I probably deserved that flame bait mod:-( I'll probably get another one for this post, but all I'm saying is what I experienced.
The short answer to your post: I need to have another look at Fedora now that version 5 is available. Up through version 4, the RPM situation was terrible. Every time I added additional repositories to up2date, I regretted it. Things would work for a while, and then up2date would just stop working, making it difficult to upgrade the system.
I've used RedHat, and later Fedora since version 6 point something. I often would add repositories to up2date, but my system got trashed so often, and rpms just failed to work so often, I eventually gave up. It was so bad, I switched to a source-only mode. I typically had dozens of source packages I compiled and installed. Source worked great, but what a pain. Even the few binary packages I did use were so hard to use with up2date, that I switched over to yum.
Fedora 1.0 was unstable, but I waited it out. 2.0 was good, and by 4.0, I liked Fedora better than RedHat Enterprise. Even by 4.0, the package situation was still bad, and I installed generally by source. Common packages such as wine just didn't work right otherwise. Complex packages such as Nvu and Open Wengo I never could get working. Having to install prerequisite packages by reading the README, downloading, then compiling, took a lot of my time. Also, I had such a hard time getting Fedora working on new lap-tops that I almost begged my company not to try it on any Compaq (we're basically a Dell house, though that is changing lately). Just over a year ago, my company needed a laptop in short order, and had to buy one from Best Buy. I think we put over a man-month into it, and I think RedHat still runs funny on it.
I never complained, but that was because I'd never experienced anything better, though I had tried other systems such as Suse. I lost another hard-disk a few months ago, and decided to give Ubuntu a go. It was amazing. Almost every package I use was included in the 'universe', including wine, Open Wengo, and even Nvu (which rocks, by the way). The package manager worked like Yum, but possibly even better, installing all the needed stuff automagically. Just as important, things basically just worked... no more fighting X.org drivers just to get X working... no more compiling kernel patches to get DVDs to play (though I did try a kernel patch to get suspend-to-disk working). It was like getting the stability of RedHat Enterprise, with the cutting-edge software of Fedora.
It was so much easier to use and so many more binary packages just worked that now I look back at my Fedora days and wonder how I ever got through them. So far, I have exactly 6 packages I've downloaded the old way (not through apt-get). 4 are binary VMWare 1.0 server things that came out last week... very cool! One is the rt2500 wireless driver that messed up my machine so badly I was glad when Ubuntu updated their kernel so I could dump my broken one. The other is a FUSE source package I don't use (I use the standard binary package). I can't remember why I downloaded it. Every other package I use (and I'm pretty much a package-fiend) I installed through the standard package system.
I hear good things about Fedora 5. Yum is finally official... how many years did that take? I might give it a spin under the new VMWare server I'm running on Ubuntu, just to see. I hope RedHat gets their act together... I live less than a 20 minute drive from their headquarters, and we could really use another good high-tech success around here. I think Ubunutu might be just the kind fire under their butts needed to get them moving in the right direction.
Nope. I get GPL... even publish some software under it.
Lawyers aren't for keeping everything above the law. They are for harassing your competitors into oblivion. I would give you very good odds that RedHat lawyers sends nasty letters to the likes of Centos. Check out:
Notice the fuzzy language: "CentOS 2, 3, and 4 are built from publically available open source SRPMS provided by a prominent North American Enterprise Linux vendor." The old web sites use to use say right out that they are derived from RedHat.
In Fedora: - RedHat controls the board that decides what goes in an what stays out. It's kind of like a "No Parking, Violators Will be Towed" thing. - RedHat directly takes over source code maintenance for any package that they decide to include. Original authors are typically out of the loop. - The old fedora.us was a user-created add-on package site for RedHat (which use to be free). Marketing at RedHat merged them, then toss them. - Enhancements to code are made by RedHat are usually only available AFTER the RedHat releases software that uses them. - Derivatives of RedHat software get no support, but I bet they get nice letters from RedHat legal.
In Ubuntu: - You are encouraged to become a "Master of the Universe", and help decide what goes in and to maintain the packages. - Ubuntu only takes control over core packages required for average end-users to have stable environments. - There's little need for a user group to build unofficial add-ons (other than EasyBuntu;-)... No ubuntu.us ever existed, and if it did, it would not get subverted. - Enhancements to open-source are fed back to the authors promptly. Authors are in the loop. - Derivatives in theory will be welcomed.
Let's face it: RedHat is a public company controlled by shareholders. Their goal is therefore to suck more money out of us than ever before, and to do it in the next 12 months, so stockholders can sell their stock at a nice profit, and get out. Ubuntu is controlled by the BDFL (one man, the right man), and has been given over to a foundation for long-term viability. Their goal is to replace Windows on the Desktop, and to worry about how that translates into obscene wealth later.
As for the value of getting real support for a derivative distribution... if I were doing a startup based on Unbuntu code, I'd sure as heck want it!
Pain in the thumb could be the corpal tunnel. That can lead to bad things I don't even want to think about. Whenever a friend of mine experiences such pain, I make an effort to talk them into making a change.
So, in short: do your best to get her to change. Switching to the left hand, use a foot mouse, or an old Alps touch pad in her lap (with tapping, not clicking) instead of a mouse on the desk. In my experience, people don't typically use good judgement here. They wait until the pain forces them to change, at which point there can already be permament damage. My rule: if typing or mousing hurts, don't do it. If you can, tell your boss you need a holiday, and take a few days off.
I was a bit luckier. As a programmer, Emacs key sequences trashed my hands and I stupidly ignored the pain until it was unbearable. I had ulnar nerve entrapments in both elbows, with significant damage. The good news is that ulnar nerve underpayments have more room to swell than carpal tunnel underpayments in the wrists, so it's easier to heal. Also, there are good ways to type that help a lot (I now use a laptop in my lap, pushed pretty far away, and make only small motions with my fingers).
When I could no longer type, my employer kindly provided a quite office for me, and paid for the machines and software I needed, and also let me take a few months to become productive again. I used Dragon Dictate for 3 years to write code for Synplicity. The original HDL Analyst was written by dication, rather than typing.
If your friend types a lot of documents, she should get Naturally Speaking, which is very productive and easy to learn. This can lower hand stress a ton. If she's a programmer, it's much harder. What I did was train about 1,600 voice commands into Dragon Dictate to drive Emacs. Then, I could code and use the bash shell all by voice.
However, this is very hard. You have to learn hundreds of voice commands and have to be fluent with them. If you know 1,600 words of a foreign language, you are close to being fluent, and certainly can carry a conversation. It takes months of hard work, not weeks. Also, learning to program while talking is surprisingly difficult. The words your saying seem to interfere with the same part of the brain that writes code. When I type code, I can hear words in my head as I type. Try coding while talking. It's just hard. I would say that learning to code while talking is harder than learning to type fast.
Hope this helps... If you want more advice from a successful programmer using voice entry, e-mail me at bill@billrocks.org. You'll need to answer a challenge question to get past my challenge/response filter.
I too loved Netscape back in the days of the war with M$. But back then, I loved it not because it was better than IE, but because it was the best out of about 10 browsers that were competing fairly in the market.
Then, Netscape became evil (in the way that Google wants not to). They had earned their lead in the market, but with it they added non-standard features, making simple HTML viewing too hard for any small company. There are many examples of this. Here's the top one I found from Google:
They embedded a custom scripting language (Javascript), and added hooks for every darned thing you can imagine. They even tried to get M$ to work with them to drive out all the little guys. In short, they used monopolistic practices to drive out the competition. In doing so, they stupidly handed over the whole market to M$.
With Firefox, the evil motives are gone, so it's no wonder that it's catching on. I love it. Rock on Firefox!
I love the idea of colonizing Mars or other planets, but let's be real. We evolved here on Earth, and there will never be anyplace else for us. We didn't spend hundreds of millions of years evolving to live anywhere else.
We may be able one day to design some sort of life form suitable for other environments, such as space, or another planet, but it wont be us. We'll die off when our environment no longer can support us.
After a spat of explicit gay insest spam, I switched to a challenge/response e-mail filter... of course, on Linux I was pretty much forced to write my own, since Evolution doesn't have one. I must say, I'm somewhat proud of the result. It's at evochallenge.sourceforge.net.
The sooner we all put them in place, the sooner the spam will go away.
Yes, I agree completely. SSTO is quite ambitious, probably one of the reasons NASA decided to push it. More practical rockets just aren't sexy. I have to say I like the ideas behind OTRAG as well (http://en.wikipedia.org/wiki/OTRAG).
So, we need cheap (and not necessarily reliable) disposable rockets for putting most of the stuff up there, with a focus of getting the cost per pound down. Then, we need something like SpaceShipOne, but grown up a bit to put people up there.
I am yet again allowing myself to be somewhat hopeful in thinking that there is a chance that NASA wont screw this up... but I've been doing that for decades, and I'm usually disappointed. My prediction: the companies that get funded by NASA will bulk up from the money, stop innovating because they'll spend too much time sucking up to NASA and doing the required paper-work, and then disappear once the money dries up. NASA will once again blame the companies themselves for the mess.
Having NASA fund projects designed to compete with NASA's core competency is just dumb. Why politicians continue to screw this up is just beyond me. This program should exist, but it should be funded through a client who just wants satellites, like the Air Force. The win for the Air Force would be more and better satellites for their money. For NASA, saving money means cutting budgets and slimming down... it's a lose.
"Our greatest primary task is to put people to work. This is no unsolvable problem if we face it wisely and courageously."
With the enormous power we have with the Internet, we could help bring together those people looking for work and those who need their services. It seems to me that the sticking point is usually that neither of them has any money. But, if they could trade some sort of on-line IOUs, and try to honor them, perhaps there is some solution.
I'm baffled as to how to go about it, but so many people simply looking for work is wrong. If a man is willing to offer his labor, there are needs out there to be filled. To leave him idle wastes his talents and damages his pride.
In the late 90's Lockheed Martin wanted to build a single-state-to-orbit (SSTO) replacement for the space shuttle. They were so confident in their design, all they asked for to build it was $100M. They would fund the rest themselves, and recoup their expenses selling the ship commercially.
NASA killed it in stages. The first stage was to take over program management of the project, which they did simply by funding it to $500M, rather than the $100M Lockheed asked for. Then, they spread development of various pieces of VentureStar to several companies, some of whom have a proven track record of failure. Finally, as various companies failed to develop their piece, they turned on the project, claiming it could never work and was a bad idea in the first place. The end result was much additional funding from Congress to continue backing NASA's stupid shuttle program.
The legacy of VentureStar was quite interesting, and seems to go back to secret SkunkWorks projects. A previous SkunkWorks director, Ben Rich, who presided over the development of the stealth fighter, wrote a book called SkunkWorks. In it, he denied that the hyper-sonic plane (referred to on the net as Aurora) exists, and further claimed that it could not be built. The skin would get too hot, and the hydrogen/oxygen engines were impractical. Not three years after publishing this book, however, Lockheed was asking for a mere $100M to build VentureStar, using technologies never publicly seen before - linear spiking hydrogen/oxygen engines, and a special metallic skin that could take the heat of reentry. Hmm....
So, Lockheed is still sitting on it's VentureStar plans. Boeing has finally built the linear spiking engines, and just to show how NASA was trying to kill the project, Lockheed's VentureStar crew built a successful fuel tank for free (NASA killed the project after another company failed in this portion of the effort).
Another cool project NASA killed was the DC-X, as well as other related SSTO vertical takeoff and landing craft. The cool thing about this rocket was how cheap it was to fly. They demonstrated on their reduced-scale prototype that they could land on gravel, run out a fuel truck, and launch again. Even though the prototype was clearly successful, NASA killed this project after the prototype fell over due to a simple hydraulic malfunction on one of it's three legs and exploded. One of the reasons stated by NASA for killing the DC-X was to focus funds on the higher performance VentureStar project!
A multi-pronged approach may have been better than NASA's single-minded shuttle focus. A DC-X technology based rocket could cheaply lift satellites and building materials for the ISS to LEO (or even lower). Focusing on low-cost, rather than reliability would greatly reduce the cost per-pound of getting stuff in orbit, but would not be suitable for human flight. Space-tugs, using ion-drive (see http://en.wikipedia.org/wiki/Ion_propulsion) could be used to haul the loads from low orbits to higher orbit, and part of the load would be additional fuel for the ion-drive. It would take weeks to months for such a space-tug trip, but that's not long for space borne projects. A separate project like VentureStar or any of the other advanced next-generation designs could be used for human flight.
Oh, well... NASA has a long history of funding and then killing good space concepts. I think this will be no different. It's probably $500M wasted. In the mean-time, thank God for the Russian rockets!
Boy, I hope you didn'g graduate in 2002 or 2003. Those years, I got hundreds of unsolicited resumes each week from people who were plenty good, but there just weren't any jobs. That's when corporate America started sending the new programming jobs overseas in earnest. Now, it just a fad. Some of it make sense, the rest is just CEOs listening to their boards and investors who don't understand how software development works. They seem to think is just like making any of that stuff WalMart sells.
Are you in Silicon Valley? Most of the people I know there who had trouble in 2002-2003 were able to find work in 2004-2005.
I wont dispute that. I will quote EETimes: From a compensation perspective, that stats looks pretty good. The mean salary for U.S.-based engineers participating in this year's survey is $99,00." Of course, that's only the guys who have jobs in the field, and I have seen many good people leave. For more info, check out:
I know it doesn't feel like it, but trust me... this is a "boom" time, and a "bust" time is around the corner. In my experience, this has always been a tough industry.
It's been 20 years, so I don't remember the exact wording, but yes, it was clearly a group of self-proclaimed communists in California trying to help grow unions and sympathy for their cause. They waited at the stop sign on Semiconductor Drive at the entrance to National Semiconductor, and handed out literature. But hey, California is still California! Plenty of fruits and nuts, including me!
Well, as for whether or not we programmers should have a union... it's not the H1-B's that worry me. It's global competition. We might be able to force our wages even higher (the average programmer in the US now makes almost $100K), but that will only increase the off-shoring rates. I'd rather keep those jobs here, for a lot of reasons, not just the programmers themselves.
I've got no problem with unions. They've helped keep corporations honest and done a lot of good for workers in America. We just don't need a union for programmers (but that's a different discussion).
I wasn't saying that old group back in 1986 were like communists... they actually were! They handed out union literature and communist literature at the same time! I also don't have any problem with communism when practiced voluntarily, rather than forced on people.
In 1986, I got a job after graduating from Berkeley with a BS in EECS for $29,500/year. The last hire I made that I'd consider comparable was an Indian student from a state university with a masters. We hired him for $60K/year. I checked out this site for inflation rates:
In short, a smart engineer with a college degree makes the same today as he did 20 years ago. Even back then, half engineers I graduated with were Indian or Chinese. It's no different today.
Sure, we engineers have to compete globally, which makes us poorer on average than doctors and lawyers. I's still rather be an engineer.
The first month after taking that first job, I was approached by communist picketers outside my workplace. They were pushing for unions, and higher wages. These Programmers Guild people are no different. It was a bad idea then, just as it is now.
I've made a few very long-term predictions that have come true... like the unilateral pullout from Gaza and the security fence/wall.
Here's another: RedHat and Microsoft will both be seriously damaged by Ubuntu.
Reasons why:
- Opensource is the only trend Microsoft can't fight with money. As technology progresses, some applications (such as Netscape, Office, and Windows) become mature, old technologies, with little money left to go after. That's when open-source takes over. I'm a Microsoft fan, but I see the writing on the wall.
- RedHat, who is practically in my back yard, and who powered my machines for the last five years, has really messed up. By splitting into Fedora and Enterprise, and then failing to support Fedora properly (actually sabotaging it), they've PO'ed the open-source community. By trying to control ALL software that their package manager can install, they've bitten off more than they can chew. By forcing their control over the entire distribution, and ingnoring many inovations being incorporated into distributions like Debian, they've lost their lead, and are now a poor overall distribution. RedHat still has a chance, but the long string of very poor decisions from RedHat are a solid indicator of more to come.
I noticed my own work productivity was quite a bit higher when working for companies who offer free coffee. Now, I offer free coffee to my employees, and drink half a pot most days myself. The fingers shake, but I still hit the keys! It's one of the ways we Americans stay ahead:-D
At our little company, we interviewed a potential employee, and somebody had the foresight to google him, and found his web site. On his home page, he was shown dressed as a girl. A prominent link showed an unusual hobby: tracking the various names given to the Devil.
Now, many companies are too stuffy to hire individuals who tend to be a bit creative around the fringes of what is considered acceptable in stuffy company. It's their loss. He was a good employee for the years we had him.
It's not as bad as people think. You need a long cable, with a good tensile strength to weight ratio. A simple steel cable will do. On one end you attach the space station, which could be as small as a single module of the current space station. The other end needs a weight, supplies, another half of the space station, space junk... whatever. Then you spin the thing. No big deal.
To dock, you pull up to the middle and grab old of the rotating cable. You then lower yourself down to the station, and enter through a hatch on 'top'.
I think for long-term living in space, it's a win. For the short term, it adds complexity and cost that nobody wants to pay for.
Sounds to me like the FCC made a resoundingly correct decision. I hope our response here will be positive. We need to encourage smart decisions. This one will put me in a good mood for the rest of the day.
The phrase "minimum cost components" was meant to describe the cheapest way to build a system. If you look inside modern consumer products, mostly you will find 180nm and larger. The cheapest transistors you can buy today are in 180nm on $600 8-inch wafers made in China. If you try to buy wafers at any smaller geometry, 130nm or smaller, you pay more per transistor. I got a quote two years ago for 12-in.13u wafers made in the US: $10K/wafer. The larger wafers in the smaller geometry have 4x the number of total transistors, but cost 16 times as much. The math doesn't work out.
In many applications, we're willing to pay more. For example, a cell phone manufacturer recently told me he shrank to 90nm to save power. Also, space is at a premium in a high-end cell phone. However, these phones are not the cheap ones. The cheapest cell phones still rely on older process technology. I think the way Moore said it, he would have used the cheaper cell phones that still use 180nm technology.
Anyway, if all we care about is the Pentium, yes Moore's Law is still on track and probably will remain so for some time. However, the impact that Moore's Law use to have on the electronics industry as a whole has already faded dramatically.
The likes of Intel will be able to afford 32nm. The rest of us are mostly still using 180nm and larger, where the cheapest consumer transistors are currently fabbed in China. The rest of the electronics industry already not benefit much from 65nm, so the great engine of Moore's has already failed us.
Slashdot Mirror
I probably deserved that flame bait mod :-( I'll probably get another one for this post, but all I'm saying is what I experienced.
The short answer to your post: I need to have another look at Fedora now that version 5 is available. Up through version 4, the RPM situation was terrible. Every time I added additional repositories to up2date, I regretted it. Things would work for a while, and then up2date would just stop working, making it difficult to upgrade the system.
I've used RedHat, and later Fedora since version 6 point something. I often would add repositories to up2date, but my system got trashed so often, and rpms just failed to work so often, I eventually gave up. It was so bad, I switched to a source-only mode. I typically had dozens of source packages I compiled and installed. Source worked great, but what a pain. Even the few binary packages I did use were so hard to use with up2date, that I switched over to yum.
Fedora 1.0 was unstable, but I waited it out. 2.0 was good, and by 4.0, I liked Fedora better than RedHat Enterprise. Even by 4.0, the package situation was still bad, and I installed generally by source. Common packages such as wine just didn't work right otherwise. Complex packages such as Nvu and Open Wengo I never could get working. Having to install prerequisite packages by reading the README, downloading, then compiling, took a lot of my time. Also, I had such a hard time getting Fedora working on new lap-tops that I almost begged my company not to try it on any Compaq (we're basically a Dell house, though that is changing lately). Just over a year ago, my company needed a laptop in short order, and had to buy one from Best Buy. I think we put over a man-month into it, and I think RedHat still runs funny on it.
I never complained, but that was because I'd never experienced anything better, though I had tried other systems such as Suse. I lost another hard-disk a few months ago, and decided to give Ubuntu a go. It was amazing. Almost every package I use was included in the 'universe', including wine, Open Wengo, and even Nvu (which rocks, by the way). The package manager worked like Yum, but possibly even better, installing all the needed stuff automagically. Just as important, things basically just worked... no more fighting X.org drivers just to get X working... no more compiling kernel patches to get DVDs to play (though I did try a kernel patch to get suspend-to-disk working). It was like getting the stability of RedHat Enterprise, with the cutting-edge software of Fedora.
It was so much easier to use and so many more binary packages just worked that now I look back at my Fedora days and wonder how I ever got through them. So far, I have exactly 6 packages I've downloaded the old way (not through apt-get). 4 are binary VMWare 1.0 server things that came out last week... very cool! One is the rt2500 wireless driver that messed up my machine so badly I was glad when Ubuntu updated their kernel so I could dump my broken one. The other is a FUSE source package I don't use (I use the standard binary package). I can't remember why I downloaded it. Every other package I use (and I'm pretty much a package-fiend) I installed through the standard package system.
I hear good things about Fedora 5. Yum is finally official... how many years did that take? I might give it a spin under the new VMWare server I'm running on Ubuntu, just to see. I hope RedHat gets their act together... I live less than a 20 minute drive from their headquarters, and we could really use another good high-tech success around here. I think Ubunutu might be just the kind fire under their butts needed to get them moving in the right direction.
Nope. I get GPL... even publish some software under it.
h p?id=2
Lawyers aren't for keeping everything above the law. They are for harassing your competitors into oblivion. I would give you very good odds that RedHat lawyers sends nasty letters to the likes of Centos. Check out:
http://www.centos.org/modules/tinycontent/index.p
Notice the fuzzy language: "CentOS 2, 3, and 4 are built from publically available open source SRPMS provided by a prominent North American Enterprise Linux vendor." The old web sites use to use say right out that they are derived from RedHat.
God, I hate lawyers.
Just compare.
;-) ... No ubuntu.us ever existed, and if it did, it would not get subverted.
In Fedora:
- RedHat controls the board that decides what goes in an what stays out. It's kind of like a "No Parking, Violators Will be Towed" thing.
- RedHat directly takes over source code maintenance for any package that they decide to include. Original authors are typically out of the loop.
- The old fedora.us was a user-created add-on package site for RedHat (which use to be free). Marketing at RedHat merged them, then toss them.
- Enhancements to code are made by RedHat are usually only available AFTER the RedHat releases software that uses them.
- Derivatives of RedHat software get no support, but I bet they get nice letters from RedHat legal.
In Ubuntu:
- You are encouraged to become a "Master of the Universe", and help decide what goes in and to maintain the packages.
- Ubuntu only takes control over core packages required for average end-users to have stable environments.
- There's little need for a user group to build unofficial add-ons (other than EasyBuntu
- Enhancements to open-source are fed back to the authors promptly. Authors are in the loop.
- Derivatives in theory will be welcomed.
Let's face it: RedHat is a public company controlled by shareholders. Their goal is therefore to suck more money out of us than ever before, and to do it in the next 12 months, so stockholders can sell their stock at a nice profit, and get out. Ubuntu is controlled by the BDFL (one man, the right man), and has been given over to a foundation for long-term viability. Their goal is to replace Windows on the Desktop, and to worry about how that translates into obscene wealth later.
As for the value of getting real support for a derivative distribution... if I were doing a startup based on Unbuntu code, I'd sure as heck want it!
It's all just evil marketting after all... Not stupidity, just good business.
Pain in the thumb could be the corpal tunnel. That can lead to bad things I don't even want to think about. Whenever a friend of mine experiences such pain, I make an effort to talk them into making a change.
So, in short: do your best to get her to change. Switching to the left hand, use a foot mouse, or an old Alps touch pad in her lap (with tapping, not clicking) instead of a mouse on the desk. In my experience, people don't typically use good judgement here. They wait until the pain forces them to change, at which point there can already be permament damage. My rule: if typing or mousing hurts, don't do it. If you can, tell your boss you need a holiday, and take a few days off.
I was a bit luckier. As a programmer, Emacs key sequences trashed my hands and I stupidly ignored the pain until it was unbearable. I had ulnar nerve entrapments in both elbows, with significant damage. The good news is that ulnar nerve underpayments have more room to swell than carpal tunnel underpayments in the wrists, so it's easier to heal. Also, there are good ways to type that help a lot (I now use a laptop in my lap, pushed pretty far away, and make only small motions with my fingers).
When I could no longer type, my employer kindly provided a quite office for me, and paid for the machines and software I needed, and also let me take a few months to become productive again. I used Dragon Dictate for 3 years to write code for Synplicity. The original HDL Analyst was written by dication, rather than typing.
If your friend types a lot of documents, she should get Naturally Speaking, which is very productive and easy to learn. This can lower hand stress a ton. If she's a programmer, it's much harder. What I did was train about 1,600 voice commands into Dragon Dictate to drive Emacs. Then, I could code and use the bash shell all by voice.
However, this is very hard. You have to learn hundreds of voice commands and have to be fluent with them. If you know 1,600 words of a foreign language, you are close to being fluent, and certainly can carry a conversation. It takes months of hard work, not weeks. Also, learning to program while talking is surprisingly difficult. The words your saying seem to interfere with the same part of the brain that writes code. When I type code, I can hear words in my head as I type. Try coding while talking. It's just hard. I would say that learning to code while talking is harder than learning to type fast.
Hope this helps... If you want more advice from a successful programmer using voice entry, e-mail me at bill@billrocks.org. You'll need to answer a challenge question to get past my challenge/response filter.
I too loved Netscape back in the days of the war with M$. But back then, I loved it not because it was better than IE, but because it was the best out of about 10 browsers that were competing fairly in the market.
e tscape.html
Then, Netscape became evil (in the way that Google wants not to). They had earned their lead in the market, but with it they added non-standard features, making simple HTML viewing too hard for any small company. There are many examples of this. Here's the top one I found from Google:
http://www.utoronto.ca/webdocs/HTMLdocs/NewHTML/n
They embedded a custom scripting language (Javascript), and added hooks for every darned thing you can imagine. They even tried to get M$ to work with them to drive out all the little guys. In short, they used monopolistic practices to drive out the competition. In doing so, they stupidly handed over the whole market to M$.
With Firefox, the evil motives are gone, so it's no wonder that it's catching on. I love it. Rock on Firefox!
I love the idea of colonizing Mars or other planets, but let's be real. We evolved here on Earth, and there will never be anyplace else for us. We didn't spend hundreds of millions of years evolving to live anywhere else.
We may be able one day to design some sort of life form suitable for other environments, such as space, or another planet, but it wont be us. We'll die off when our environment no longer can support us.
After a spat of explicit gay insest spam, I switched to a challenge/response e-mail filter... of course, on Linux I was pretty much forced to write my own, since Evolution doesn't have one. I must say, I'm somewhat proud of the result. It's at evochallenge.sourceforge.net.
The sooner we all put them in place, the sooner the spam will go away.
Yes, I agree completely. SSTO is quite ambitious, probably one of the reasons NASA decided to push it. More practical rockets just aren't sexy. I have to say I like the ideas behind OTRAG as well (http://en.wikipedia.org/wiki/OTRAG).
So, we need cheap (and not necessarily reliable) disposable rockets for putting most of the stuff up there, with a focus of getting the cost per pound down. Then, we need something like SpaceShipOne, but grown up a bit to put people up there.
I am yet again allowing myself to be somewhat hopeful in thinking that there is a chance that NASA wont screw this up... but I've been doing that for decades, and I'm usually disappointed. My prediction: the companies that get funded by NASA will bulk up from the money, stop innovating because they'll spend too much time sucking up to NASA and doing the required paper-work, and then disappear once the money dries up. NASA will once again blame the companies themselves for the mess.
Having NASA fund projects designed to compete with NASA's core competency is just dumb. Why politicians continue to screw this up is just beyond me. This program should exist, but it should be funded through a client who just wants satellites, like the Air Force. The win for the Air Force would be more and better satellites for their money. For NASA, saving money means cutting budgets and slimming down... it's a lose.
To quote Franklin D. Roosevelt:
"Our greatest primary task is to put people to work. This is no unsolvable problem if we face it wisely and courageously."
With the enormous power we have with the Internet, we could help bring together those people looking for work and those who need their services. It seems to me that the sticking point is usually that neither of them has any money. But, if they could trade some sort of on-line IOUs, and try to honor them, perhaps there is some solution.
I'm baffled as to how to go about it, but so many people simply looking for work is wrong. If a man is willing to offer his labor, there are needs out there to be filled. To leave him idle wastes his talents and damages his pride.
Count on NASA to screw this project up, too.
In the late 90's Lockheed Martin wanted to build a single-state-to-orbit (SSTO) replacement for the space shuttle. They were so confident in their design, all they asked for to build it was $100M. They would fund the rest themselves, and recoup their expenses selling the ship commercially.
NASA killed it in stages. The first stage was to take over program management of the project, which they did simply by funding it to $500M, rather than the $100M Lockheed asked for. Then, they spread development of various pieces of VentureStar to several companies, some of whom have a proven track record of failure. Finally, as various companies failed to develop their piece, they turned on the project, claiming it could never work and was a bad idea in the first place. The end result was much additional funding from Congress to continue backing NASA's stupid shuttle program.
The legacy of VentureStar was quite interesting, and seems to go back to secret SkunkWorks projects. A previous SkunkWorks director, Ben Rich, who presided over the development of the stealth fighter, wrote a book called SkunkWorks. In it, he denied that the hyper-sonic plane (referred to on the net as Aurora) exists, and further claimed that it could not be built. The skin would get too hot, and the hydrogen/oxygen engines were impractical. Not three years after publishing this book, however, Lockheed was asking for a mere $100M to build VentureStar, using technologies never publicly seen before - linear spiking hydrogen/oxygen engines, and a special metallic skin that could take the heat of reentry. Hmm....
So, Lockheed is still sitting on it's VentureStar plans. Boeing has finally built the linear spiking engines, and just to show how NASA was trying to kill the project, Lockheed's VentureStar crew built a successful fuel tank for free (NASA killed the project after another company failed in this portion of the effort).
Another cool project NASA killed was the DC-X, as well as other related SSTO vertical takeoff and landing craft. The cool thing about this rocket was how cheap it was to fly. They demonstrated on their reduced-scale prototype that they could land on gravel, run out a fuel truck, and launch again. Even though the prototype was clearly successful, NASA killed this project after the prototype fell over due to a simple hydraulic malfunction on one of it's three legs and exploded. One of the reasons stated by NASA for killing the DC-X was to focus funds on the higher performance VentureStar project!
A multi-pronged approach may have been better than NASA's single-minded shuttle focus. A DC-X technology based rocket could cheaply lift satellites and building materials for the ISS to LEO (or even lower). Focusing on low-cost, rather than reliability would greatly reduce the cost per-pound of getting stuff in orbit, but would not be suitable for human flight. Space-tugs, using ion-drive (see http://en.wikipedia.org/wiki/Ion_propulsion) could be used to haul the loads from low orbits to higher orbit, and part of the load would be additional fuel for the ion-drive. It would take weeks to months for such a space-tug trip, but that's not long for space borne projects. A separate project like VentureStar or any of the other advanced next-generation designs could be used for human flight.
Oh, well... NASA has a long history of funding and then killing good space concepts. I think this will be no different. It's probably $500M wasted. In the mean-time, thank God for the Russian rockets!
Oops... this one works.
a rticleID=169400754
http://www.eet.com/news/latest/showArticle.jhtml?
Boy, I hope you didn'g graduate in 2002 or 2003. Those years, I got hundreds of unsolicited resumes each week from people who were plenty good, but there just weren't any jobs. That's when corporate America started sending the new programming jobs overseas in earnest. Now, it just a fad. Some of it make sense, the rest is just CEOs listening to their boards and investors who don't understand how software development works. They seem to think is just like making any of that stuff WalMart sells.
Are you in Silicon Valley? Most of the people I know there who had trouble in 2002-2003 were able to find work in 2004-2005.
I wont dispute that. I will quote EETimes: From a compensation perspective, that stats looks pretty good. The mean salary for U.S.-based engineers participating in this year's survey is $99,00." Of course, that's only the guys who have jobs in the field, and I have seen many good people leave. For more info, check out:
http://www.eet.com/news/latest/showArticle.jhtml ;jsessionid=BIVESULJCG5LMQSNDLPSKHSCJUNN2JVN?artic leID=169400754&pgno=2
I know it doesn't feel like it, but trust me... this is a "boom" time, and a "bust" time is around the corner. In my experience, this has always been a tough industry.
It's been 20 years, so I don't remember the exact wording, but yes, it was clearly a group of self-proclaimed communists in California trying to help grow unions and sympathy for their cause. They waited at the stop sign on Semiconductor Drive at the entrance to National Semiconductor, and handed out literature. But hey, California is still California! Plenty of fruits and nuts, including me! Well, as for whether or not we programmers should have a union... it's not the H1-B's that worry me. It's global competition. We might be able to force our wages even higher (the average programmer in the US now makes almost $100K), but that will only increase the off-shoring rates. I'd rather keep those jobs here, for a lot of reasons, not just the programmers themselves.
I've got no problem with unions. They've helped keep corporations honest and done a lot of good for workers in America. We just don't need a union for programmers (but that's a different discussion).
I wasn't saying that old group back in 1986 were like communists... they actually were! They handed out union literature and communist literature at the same time! I also don't have any problem with communism when practiced voluntarily, rather than forced on people.
In 1986, I got a job after graduating from Berkeley with a BS in EECS for $29,500/year. The last hire I made that I'd consider comparable was an Indian student from a state university with a masters. We hired him for $60K/year. I checked out this site for inflation rates:
http://eh.net/hmit/compare/
In short, a smart engineer with a college degree makes the same today as he did 20 years ago. Even back then, half engineers I graduated with were Indian or Chinese. It's no different today.
Sure, we engineers have to compete globally, which makes us poorer on average than doctors and lawyers. I's still rather be an engineer.
The first month after taking that first job, I was approached by communist picketers outside my workplace. They were pushing for unions, and higher wages. These Programmers Guild people are no different. It was a bad idea then, just as it is now.
I've made a few very long-term predictions that have come true... like the unilateral pullout from Gaza and the security fence/wall.
Here's another: RedHat and Microsoft will both be seriously damaged by Ubuntu.
Reasons why:
- Opensource is the only trend Microsoft can't fight with money. As technology progresses, some applications (such as Netscape, Office, and Windows) become mature, old technologies, with little money left to go after. That's when open-source takes over. I'm a Microsoft fan, but I see the writing on the wall.
- RedHat, who is practically in my back yard, and who powered my machines for the last five years, has really messed up. By splitting into Fedora and Enterprise, and then failing to support Fedora properly (actually sabotaging it), they've PO'ed the open-source community. By trying to control ALL software that their package manager can install, they've bitten off more than they can chew. By forcing their control over the entire distribution, and ingnoring many inovations being incorporated into distributions like Debian, they've lost their lead, and are now a poor overall distribution. RedHat still has a chance, but the long string of very poor decisions from RedHat are a solid indicator of more to come.
I noticed my own work productivity was quite a bit higher when working for companies who offer free coffee. Now, I offer free coffee to my employees, and drink half a pot most days myself. The fingers shake, but I still hit the keys! It's one of the ways we Americans stay ahead :-D
Oh, we didn't fire him. We saw the site before we hired him. He left our company a couple years later, looking for better things.
At our little company, we interviewed a potential employee, and somebody had the foresight to google him, and found his web site. On his home page, he was shown dressed as a girl. A prominent link showed an unusual hobby: tracking the various names given to the Devil.
Now, many companies are too stuffy to hire individuals who tend to be a bit creative around the fringes of what is considered acceptable in stuffy company. It's their loss. He was a good employee for the years we had him.
Whoa! I've been running the new release for two days and didn't even know it! I thought I was still running the beta.
It's not as bad as people think. You need a long cable, with a good tensile strength to weight ratio. A simple steel cable will do. On one end you attach the space station, which could be as small as a single module of the current space station. The other end needs a weight, supplies, another half of the space station, space junk... whatever. Then you spin the thing. No big deal.
:-D
To dock, you pull up to the middle and grab old of the rotating cable. You then lower yourself down to the station, and enter through a hatch on 'top'.
I think for long-term living in space, it's a win. For the short term, it adds complexity and cost that nobody wants to pay for.
It's not rocket science
Sounds to me like the FCC made a resoundingly correct decision. I hope our response here will be positive. We need to encourage smart decisions. This one will put me in a good mood for the rest of the day.
The phrase "minimum cost components" was meant to describe the cheapest way to build a system. If you look inside modern consumer products, mostly you will find 180nm and larger. The cheapest transistors you can buy today are in 180nm on $600 8-inch wafers made in China. If you try to buy wafers at any smaller geometry, 130nm or smaller, you pay more per transistor. I got a quote two years ago for 12-in .13u wafers made in the US: $10K/wafer. The larger wafers in the smaller geometry have 4x the number of total transistors, but cost 16 times as much. The math doesn't work out.
In many applications, we're willing to pay more. For example, a cell phone manufacturer recently told me he shrank to 90nm to save power. Also, space is at a premium in a high-end cell phone. However, these phones are not the cheap ones. The cheapest cell phones still rely on older process technology. I think the way Moore said it, he would have used the cheaper cell phones that still use 180nm technology.
Anyway, if all we care about is the Pentium, yes Moore's Law is still on track and probably will remain so for some time. However, the impact that Moore's Law use to have on the electronics industry as a whole has already faded dramatically.
The likes of Intel will be able to afford 32nm. The rest of us are mostly still using 180nm and larger, where the cheapest consumer transistors are currently fabbed in China. The rest of the electronics industry already not benefit much from 65nm, so the great engine of Moore's has already failed us.