. And then the "next big thing" will come along, and the Internet will be no more interesting than a ham radio today.
Your stupid straw man that you can't tell if it's Gnome or KDE is just that, a stupid straw man.
Ok.
Write a program for END USERS - not tech weenies administering a server. I'm talking about accountants, teachers, lawyers, doctors, and the like. Sell it. Write an installer for it, that works with RedHat, Novell, SUSE, Ubuntu, and Debian - recent versions sold in the past 5 years. And since Linux is what we're talking about, it can't be web-based.
And support it.
Look, they could issue (for $100? or how much it costs...
Ok. 300 Million people in the USA. Times $100. That's $30 BILLION dollars. So much for cheap.
to people devices which are able to sign with a private key a short string of digits (16? 20?) that they dictate to you over the phone. You dictate back the 20 digits of the signature.
Ever enter a WEP key? It's 26 letters long. I have to retype one at LEAST 2 or 3 times TWICE in order to get it to work, when I have the key printed right in front of me. Do you REALLY think that's going to work reliably over the phone?
No complication, no computer needed.
Eh, let's see. We're going to relay a 20-character random text key twice over the phone, in and out of a $100 computing device. How is this either one of "No complication" or "no computer needed" !?!?!? What is that $100 thingy if not a limited-function computer?
What happens if you lose your $100 thingy?
Ultimately secure? Not.
Meaning, it isn't even a particularly good assurance of what you're after.
But anything like this would be VASTLY better than relying on the same 9-digit fixed number (the SSN) that appears in cleartext on every kind of document, and of which there are hundreds of copies lying around in offices around the country, from banks to insurance companies to medical offices to schools to universities to... you get the idea.
The problem is that you are trying to solve a social problem with a technical solution. You can't do that. No amount of technology usage would eliminate crime. Your solutions is simply too complicated and expensive to work well. Furthermore, it doesn't fail gracefully. Somebody gets your $100 thingie, and they suddenly can do whatever they want with your bank accounts and whatnot.
I STRONGLY recommend that you read some of Bruce Schnier's work. He started out like you - thoroughly convinced that the proper use of encryption could solve all of society's security ills, through his best-selling book "Applied Cryptography".
But then, the real world showed him how he was simply wrong. He was smart enough to swallow his pride and learn his lessons, and he's subsequently become one of the worlds leading experts on system security. Some of his best works include "Secrets and Lies", and his most recent: "Beyond Fear".
Give it a chance. You could make a 6-figure career by applying his principles!
If I am supporting application software on a GNU/Linux system, presumably because I am under contract to do so, I will specify the supported system(s). "We support RedHat AS 3.0 and above" for example.
When's the last time Roxio was "under a contract" to support their application software on ANY system for their Easy CD-Creator product? Never? I didn't think so.
Also, I don't believe users install an operating system without knowing which one it is. If they do, all bets are off.
You may do well to remember this little tidbit:
"Tis far better to keep your mouth shut and appear that you may not know what you are talking about, than to open it and remove all doubt".
The point is just that. All bets are off. Supporting desktop software on Linux is next to impossible.
And if I am supporting application software and the user can't tell me what the operating system is, I am well within my rights to say "I can't help you unless you tell me exactly what operating system you're running".
I'll just be glad that you aren't running my company! If you said this to your users, you'd go out of business rather quickly!
If you assume that the people who use your product are the most incompetent, utterly, incoherently, mind-numbingly ignorant beings to ever grace a keyboard, and you realize that the users of your software are actually quite a bit dumber than that, you might begin to get an approximation of just how bad reality is.
They'll ask for a manual until they get one. But they won't read it. They'll ask for help files, and you'll provide them. And they won't look at them. They'll ask you to warn them before they do something stupid. And they'll click OK despite the siren, flashing icons, and red background on the warning message, without even reading it. They'll even deny that the message was there in the first place.
Worse still, you'll have people who refuse to read these messages, behaving as though they were truly illiterate, even though they are teachers teaching a college English class!
You just don't have a clue. That doesn't make you bad, just clueless.
Linux is a great server operating system. It does amazing in the embedded space, as well. It's rock solid, quite secure, and is a dream come true if you are a programmer. But it is NOT a consumer operating system.
Unix begat OSX, so the potential exists. But it ain't there yet!
And all the while, better tools for identifications are widely available. I could identify myself to my bank simply by sending them a PGP-signed email: all that this requires of me is to click on the "sign it" button in Thunderbird - and I get incredibly better security than monkeying around with SSNs.
Yes, and no. You get better security, as long as your system isn't trojaned, wormed, or compromised. (And no, running Linux or OSX doesn't make you immune to these problems, though it helps) And so long as a multitude of other factors are considered. Such as:
1) Does your private key reflect sufficient randomness?
2) Does the 1-way function used to generate your private key have a "back door" making for trivial penetration?
3) Is your private key sufficiently private?
4) Is your bank USING PGP to authenticate?
5) Is THEIR private key really private? (If not, there's room for a man-in-the-middle attack)
But, even if those issues didn't exist, this solution simply doesn't scale well. What about people who don't have computers? What about people who can barely turn them on? What about people who are illiterate? What about people who don't speak english? How do you make sure that this works when the power is out?
And, if you think phishing is a problem now, boy, just wait until word gets around that private keys are such a big deal!
Our legislation, and public services, are late some 20 years regarding identity management. The scandal is that they are not brought up to date faster, not that some people are selling email footers that we send around for free.
A great sound bite. Unfortunately, it's just not true. You haven't presented a solution that works well, is cheap, widely understandable, fails gracefully, and is in the reach of the average (non-techie) Joe.
What solution presents all of these?
Certainly not your PGP "Web of Trust".
I'm a techie-type, who tends towards paranoia in security, and I've never set it up. It simply offers no real value. Hardly anybody else uses it, and if they did, they wouldn't care about the signed email. Realistically, nobody's going to say "Yes, I knowed it was you, because the Email was SIGNED!!!".
If somebody spoofs an email, it's pretty easy to look in the headers to identify that it wasn't me, and nobody has ever spoofed me to my detriment. Nor do I know anybody who's been so spoofed to their detriment, either. I've seen SPAM go out with forged from: addresses, but nobody believes that the penis pillz offer actually came from that person. Additionally, even if you encrypt an email so that only the recipient can see it, that recipient is then free to forward your message (without encryption) to whomever they like. So, your email is still a matter of public record. The rules of the game are simple: don't send an email that would be a problem if forwarded.
So what was this PGP thingy supposed to do for me, again?
If a user doesn't know what distro they're using, then they wouldn't know what version of Windows they were using either.
But, on either Windows or Macintosh, it's EASY to tell them how to find out what flavor they're running. That is not true for *nix. And so long as this is the case, the "Year of the Linux Desktop" will never happen.
There's no such consistency when using Linux. Are they running KDE 2/3/x? Are they running Gnome 1.x/2.x? Stuff moves around constantly - there's the "normal" Gnome, the "RedHat" Gnome, the "Suse" Gnome, etc.
And there's no promise the user is even using either of the 2 big ones. My example above assumes the user is running IceWM.
Yes, I've swallowed the Blue OSS pill - and I don't regret it. But I defy you to try to support end-user software on the Linux platform.
I used to buy a lot of their stuff. Not any longer. And I'm not alone.
......
But you're in a very small room. Most people don't know or care about stuff like this. It measures somewhere between a traffic fine for accidentally running a red light and being late for a video rental.
And how many people do you represent? Do you buy for a corporation? Large group? Somehow, I doubt it.
I'm an OSS kinda guy (I write this on my Fedora Core system, using Mozilla) and love it, and have even made sure that our software works on Windows, Mac, and Linux - but none of our customers have *EVER* used our Linux software. A small (but meaningful) percentage of our users are on Macs.
Truth is, much as we who are interested in this stuff might like otherwise, this stuff just doesn't matter to most people - and to those whom it does, Microsoft really is cheaper.
Ever try to support desktop software? Yes, it's getting worse on Windows, but it's still not too bad, compared to supporting some XYZ linux flavor:
Q. What Operating System are you using?
A. Linux
Q. Ok, what UI are you using?
A. What?
Q. I mean, what Window Manager?
A. What's that?
Q. When you click on the start button, what do you see?
A. There is no "Start" button...
Q. Is there a button where you click on to run a program?
A. Yeah.
Q. When you click on it, what does it say?
A. Enter Command
Q. That's it, "Enter Command"?
A. Yes.
Q. So how do you do stuff?
A. What kind of stuff?
Q. You know, look at a website.
A. Oh, a website! I use Firefox!
Q. Good, how do you find FireFox?
A. It's on my desktop!
Q. So are you using Gnome or KDE?
A. I don't know what you're talking about.
Q. (deep sigh)
A. So, you're looking at a screen, right?
Q. Yes.
A. And there's a task bar on it, right?
Q. No.
See where this is going? Linux is not for end users. It probably could be - but it just isn't there now. Ubuntu just might be getting there. Macintosh OSX is there. But for end users, only through some very controlled interface, and in some limited capacity.
Now, I was talking with my father-in-law the other day, and he indicated that he would *never* use Linux. I laughed, and told me that he did, every day. And not only that, but he raved to me about it!
With a look of surprise, he asked me how/where - and I pointed to his Dish DVR. (which is Linux-based, all the way down to an ext2/3 filesystem)
What happens?
nothing.
See, an airplane is NOT A SOLID OBJECT. It is designed to be very lightweight - more like a kite than a car.
If you were to punch your average plane pretty much anywhere, you'd leave a good-sized dent. There are only a few parts of a plane really suitable to stand on.
On a Cessna 172, you can get inside the plane, step on the foot-steps that are part of the tire, and there's a special place to step up on the fuselage in order to do a visual fuel check.
But if you were to stand on just about anything else, you'd probably cause expensive structural damage. So, if such a lightweight craft were to hit a carbon nanotube line, the plane would most likely be sliced right in half, like so much cheese or warm butter.
The whole point of carbon nanotubes is that they are many times stronger than a steel cable!
The barebones machine in this article is worthless, overhyped garbage. However, I think there is a real market for a silent-as-a-good-DVD-player HTPC that "just works" out of the box and which is unencumbered by DRM. I'd like to see someone make such a system, using the same hardware and disk image on each system...since there's no way I'm going to go through the headache of trying to get a Myth box to work.
/. sig. Etc.
It's a free market - why don't YOU do that? If you are right, and there's really a need for this, it'll sell well. Really, it just starts with BUILDING the darn thing - get the right parts together, test them out, make sure you have a source, and then begin promoting it. Start by selling them on Ebay. Get a decent website. Update your
If, as you say, there is a "real market" for these things, you'll become rich indeed.
Isn't having too much swap potentially harmful, though?
Sure - in much the same way that having a red Pacer is harmful.
Yes, something could go wrong. Your dog could go rabid, bite you, and you could die. That's probably many times more likely than the scenario you paint. In short, if you like red Pacers, go for it - it sure won't hurt you any more than any other Pacer.
The only people who should care about how much VM to allocate already know how to allocate it. This is something that should be automated by your distro installer, and should be something that you aren't informed of anymore than the value of SHMMAX.
Yes, there are times where SHMMAX matters. And if you are asking about it, you should already know how to get the right answer, otherwise you are just seriously out of your league.
My strategy generally is to use a file for swap rather than a partition, even in linux.
What I find curious is that you have a strategy. On what relevant experience do you base this strategy? 1 GB of disk space costs less than $0.50. Set up 3 GB of VM if it makes you feel good. The latte you drink while you set it up costs more than the extra disk space!
So go for it!!! Who cares what you do? Heck, give yourself 10x the RAM and see if it actually makes any difference!!! (it won't)
This is sort of like asking: "Which goes faster: the yellow Pacer or the red Pacer?"!
I think you misunderstand what the future most likely holds.
See, you think that you'll be "held hostage" by some outside company that holds you at ransom for your data, and if you don't pay up, they string you up and cut you off.
I view the future a bit differently: more like a VM.
Imagine a future where you could (securely) get to "your" computer, from any location on the whole planet.
Imagine a future where you never have to worry about hardware failures, or backups.
Imagine a future where performance is automatic - if your webserver gets slashdotted, your VM merely takes more time on a hosted cluster and you pay usage. It's "your" VM, so if you don't like your hosting provider, you switch to another at any time.
Why is this unreasonable? It already is in many contexts...
I've solved this in the past by making sure that their ownership of my ideas is restricted to items produced in the course of (as opposed to during the term of) working for them.
Wow. That sucks!
I've solved this in the past by making sure that none of my ideas are ever owned by anybody but me. I produce software and routinely borrow code and ideas from past projects. It's called "experience" and it's a big part of what I bring to the table. In exchange, I don't do work unless I get ownership of the resulting code. Then, I grant whatever license is appropriate for the use of the resulting code.
This has worked out amazingly well for me. When positioned as a mutual "sharing of ideas", I've had little/no trouble getting people that I work with to agree to these terms, as either a company owner / consultant or an indie consultant.
But then, I'm not going for a job, never have, and I don't work with big companies, though my software company looks to be set to break $1 million in annual sales in the next year or so, with me as CTO.
I personally would not hire an employee on these terms, so YMMV. Decide what you want and go for it. If you want to be a consultant, make sure that you have both good technical and salesmanship skills.
Maybe it's best because it doesn't force you to run super vulnerable buggy OS, or maybe it's best because it doesn't cost a arm and a leg and you have to sell your soul to MS, that is if your key actually works.
None of these reasons imply he actually DOES ANYTHING with the software. A $150,000 truck means nothing to somebody who makes his living hauling goods. The more expensive truck gets the job done, and provides a reliable income.
But the diletante is content with the pickup. Yeah, it hauls stuff, like a couch from the LA-Z boy dealership, but it doesn't compare with the moving van used by professionals.
Gnumeric is a pickup truck. Worse, a mini-pickup. Excel/OO-Calc is more like the moving van.
If what he wants is a spreadsheet, Gnumeric [gnome.org] is the best spreadsheet application I've seen anywhere, and that's including Excel.
Oh, come on, for god's sakes. This is just absurd. What, have you *NEVER* done any decent amount of spreadsheeting?
Why am I asking - OF COURSE YOU HAVEN'T. Gnumeric is a decent spreadsheet, I suppose - when compared to spreadsheets circa 1991. Rows, colums, cells, formulas, all that jazz. Nifty!
Sorry, but suggesting that Gnumeric is *better* that Excel just shows an astonishing tendency to pop off at the mouth whilst having the barest hint of a clue of what the words coming out actually mean.
Sorry, but this is just ridiculous. Had you said OO Calc, I might have bought it. But Gnumeric?!?!?!
It's like saying the wordpad is SOOOO much better than MSWord and OO Writer together! It's like saying that AbiWord is more feature-packed than OO Writer.
Absurdity!
Other than freezing an old hard drive to free up bearings to get data off before you throw it out, it's not worth the aggravation.
Wow. When I had to deal with stiction, I did something very different!
1) Unmount the drive.
2) Plug in the drive with power/signal so it's plugged in but not mounted anymore.
3) Turn on the power to the computer.
4) Immediately after power on, twist the drive in a planar spin (like it was a pizza crust you were spinning) so that it frees up and starts spinning.
5) You can tell when the drive is spinning up, because the gyroscopic effect starts to kick in.
So much faster, and has worked for me many times.