One big obstacle is that too many applications I see require administrator privileges not just to install but also to run. Your end users figure that out, set themselves up as administrators, and leave it at that.
At that price, supporting free software is a mixed bargain if I ever heard of one. Note that it supports Linux binaries, but it is not Linux as we know it.
I have been doing independent contracting for a couple of years now, and the biggest expense by far has been health insurance. Bear in mind that my wife and I are in our mid 50s, and rates are higher for us. We were able to continue group coverage through a state administered program, but the price for the two of us is roughly US$975 / month.
Whatever you do, do not just try to "go bare" without that insurance. I have been perfectly healthy, but turned yellow in September, had major cancer surgery in October (Whipple Procedure, for colangeocarcinoma (bile duct tumor)), and am seeing bills totalling about US$75000 for five hours on the operating table and a week of recovery at the UAB Hospital. The insurance is paying most of it (about 90%) and the remainder is still a large price. It is still better than having been uninsured when it happened!
So, this week I am back to knocking on my contacts' doors looking for more contract work, a bit poorer but very glad I paid the insurance premiums.
Because I have a corporate past, some of my works must be published under a pseudonym. The honorable history of the "nom de plume" descends from this and other crazy rulings.
Does the record label own all the works of "Joe Skunk?" Fine, release your nest record as "Joseph Weasel" and they will never know.
Does your employer prohibit your publications without prior review, and rejects everything you say? Fine, publish under another name.
Does anyone remember the Ada language books by "Do While Jones?" They were published under a false name for just this sort of reason. (And, no, I am not Do While Jones.)
Moral? Say what you please, release what you will, but misdirect them as to who was saying it. Sometimes freedom comes with a strange price.
After 25 years working as an electronics engineer, the last company I worked for went into technical bankruptcy, stopped meeting payroll, and I was forced to reconsider whether I wanted to continue in this line of work. Result? I decided to take the savings, 401K, and such and put it into a more sane business.
So my wife and I expanded her business (one of those "horribly overpayed wedding photographers") and now I work full time selling portraits, photographing weddings, doing bookeeping, and such. I couldn't be happier!
The life as an engineer was (excuse me) pathetic. Why should I spend all my life chained to a desk, living in a cube farm, and putting up with the Boss from Hell who figured he owned me as so much chattel property? Life is much better now.
So tell me again why I would even talk any teenager into becoming an engineer? They would be fools to do so.
This really looks to me like tilting at windmills. Red Hat Tried this with their eminently forgettable "Blue Curve" standard look and feel, and the result pleased no one that I have asked. It is possible to skin them to look alike, of course, but below decks there is little enough similarity to make them mix as well as oil and water.
The real question is "Why Bother?" If both libraries are present, apps from both work well enough together to make the whole question moot. This is a marketing driven decision, with no real respect to the technical merits of the question.
This new "feature" causes a dilemma for the professional photographic community. Image if you will the wedding where the bridesmaids' dresses are in a lovely shade of "banknote green" (quite possible given the wild colors we see at weddings) and that the printer decides that it must put banding in the proof prints, because it might be counterfiet money. Now, imagine explaining to the the bride's mother why the stripes in the pictures are there. Ugh. HP broke their printers intentionally, and it will come back and bite them in strange and wonderful ways.
Yes, what they describe may indeed work great for the intended purpose of reducing the accuracy of their printers under certain circumstances, but the fact of reducing their output quality will sometimes cause user problems which are totally unrelated to counterfeiting. Their software simply cannot be smart enough to avoid the false positives which will most certainly occur.
A good friend of mine came by and presented me with a wrapped item, about 1 meter long with bulges on each end. I tried to guess the contents, but to no avail. It turned out to be a beat-up air horn off of a large truck - which he had found in a junk yard. It didn't take long to find the fittings in the junk box to wake up the neighbothood with it.
Loud? Oh my! The 100 PSI shop air will make it sing. Now, where can I install it?
Long ago and far away, while I was in college full time (Cal Poly Pomona) I payed for it by working full time swing shift at the Perkin Elmer plant in Pomona, California. As an environmental test technician, I got to see designed and built the mass spectrometer which was used in the Viking Landers, which successfully landed on Mars, and which worked when they arrived.
The thing which stands out about these old birds (this was the mid-1970s, mind you) is that they were very rugged, and very simple electronically, by our standards. Most of the electronics were analog, and the electronic technologies used were huge, robust, massive pieces of silicon - by today's standards. The components were all tested beyond all reason, the modules were tested just as hard, and the final assemblies were tested more so. It cost a fortune - but it did work when it got there.
Mars is a hard target. We know that now, and it has become apparent that the statistics speak against getting there on the cheap.
We knew that sooner or later this requirement would be removed. Those of us who love CW (Morse Code) still use it, and others will continue to do so, if only because it is simple, it works, and it overcomes real language barriers.
Still, even though we may love it, it is an anachronism, and the requirement will be dropped, like it or not.
From the RMS perspective, this makes perfect sense. One of his charms, if you will, is that he does not deviate from his ideals, even when it offends a large group. Free is free, and anythoing that compromises that is less than perfect.
Like any other outspoken issue-perfectionist, this grates on those who are less tough about that issue. But make no bones about it, he would be less respected in the end if he compromised.
From what I have seen, the person who installs GNU/Linux/BSD/etc on a work desktop is generally in the power-user category in the first place. Your average corporate user is not the "culprit" as she is not sufficiently interested in what is under the hood. Your early adopters (who adopted free software years ago) are more likely to install what they need, rather than what the corporate leviathan requires.
It was interesting to hear that my daughter (Ph.D. Candidate at University of Washington, Seattle) brought her own system into the lab, set it up with Linux, and uses it rather than the department standard systems. The reason? Lyx! UW has the standard dissertation formats all set up for lyx/latex, and so this is the best solution for her. The other reason is that all of the rest of the lab users do not get beyond her login prompt, and so her system does not get used by others and messed up regularly. The IT folks allow her access to the network, so she is fully equipped.
Quite so. As an old guy, I learned Morse long ago, and use it to this day, but only on the radio. It is fun to be able to send and receive it (in my head - I do not bother writing it down) as fast as I can type, or even faster. I can listen in and follow along with the conversation, without having to take my eyes off of my work.
On the other hand, the most efficient communication I have ever been involved in involved using a sound board on my PC, hooking it up to the audio in/out of the radio transceiver, and using the computer to generate PSK31 encoded signals.
Hansi Reiser has written linux software for doing this: http://www.qsl.net/dl9rdz/#psk
For better or worse, this battle is based as much on the court of public opinion as anything else. The repeated accusations, the repeated lies on SCO's part, will do their damage, even though all of it may well prove to be baseless.
As Deep Throat (of Watergate fame) said: "Follow the money."
You know, that is the funny part. This system is a picture editing system which spends most of its time running Adobe Photoshop on Windows, but which I used to test a recent distribution (SuSE 8.2) as a dual-boot. The SuSE installer found _everything_ in the way of hardware and configured it to work fine on the first try. Even the SanDisk flash-card reader works. Life is good!
This system is based on a Gigabyte GA-7DXR+, and yes, the sound is on board - a Creative CT5880 chip. Perfect? Well no, but certainly more than adequate for my actual needs. And yes, it works under Linux.
Those of us in the user/developer community have the source for gnu/linux and all of the parts and pieces, but we do not have access to the SCO source code. So we cannot directly run the analysis to compare what may be similar and why it is that way.
However, as an SCO licensee, would not IBM have the SCO source code and be able to do a full analysis as to the similarities and where exactly they occur? They most certainly have the computing horsepower to check it every way possible, and in a flat hurry. Assuming this to be true, we can safely assume that IBM knows of each and every match and similarity, and has already done considerable analysis on the results. Perhaps this accounts for their apparent unruffled confidence going into this legal battle. Something tells me that they know what SCO sees, and already know better than SCO where it all came from. 4.4BSDLite? FreeBSD? Caldera during the cooperative period? Hardware manufacturer driver code? Sample code from Knuth? The Dragon Book? Some issue of the ACM Proceedings? We may not know, but IBM probably already has the paper trail to prove the source for every similarity that good software tools could find. IBM is nothing if not thorough in legal matters. Thus, their confidence in the outcome. IBM knows, and is keeping that as a closely guarded secret to be released in due time.
This also explains SCO's strained unwillingness to point out in public the specific parts of the code which they accuse of being copied. The distributed institutional memory of the entire UNIX/GNU community will find the true source of any similarities in due course (and rather quickly). If the common source is from some published literature, it will be found. If it is from some third party, that will come out. Someone remembers, and the truth will come out. SCO needs the theater to push stock prices up, and so does not want an early resolution.
(By the way, could this access to the SCO source code apsect be another reason Microsoft recently bought the SCO license? They are most assuredly curious to know the outcome before the trial, and this would give them a leg up. This foreknowledge requires access to the SCO source code, which Microsoft has just purchased.)
Current SCO management does not appear to understand software development particularly well. (The modern business schools seem to teach that a good manager can manage absolutely anything - a naked lie, but a lie which many of us have witnessed firsthand - i.e. the PHB.) There has arisen a generation which knows finance better than technology,. and this crowd is in control. As such, the possibilities of what actually happened are not apparent to them, and they can only see one-way transfers of their precious code to the unworthies, and not that it well could have occurred the other way around. The fact that CVS repositories could actually prove such a thing has escaped their notice. Discovery should be a hard time for this crowd.
All we see now is posturing and bluff. But in this battle I would say that IBM is actually the cat and SCO the mouse. IBM now gets to choose whether to go for the quick kill, or play with their mouse for while first. SCO will go down - and will not need to be bought off to do so. It is merely a matter of what torment they will endure in the process.
The ability to run one or more concurrent instances of Linux (or whatever, quite frankly) internally to one of the Xilinx Virtex II parts is seriously amazing. Ignore the board it comes on for development for now - that is just cruft. The Virtex II is probably the most powerful instantly reconfigurable DSP engine in existence (think audio, video manipulation at real time speeds). They have internal hardware to perform from 16 to 128 simultaneous 16x16 multiply/accumulate operations simultaneously, _in_one_clock_cycle_. And if you don't like what it is doing, you can change it, time and time again, forever. Raw Power. Complete Reconfigurability. Sweet!
Combine this kind of power with multiple PPC processors on the same die, and the possibilities are incredible. The big difficulty is that the operation of the hardware and software can be so tightly tied together that it is difficult to program and debug. Everything is controlled by software (both the software and the VHDL or Verilog based FPGA code) and so the possibilities are limitless.
Kudos to Jim Ready and the folks at Monta Vista for supporting this kind of device with development tools for Linux.
I have to remind my fellow photographers to charge enough for their work. Why? Professional photography is a luxury item, one which no one actually needs, but some people actually want enough to pay for. Think of it: absolutely no one goes hungry (except the professional photographer) if a photograph is not made. It is a luxury.
Think of it: anyone can go to Wal-Mart and buy a rather good single-use camera, and if they are careful, get excellent results. For less than US$20, they can have the camera and double prints, and get the prints back today. Why, then, does anyone go to a professional photographer to have done what they could do themselves? The reason is usually that they realize that they cannot be in the pictures they shoot themselves, and they need someone to do the job for them, preferably someone who has done it enoough before to be sure to get the expected pictures of The Big Event. So we as professional photographers provide a service, and a luxury at that.
Now most of us have changed over from film to digital imaging, and the originals are just jpg files. Why don't we just sell the original files? Well, I have argued with some of the locals that we could do that, but the average person is not prepared to deal with the 500+ multi-megabyte jpg files generated at a simple wedding. The raw files are just that - raw. We typically spend much of a week's time doing the pre-print processing before a customer sees them. There is color correction, cropping, eliminating the glasses glare, swapping heads - all of the usual retouching is done, and from a 500 raw picture wedding we may present 150 images.
Now we gave up on paper proofing some time ago. We show the images by projection (Epson LCD projectors are great for this), and the customer gets to decide what she (usually it is a woman) wants - we provide just about any print format anyone could want, and can frame them, bind them into books, and such. Whatever prints were ordered are delivered at the end of the process.
An average wedding takes two of us probably 30 hours each to do right. The results are gorgeous, by the way. Could we do weddings for less? Sure, but not at this quality level. Could we just shoot weddings and hand the customer a couple of CDs of jpg files and our blessings to use however they wish? Sure, but the results in general would not be as good, as most folks (even uber-geeks) are not all that good with gimp or photoshop to do the post-processing themselves.
In the end, you get what you pay for. Professional photography is a luxury item. Deal with it.
The local phone service I get (through my cable provider) comes with a bill broken out according to every mandatory fee and tax, and the mandatory fees and taxes are larger then the phone service cost itself.
Whatever they may be for, the combination of added fees and taxes on phone service is exhorbitant already. Adding them to other net services is just another revenue stream for someone else.
Phbbbbbt!
IPCop as a quick solution to firewalling
on
IPCop 0.1.1 Review
·
· Score: 5, Informative
We have tried IPCop 0.1.1 at the office, and it has one very big advantage over using a general purpose distribution: it installs and comes up running very quickly. From inserting the CDROM to completion of the install on a typical system (200MHz Pentium with 64MB memory) it took about 14 minutes to having it running.
We use it as a three-way firewall with a DMZ, and that is stone-cold simple to install. Slick, with no problems.
IPCop is excellent for probably 90% of the firewall needs for individuals and small businesses. It is based on linux kernel 2.2.20 and ipchains. It is GPLed, has a quality web interface, and installs fast and easy. Furthermore, the user list is friendly and helpful. I downloaded the iso for it, wrote it to a cd, and then took about 15 minutes start to finish with the initial installation. After that, the fine tuning was handled over a very intuitive web interface. I would rate it a 9.5/10.
Many years ago I worked in a lab where shipboard sonar equipment was being built. Their cooling system for the high power units consisted of aluminum plates with embedded copper tubing carrying filtered sea water right through the rack of electronics. It carried away the heat quite well, but the whole thought of using sea water was antithetical to everything I would normally consider.
On the other hand, large high power vacuum tubes have been water cooled for many decades. The most impressive were the vapor-phase cooling units in which the cooling was done by boiling the water off of the external anode of the tube. The steam was then condensed back into water in an outdoor cooling tower.
The assumption that all development is in the corporate realm is most evident in this sort of article. Companies come and companies go, but the development of free software and hardware on which it can run goes on with or without corporate backing.
We have a project to develop free hardware designs to which free software is being ported, with no corporate backing. For an example of an embedded controller to which Linux is being ported, look here:
Slashdot Mirror
One big obstacle is that too many applications I see require administrator privileges not just to install but also to run. Your end users figure that out, set themselves up as administrators, and leave it at that.
This is nothing new...
At that price, supporting free software is a mixed bargain if I ever heard of one. Note that it supports Linux binaries, but it is not Linux as we know it.
I have been doing independent contracting for a couple of years now, and the biggest expense by far has been health insurance. Bear in mind that my wife and I are in our mid 50s, and rates are higher for us. We were able to continue group coverage through a state administered program, but the price for the two of us is roughly US$975 / month.
Whatever you do, do not just try to "go bare" without that insurance. I have been perfectly healthy, but turned yellow in September, had major cancer surgery in October (Whipple Procedure, for colangeocarcinoma (bile duct tumor)), and am seeing bills totalling about US$75000 for five hours on the operating table and a week of recovery at the UAB Hospital. The insurance is paying most of it (about 90%) and the remainder is still a large price. It is still better than having been uninsured when it happened!
So, this week I am back to knocking on my contacts' doors looking for more contract work, a bit poorer but very glad I paid the insurance premiums.
I would advise you to do the same.
Because I have a corporate past, some of my works must be published under a pseudonym. The honorable history of the "nom de plume" descends from this and other crazy rulings.
Does the record label own all the works of "Joe Skunk?" Fine, release your nest record as "Joseph Weasel" and they will never know.
Does your employer prohibit your publications without prior review, and rejects everything you say? Fine, publish under another name.
Does anyone remember the Ada language books by "Do While Jones?" They were published under a false name for just this sort of reason. (And, no, I am not Do While Jones.)
Moral? Say what you please, release what you will, but misdirect them as to who was saying it. Sometimes freedom comes with a strange price.
After 25 years working as an electronics engineer, the last company I worked for went into technical bankruptcy, stopped meeting payroll, and I was forced to reconsider whether I wanted to continue in this line of work. Result? I decided to take the savings, 401K, and such and put it into a more sane business.
So my wife and I expanded her business (one of those "horribly overpayed wedding photographers") and now I work full time selling portraits, photographing weddings, doing bookeeping, and such. I couldn't be happier!
The life as an engineer was (excuse me) pathetic. Why should I spend all my life chained to a desk, living in a cube farm, and putting up with the Boss from Hell who figured he owned me as so much chattel property? Life is much better now.
So tell me again why I would even talk any teenager into becoming an engineer? They would be fools to do so.
This really looks to me like tilting at windmills. Red Hat Tried this with their eminently forgettable "Blue Curve" standard look and feel, and the result pleased no one that I have asked. It is possible to skin them to look alike, of course, but below decks there is little enough similarity to make them mix as well as oil and water.
The real question is "Why Bother?" If both libraries are present, apps from both work well enough together to make the whole question moot. This is a marketing driven decision, with no real respect to the technical merits of the question.
This new "feature" causes a dilemma for the professional photographic community. Image if you will the wedding where the bridesmaids' dresses are in a lovely shade of "banknote green" (quite possible given the wild colors we see at weddings) and that the printer decides that it must put banding in the proof prints, because it might be counterfiet money. Now, imagine explaining to the the bride's mother why the stripes in the pictures are there. Ugh. HP broke their printers intentionally, and it will come back and bite them in strange and wonderful ways.
Yes, what they describe may indeed work great for the intended purpose of reducing the accuracy of their printers under certain circumstances, but the fact of reducing their output quality will sometimes cause user problems which are totally unrelated to counterfeiting. Their software simply cannot be smart enough to avoid the false positives which will most certainly occur.
A good friend of mine came by and presented me with a wrapped item, about 1 meter long with bulges on each end. I tried to guess the contents, but to no avail. It turned out to be a beat-up air horn off of a large truck - which he had found in a junk yard. It didn't take long to find the fittings in the junk box to wake up the neighbothood with it.
Loud? Oh my! The 100 PSI shop air will make it sing. Now, where can I install it?
Long ago and far away, while I was in college full time (Cal Poly Pomona) I payed for it by working full time swing shift at the Perkin Elmer plant in Pomona, California. As an environmental test technician, I got to see designed and built the mass spectrometer which was used in the Viking Landers, which successfully landed on Mars, and which worked when they arrived.
The thing which stands out about these old birds (this was the mid-1970s, mind you) is that they were very rugged, and very simple electronically, by our standards. Most of the electronics were analog, and the electronic technologies used were huge, robust, massive pieces of silicon - by today's standards. The components were all tested beyond all reason, the modules were tested just as hard, and the final assemblies were tested more so. It cost a fortune - but it did work when it got there.
Mars is a hard target. We know that now, and it has become apparent that the statistics speak against getting there on the cheap.
Faster, better, cheaper - which two did you want?
We knew that sooner or later this requirement would be removed. Those of us who love CW (Morse Code) still use it, and others will continue to do so, if only because it is simple, it works, and it overcomes real language barriers.
Still, even though we may love it, it is an anachronism, and the requirement will be dropped, like it or not.
73
W4TI
From the RMS perspective, this makes perfect sense. One of his charms, if you will, is that he does not deviate from his ideals, even when it offends a large group. Free is free, and anythoing that compromises that is less than perfect.
Like any other outspoken issue-perfectionist, this grates on those who are less tough about that issue. But make no bones about it, he would be less respected in the end if he compromised.
So be it.
From what I have seen, the person who installs GNU/Linux/BSD/etc on a work desktop is generally in the power-user category in the first place. Your average corporate user is not the "culprit" as she is not sufficiently interested in what is under the hood. Your early adopters (who adopted free software years ago) are more likely to install what they need, rather than what the corporate leviathan requires.
It was interesting to hear that my daughter (Ph.D. Candidate at University of Washington, Seattle) brought her own system into the lab, set it up with Linux, and uses it rather than the department standard systems. The reason? Lyx! UW has the standard dissertation formats all set up for lyx/latex, and so this is the best solution for her. The other reason is that all of the rest of the lab users do not get beyond her login prompt, and so her system does not get used by others and messed up regularly. The IT folks allow her access to the network, so she is fully equipped.
Free software? Life is good!
Quite so. As an old guy, I learned Morse long ago, and use it to this day, but only on the radio. It is fun to be able to send and receive it (in my head - I do not bother writing it down) as fast as I can type, or even faster. I can listen in and follow along with the conversation, without having to take my eyes off of my work.
On the other hand, the most efficient communication I have ever been involved in involved using a sound board on my PC, hooking it up to the audio in/out of the radio transceiver, and using the computer to generate PSK31 encoded signals.
Hansi Reiser has written linux software for doing this: http://www.qsl.net/dl9rdz/#psk
73,
W4TI
For better or worse, this battle is based as much on the court of public opinion as anything else. The repeated accusations, the repeated lies on SCO's part, will do their damage, even though all of it may well prove to be baseless.
As Deep Throat (of Watergate fame) said: "Follow the money."
You know, that is the funny part. This system is a picture editing system which spends most of its time running Adobe Photoshop on Windows, but which I used to test a recent distribution (SuSE 8.2) as a dual-boot. The SuSE installer found _everything_ in the way of hardware and configured it to work fine on the first try. Even the SanDisk flash-card reader works. Life is good!
This system is based on a Gigabyte GA-7DXR+, and yes, the sound is on board - a Creative CT5880 chip. Perfect? Well no, but certainly more than adequate for my actual needs. And yes, it works under Linux.
I spent so many years with no audio on any system, that the first hardware that had it was a shock: KDE starts with a bongo riff?!
All those years I thought those gears made a different sound.
Those of us in the user/developer community have the source for gnu/linux and all of the parts and pieces, but we do not have access to the SCO source code. So we cannot directly run the analysis to compare what may be similar and why it is that way.
However, as an SCO licensee, would not IBM have the SCO source code and be able to do a full analysis as to the similarities and where exactly they occur? They most certainly have the computing horsepower to check it every way possible, and in a flat hurry. Assuming this to be true, we can safely assume that IBM knows of each and every match and similarity, and has already done considerable analysis on the results. Perhaps this accounts for their apparent unruffled confidence going into this legal battle. Something tells me that they know what SCO sees, and already know better than SCO where it all came from. 4.4BSDLite? FreeBSD? Caldera during the cooperative period? Hardware manufacturer driver code? Sample code from Knuth? The Dragon Book? Some issue of the ACM Proceedings? We may not know, but IBM probably already has the paper trail to prove the source for every similarity that good software tools could find. IBM is nothing if not thorough in legal matters. Thus, their confidence in the outcome. IBM knows, and is keeping that as a closely guarded secret to be released in due time.
This also explains SCO's strained unwillingness to point out in public the specific parts of the code which they accuse of being copied. The distributed institutional memory of the entire UNIX/GNU community will find the true source of any similarities in due course (and rather quickly). If the common source is from some published literature, it will be found. If it is from some third party, that will come out. Someone remembers, and the truth will come out. SCO needs the theater to push stock prices up, and so does not want an early resolution.
(By the way, could this access to the SCO source code apsect be another reason Microsoft recently bought the SCO license? They are most assuredly curious to know the outcome before the trial, and this would give them a leg up. This foreknowledge requires access to the SCO source code, which Microsoft has just purchased.)
Current SCO management does not appear to understand software development particularly well. (The modern business schools seem to teach that a good manager can manage absolutely anything - a naked lie, but a lie which many of us have witnessed firsthand - i.e. the PHB.) There has arisen a generation which knows finance better than technology,. and this crowd is in control. As such, the possibilities of what actually happened are not apparent to them, and they can only see one-way transfers of their precious code to the unworthies, and not that it well could have occurred the other way around. The fact that CVS repositories could actually prove such a thing has escaped their notice. Discovery should be a hard time for this crowd.
All we see now is posturing and bluff. But in this battle I would say that IBM is actually the cat and SCO the mouse. IBM now gets to choose whether to go for the quick kill, or play with their mouse for while first. SCO will go down - and will not need to be bought off to do so. It is merely a matter of what torment they will endure in the process.
You are correct, of course.
But it does beg the question: Were all versions before the "Pro" just amateur?
The ability to run one or more concurrent instances of Linux (or whatever, quite frankly) internally to one of the Xilinx Virtex II parts is seriously amazing. Ignore the board it comes on for development for now - that is just cruft. The Virtex II is probably the most powerful instantly reconfigurable DSP engine in existence (think audio, video manipulation at real time speeds). They have internal hardware to perform from 16 to 128 simultaneous 16x16 multiply/accumulate operations simultaneously, _in_one_clock_cycle_. And if you don't like what it is doing, you can change it, time and time again, forever. Raw Power. Complete Reconfigurability. Sweet!
Combine this kind of power with multiple PPC processors on the same die, and the possibilities are incredible. The big difficulty is that the operation of the hardware and software can be so tightly tied together that it is difficult to program and debug. Everything is controlled by software (both the software and the VHDL or Verilog based FPGA code) and so the possibilities are limitless.
Kudos to Jim Ready and the folks at Monta Vista for supporting this kind of device with development tools for Linux.
I have to remind my fellow photographers to charge enough for their work. Why? Professional photography is a luxury item, one which no one actually needs, but some people actually want enough to pay for. Think of it: absolutely no one goes hungry (except the professional photographer) if a photograph is not made. It is a luxury.
Think of it: anyone can go to Wal-Mart and buy a rather good single-use camera, and if they are careful, get excellent results. For less than US$20, they can have the camera and double prints, and get the prints back today. Why, then, does anyone go to a professional photographer to have done what they could do themselves? The reason is usually that they realize that they cannot be in the pictures they shoot themselves, and they need someone to do the job for them, preferably someone who has done it enoough before to be sure to get the expected pictures of The Big Event. So we as professional photographers provide a service, and a luxury at that.
Now most of us have changed over from film to digital imaging, and the originals are just jpg files. Why don't we just sell the original files? Well, I have argued with some of the locals that we could do that, but the average person is not prepared to deal with the 500+ multi-megabyte jpg files generated at a simple wedding. The raw files are just that - raw. We typically spend much of a week's time doing the pre-print processing before a customer sees them. There is color correction, cropping, eliminating the glasses glare, swapping heads - all of the usual retouching is done, and from a 500 raw picture wedding we may present 150 images.
Now we gave up on paper proofing some time ago. We show the images by projection (Epson LCD projectors are great for this), and the customer gets to decide what she (usually it is a woman) wants - we provide just about any print format anyone could want, and can frame them, bind them into books, and such. Whatever prints were ordered are delivered at the end of the process.
An average wedding takes two of us probably 30 hours each to do right. The results are gorgeous, by the way. Could we do weddings for less? Sure, but not at this quality level. Could we just shoot weddings and hand the customer a couple of CDs of jpg files and our blessings to use however they wish? Sure, but the results in general would not be as good, as most folks (even uber-geeks) are not all that good with gimp or photoshop to do the post-processing themselves.
In the end, you get what you pay for. Professional photography is a luxury item.
Deal with it.
Marty
The local phone service I get (through my cable provider) comes with a bill broken out according to every mandatory fee and tax, and the mandatory fees and taxes are larger then the phone service cost itself.
Whatever they may be for, the combination of added fees and taxes on phone service is exhorbitant already. Adding them to other net services is just another revenue stream for someone else.
Phbbbbbt!
We have tried IPCop 0.1.1 at the office, and it has one very big advantage over using a general purpose distribution: it installs and comes up running very quickly. From inserting the CDROM to completion of the install on a typical system (200MHz Pentium with 64MB memory) it took about 14 minutes to having it running.
We use it as a three-way firewall with a DMZ, and that is stone-cold simple to install. Slick, with no problems.
Highly recommended!
IPCop is excellent for probably 90% of the firewall needs for individuals and small businesses. It is based on linux kernel 2.2.20 and ipchains. It is GPLed, has a quality web interface, and installs fast and easy. Furthermore, the user list is friendly and helpful. I downloaded the iso for it, wrote it to a cd, and then took about 15 minutes start to finish with the initial installation. After that, the fine tuning was handled over a very intuitive web interface. I would rate it a 9.5/10.
Many years ago I worked in a lab where shipboard sonar equipment was being built. Their cooling system for the high power units consisted of aluminum plates with embedded copper tubing carrying filtered sea water right through the rack of electronics. It carried away the heat quite well, but the whole thought of using sea water was antithetical to everything I would normally consider.
On the other hand, large high power vacuum tubes have been water cooled for many decades. The most impressive were the vapor-phase cooling units in which the cooling was done by boiling the water off of the external anode of the tube. The steam was then condensed back into water in an outdoor cooling tower.
The assumption that all development is in the corporate realm is most evident in this sort of article. Companies come and companies go, but the development of free software and hardware on which it can run goes on with or without corporate backing.
We have a project to develop free hardware designs to which free software is being ported, with no corporate backing. For an example of an embedded controller to which Linux is being ported, look here:
http://freeio.org/library/toast.htm