Your statement is similar to condemning Intel's current lineup of processors for the old Pentium division bug. Or condemning a 2000 Chevy Cavalier, because you saw a 1983 model break down last week.
Of course a cheap PC would beat any computer whose memory was flawed. However, the UltraSPARC IIs have been fixed, and they are back to their 24/365 behavior.
There are other file managers that can be used with KDE that do not have built-in browsers I think.
Yes, I enjoy using/bin/sh within/usr/X11R6/bin/xterm. This makes a great file manager for KDE. It does have a "built-in" web browser, called lynx, but that is purely optional, too.
Printing $50,000 Solaris CD binders is a major source of profit for Sun...
Since when did Solaris media cost $50,000?
Today, $50,000 would be about 500 to 1000 licenses...this is probably the top 3% of Sun installations. I'd figure Sun gets more money from that other 97%.
There are reasons Sun still uses its own processor: scalability, ECC on all interfaces, secondary diagnostic busses, and binary compatibility.
If the AMD Hammer can guarantee the same uptime that the UltraSPARC processors do, then you have a point, but, in general, there is more to a computer than just the CPU.
Not when your professional reputation is at stake.
as well as getting all the power of Open Source, which is making Linux more powerful every single day.
Remember that most Open Source software works well on Solaris, also.
So what can Sun do?
Both. Solaris and Linux can each be a perfect fit for diffent Sun customers. Would I buy an entry-level server from Sun with Linux, if Linux were my OS preference? Certainly. Would I buy an entry-level server from Sun with Solaris, if Solaris were my OS preference? Certainly.
My point is that Sun is a hardware company, who produced Solaris to fully support their hardware. The hardware speaks for itself. CPU2000 zealots out there, who think the Pentium 4 is the Supreme Being, don't see the larger picture, which is that Sun hardware is typically very well-rounded and very well-engineered. They build their hardware from the CPU innards on out to be consistent and robust.
I do bet my reputation on a well-configured network of Sun servers. I could also bet my reputation on a network of Intel-based servers, but I know from experience this is a riskier choice. And this remains fairly independent of whether I am using Solaris or Linux, because the hardware has its own merits.
There is little or nothing Sun's OS can do now that Linux can't.
Careful, here.
Solaris and Linux each satisfy the needs of different markets.
Solaris allows Sun Enterprise and Sun Fire servers to break, be dismantled, and put back together--all while the server remains available to its users.
Solaris scales extremely efficiently with the size of the server. I would not hesitate to put Solaris on a server with hundreds of CPUs. Linux is probably more appropriate for a cluster of smaller servers.
Solaris also fully supports all of the subtle, yet important, RAS features that Sun servers have.
In short, if you have a Sun server, Solaris is the only OS that allows you to use it to its full potential. Linux is extremely useful in a wide variety of applications, but, when I last checked, the SPARC ports still didn't support the same scope of Sun hardware that Solaris does.
Sun will sooner or later have to realise that Linux will dominate the Unix OS market to an ever greater extent in the future.
Be aware that Linux isn't the only non-commercial UNIX OS in the world. There are other kernels, such as GNU HURD, or even other OS models, such as Plan 9, that may suprise us one day. Of course Sun would be negligent to be unaware of Linux, but Linux isn't the one-and-only thing Sun needs to keep tabs on.
If it does, then you should have no problem. RedHat has a very easy default ghostscript setup. Other popular systems should have something similar, or you can just roll your own ghostscript command line. Or you can buy a PostScript printer; they are pretty inexpensive now-a-days.
With ghostscript you can also print PostScript on Windows, so you can keep a consistent system going across platforms.
From what I've heard, high-test medical software runs on mainframe-type machines. You may need to go with something from IBM, just because that is the platform the software is designed for. Medical computing is old enough that its roots really are in the mainframe era, not the relatively new UNIX/Linux era. Keep in mind that you won't need a big mainframe, just a desk-side model.
Also, I don't know if it was MediSoft, but I knew someone who worked with a Windows based billing program at one time--and hated it. It was down more than once every day; the the lost productivity probably made it much much more expensive than just shelling out the cash for a real mainframe. Just think of it, a whole medical lab stopped in its tracks while one person argues with tech support over the phone. Terrible.
If a bank manager decided to leave the vault and the office doors wide open for one night, and the following day everything had been stolen, who is to be held responsible? The bank manager or the thieves?
The answer is all of them.
Unfortunately, liability has yet to be soundly worked out in the software industry, and Microsoft continues to walk free for what should be considered criminal negligence. "How were they negligent", someone asks? Well, marketing software as an idiot-proof point-and-click haven, when it clearly isn't, is simply negligent.
Microsoft really should be taken to court by those damaged by these viruses. Microsoft can't claim they didn't know about security holes, when the holes have been so obvious for years, now. This is no different than an auto company putting prices on passenger's lives to improve margins or tobacco companies continuing to sell cigarettes when they are clearly harmful. Microsoft simply cares more about profit than protecting its customers.
is for the World to begin the arduous and expensive task of removing Microsoft software from their computers.
The first step is to eliminate Outlook for e-mail. There are other options, even Emacs, that really aren't too user unfriendly.
The second step is to eliminate Office for shared documents. There are other options, perhaps Open Office, that will be less prone to viruses and will be more maintainable over time.
The third step is to begin evaluating other operating systems besides Windows. This is harder, because it will be difficult to replace all the software that was useful in Windows. Over time, however, a fairly comprehensive list can be developed, and a plan can be made to make the switch to a non-Windows OS.
The fourth step is to take the plunge and dump Windows entirely. This may be the hardest step, because this is where the most learning needs to take place. But it is just a matter of time before users adapt to the new environment.
This is what I have been doing at home and know it isn't easy to make a full transition. However, I have found adequate replacements for nearly everything and am pretty satisfied with the results.
This doesn't have to be an all-Free-all-the-time solution, either, because there really is a way to mix open and closed software to meet your needs. It just takes research, time, and patience to find that Microsoft really doesn't rule the world at all--they just want us to think they do.
It must depend on which portion of the goverment you work for. I work indirectly for the goverment and am continually amazed--and frustrated--at how politics and people who think they are in-the-know continually muck up what could be a very nice software project. I am also continually amazed that people are not usually promoted for merit but tenure (interesting that tenure is two letters away from manure!). Sadly, this often happens in the private sector, too. Oh, well.
Actually the speed improvement is due to javac not Ant, and the same result can be achieved in a makefile. The difference, here, is that Ant is intended for Java programs and has built in this optimization already.
This is unheard of in the US... Who doesn't finish high-school? Who hasn't used a computer by then?
Perhaps you need to visit the prosperous southern U.S., where every student is equally denied access to a good education? People not finishing high school in the U.S.A. might be more common than you think.
One thing I really don't understand is why Ant reinvents the wheel.
Many people have been using makefiles successfully for years, and, now, part of the Java community comes out and says, "Make sucks. Let's just throw it away (even though it is really mature) and start over from scratch (instead of implementing Make in Java)."
I have found that the common arguments against Make really aren't very strong. The tab issue is really minor. The only issue that bugs me is that parallel builds aren't possible with most implementations. There are other more theoretical issues, such as the dependency graphing, etc., but I don't see Ant taking on these much more complex issues.
My other peeve is that Ant is fairly Java-specific, while Make is really language-independent (I use it for Java, C, C++, release management, testing, etc.).
Is the popularity of Ant really just a case of jumping on the bandwagon?
Companies in the U.S. do this, too. It really boils down to whether theft can cripple the company or whether materials can be resold for large amounts of money. For example, I knew someone who worked at a catalyst manufacturer (chemicals, not routers) and had to go through metal detectors and searches before leaving the factory. The reason: the metals that went into the catalysts were so valuable that a person could become wealthy by smuggling small amounts out over time.
... Microsoft Office is one of the best software packages ever produced.
At first, I thought that this may have been valid at one time or another, but, then, I realized it isn't. A recent example: even with its auto-super-helpful features disabled, Word 2000 still makes "decisions" for me and crashes frequently. An earlier example: I remember when formatting a nice paper was a piece of cake in Lotus Ami Pro (or even LaTeX, for that matter), but when I tried to do the same basic operations, such as foot notes, in Word, I wasted an entire evening trying. I quickly found that the Word user interface is really pretty crappy.
You may have had better experience with Excel, and that is fine. But, as a whole, Microsoft Office is at best average and more likely mediocre. It solves many problems poorly and a very few problems well. I'd argue that it has caused more problems via its unreliable file maintenance (trust me, I've seen files disappear and files that simply go bad during normal use) than it has solved through its encyclopedic set of "features".
What will the economic effects for the software industry be if software such as Office (which is a huge driver for software revenue and profits) is given away for free?
In the long run, the economic effect can only be positive.
For starters, the Market has made it very clear that office productivity suites are desirable. This is evident in the massively wide-spread use of them and the many many different office suites that came and went.
However, Microsoft, over the years, has taken steps to ensure that its office suite is the one used by pretty much everyone. The end result is that Microsoft Office file formats have become a defacto standard communications protocol. One may think that this is a good thing, but I don't.
Currently, this widely used communications protocol is proprietary, which means you really don't know how your information is transmitted. You don't know how to fix it if it breaks, so, if your file becomes corrupted, a text editor won't be able to save you (let's hope your IT dept. keeps good daily backups). If you have a very large number of documents, you don't know how to query them efficiently for various data (how does one search what are effectively random binary files?). How will you access these files in ten years, when there isn't even a guarantee that Microsoft will still be here (this is true of any company)? If Microsoft dissolves and you didn't save the CD-ROM for Office 95...well, those files might just as well be deleted.
My argument, here, is that using the proprietary Microsoft communications protocol in Office is risky. Very risky. For some reason, our society at large has not grasped that our important data is simply not accessible to us without Microsoft getting it for us. From a risk management point of view, this is a terrible position for any company or individual to be in.
Reducing this risk is why I choose do document software, write e-mail, take notes, etc. in plain text or plain-text-based file formats, such as TeX and SGML. This way, if I have the file in hand--but the software that created it is unavailable--it is trivial for me to write my own program to decode it again if all else fails. From a risk management point of view, this is nearly ideal.
I think many people spend more time than they will admit to dealing with this proprietary communications protocol. Dealing with subtle incompatibilties. Dealing with data corruption. Doing everything manually, for cripes sake, when a text-based format would allow automation through scripts.
When--not if--we are finally using open file formats, such as the XML formats with Open Office, we will notice a general improvement in the quality of our communications. These open files allow for flexibility that can be invaluable when large amounts of data need to be processed or when the office suite isn't available and we just need a few tidbits of information. It may not be possible to quantify the impact of these improvements, but they will certainly be good for our society.
How many engineers (and for that sake.. people like managers and support staff) are involved in the MS Office product? Tens of thousands make their living of that product.
Out of the millions of engineers, managers, and support staff in the world, a few thousand displaced is kinda sad, but it really is a small number of people. The people who used to make a living selling Microsoft Office will adjust. What happens to them is no different than what happens to realestate agents or car salesmen when segments of our economy take a dive. Volatility is nothing new. That's why community colleges are successful. They are an integral part of retraining our workforce as the markets evolve.
While I must say that it is very nice to have free software such as operating systems, compilers etc available instead of having to buy (or copy..) expensive software, I think that this is doing more damage than good to the people involved in software development.
Free Software is a real part of the software industry. Yes, it does affect the commercial parts of that industry. However, I, a software engineer, don't mind. I will adapt to the presence of Free Software. All Free Software does is alter what is marketable. I will find something new to produce and sell, and I might be successful at it. This is how the free market worked before and nothing has changed.
We shouldn't try to do something foolish, such as suppressing Free Software in the hopes of creating jobs, when the Market obviously wants something else. This is what the U.S. goverment tries frequently often with debatable results.
I used to think that it was great to be able to set up an entire advanced Internet-system for free...
It still is. The Internet is about community; it is part of our civilisation's infrastructure.
... do you all want to work as sysadmins on Linux and databaseadmins on MySQL instead of software engineers & technical managers on projects that aim to sell the software you have created?
No. As I stated above, there will always be marketable software that remains unwritten. There will always be new problems that need solving, and there will always be people (sometimes called entrepeneurs) who are up to the task.
Unsupported? The collective experience among people who purchase and employ Sun workstations, for example, speaks for itself. In my office, 7-year-old Sun workstations are still useful, are still upgradable, and they are as predictable as night following day. Meanwhile, the 7-year-old PCs...wait a minute...where are they?
The RISC-based hardware by folks like Sun, IBM, SGI, HP, whatever is a real investment. In a hard-working business, their flexibility and durability pays both in time saved and frustration saved. I'm not making this up, believe me!
Why? They are really well engineered and are solid workhorses. They really aren't expensive, either (unless you buy directly from Sun). Also, an Ultra 2 is a much better workstation than an Ultra 5 or 10, unless you really need a PCI bus.
An Ultra 60, while sounding considerably more impressive than the modest 10, isn't worth much more than a 10 unless it has either the Elite 3D graphics card or dual processors
Ultra 60s come with dual-channel UltraSCSI controllers, 4MB cache per CPU, and dual UPA framebuffer slots. Ultra 10s have IDE disks and 2MB cache per CPU. The Ultra 60 also has the CPUs mounted on their own daughter cards, which makes it considerably more flexible than the Ultra 10. They really are different beasts.
It isn't just the technical shortcomings of the x86 processor. Consider, also, that most PCs are hodge-podge assemblies of cheap components. One reason why many people choose brand-name RISC workstations is that they really are simpler to work with, they break less, they perform consistently, they have longer useful lifetimes, and they have real firmware.
I have been using Sun workstations for a few years, now, and when my PC breaks, I'm replacing it with an Ultra. What about games, someone asks? Why, that's why I have a PlayStation. What about Windows, someone asks? Well, Windows is a toy and has nothing I need, and I need to get real work done.
Take a look at the secondary market of Sun hardware. For less than $1000, you can have an Ultra 2 workstation with SCSI disks and SMP capability. Or you get an older SPARCstation 10 or 20 that still supports SCSI-2 and up to 4 CPUs.
While these computers won't win CPU2000 flame wars, they really are beautiful machines that have full firmware, super-clean layout, and integrated Ethernet and SCSI. Also, you can run Solaris 8, Linux, NetBSD, and OpenBSD on them. They make great personal workstations (I have KDE on a 40MHz SS10--still usable) or great file or web servers. On top of that, they run forever (my SS10 is now 10 years old). Because they're SCSI, you can put big disks into them (9GB, no problem) and connect external tapes, CDROMs and Zip drives to them. Even the old ones support gobs of ram (at least 512MB). If you can figure it out, the SS10s even have integrated ISDN interfaces.
In short, they are a joy to work with.
There are many vendors, so be sure to get several quotes. Some vendors are arrogant and still think they can charge an arm and a leg for old hardware. Don't let them get you down, because you will find a good price if you are persistent. Also, try eBay or other auctions.
Your talking about regulating what a company can sell. Tell me this isn't about regulation again?
Others may have argued for regulating Microsoft, I have not. I simply want more options for consumers. The DOJ may find that regulations are a way to achieve this, but there are other emerging market forces, such as GNOME and KDE, that may do this for me.
You do have 10 models to choose from. You have FreeBSD, OS2, Linux, Solaris X86, FreeDos, DOS, CP/M, Netware, Darwin and tons of other OS's to choose from.
The original argument concerned consumer-grade operating systems, such as Windows and MacOS. UNIX and its derivatives, for example, are excellent operating systems, but they are simply not intended for Mr. and Mrs. Average Consumer. For these people, there is still only one dominating choice: Windows. MacOS is still a small player. Other promising consumer-grade options, such as OS/2 and BeOS, were simply crushed by the market dominance of Windows.
I can't take the spark plugs out of my rx-7 and fit them in my tiburon.
The point is that Microsoft wants to own the roads themselves. This is much more fundamental than whether certain components are interchangable, this is an issue of whether different people can even share basic information without Microsoft software intervening.
you have problems if your going to commit suicide because of windows
If Bill Gates is able to fufill his visions, then we will basically be living in an information dictatorship--one that I will certainly be looking for a way out of if it occurs. This doesn't imply suicide; rather, I may just stop using computers and change professions.
Prove to me how windows is a Kludge?
Why is my Windows 2000 installation directory nearly 900MB in size? Why is it comprised of 40 million lines of source code? How many tens of thousands of known bugs are there? How many hundreds of thousands of unknown bugs are there? How many thousands of security holes are there? Why can I not uninstall the software I don't want? What is that in the registry? What's with the multi-rooted file system?...
From a software engineering standpoint, this is a kludge, where the complexity is simply not justified. There is no way I would use Windows in an application where someone's life depended on it. It's hard enough to see my family and friends trust their important information with it.
It is wrong for microsoft to give away internet explorer but it is fundamentally right for people to give away an entire OS for free?
Microsoft crossed the line, where they used IE to dominate a market. Others package things or give some things away for free as a matter of survival in a competitive marketplace. There really is a difference between "value added" and "value mandated".
Yes, you can choose what kind of cars you want, where to buy cars, what color you want it and low and behold when you get it home it is still a car no matter who built it.
With Microsoft, they choose what you want and what color it comes in.
But *I* for one don't want the computer industry regulated like the car industry.
This isn't about regulation. It's about competition.
I don't want DellXP, CompaqXP, MSXP, GatewayXP. I don't want a stripped down car either.
Many other people do. Having options makes some decisions harder, but our lives are better as a result. I'd rather have 10 models to choose from than one. Let the companies scramble for my business, and let me put them in their place. This is what happens when the free market is in good health.
Don't let this choice BS get to your head.
Without choice, is my life worth living?
Windows is CHEAP, Affordable and RUNS JUST FINE.
Windows is not cheap, and it is a kludge. It does not run fine. In fact, it's behavior is so inconsistent sometimes that I want to punch my monitor.
You can CHOOSE YOUR OWN GODDAMN ROAD THOUGH.
Not when all roads lead to Microsoft.
Just remember you do get what you pay for, and you don't get something for nothing.
When what I'm buying is selling for its true market value. Operating systems used to be expensive, but the market has spoken. Other companies have accepted this fact. For example, I can get Solaris, RedHat Linux, and OpenBSD media for less than $50 (one of these used to be really expensive).
However, there is so much competition in the auto industry that the quality of new cars has improved greatly over the years. People shopping for cars, now, have a pretty level field to choose from, and they bicker over prices and features. In today's auto market, the consumer has the edge over the salespeople ("You won't come down in price?? Well, I just go across the street.").
How many models of the standard 4-door family mover are there in the U.S.A.: GM has a few, Ford has a few, Chrysler has a few, Toyota, Nissan, Honda, Hyundai, Kia, Daewoo, VW, BMW, Volvo, Saab, Mercedes, and more I can't remember.
How many models of consumer-grade operating systems are there: Microsoft has a few (>85% share), Apple has a couple (<15% share) ,... hmmm, that is about it.
Consumer to Microsoft salesperson, "You won't come down in price?? Well, okay, who do I make the check out to?"
Also, no one is forced to buy a new car. A technically-inclined person can go scavenge a junk yard and rebuild a classic. The laws work so that he can get by with older technology, too, with just a few restrictions.
The car-road interface has been standardized well enough, that we don't have to worry about suddenly having to drive on rails or fly on tethers. In software, however, Microsoft wants to own the roads and dictate that only Microsoft tires can achieve traction on those roads. They want us to be under their control.
Slashdot Mirror
The e-cache issue is old and has been dealt with.
Your statement is similar to condemning Intel's current lineup of processors for the old Pentium division bug. Or condemning a 2000 Chevy Cavalier, because you saw a 1983 model break down last week.
Of course a cheap PC would beat any computer whose memory was flawed. However, the UltraSPARC IIs have been fixed, and they are back to their 24/365 behavior.
There are other file managers that can be used with KDE that do not have built-in browsers I think.
/bin/sh within /usr/X11R6/bin/xterm. This makes a great file manager for KDE. It does have a "built-in" web browser, called lynx, but that is purely optional, too.
Yes, I enjoy using
Printing $50,000 Solaris CD binders is a major source of profit for Sun...
Since when did Solaris media cost $50,000?
Today, $50,000 would be about 500 to 1000 licenses...this is probably the top 3% of Sun installations. I'd figure Sun gets more money from that other 97%.
There are reasons Sun still uses its own processor: scalability, ECC on all interfaces, secondary diagnostic busses, and binary compatibility.
If the AMD Hammer can guarantee the same uptime that the UltraSPARC processors do, then you have a point, but, in general, there is more to a computer than just the CPU.
Cheaper
Not any more.
runs on cheaper hardware
Not when your professional reputation is at stake.
as well as getting all the power of Open Source, which is making Linux more powerful every single day.
Remember that most Open Source software works well on Solaris, also.
So what can Sun do?
Both. Solaris and Linux can each be a perfect fit for diffent Sun customers. Would I buy an entry-level server from Sun with Linux, if Linux were my OS preference? Certainly. Would I buy an entry-level server from Sun with Solaris, if Solaris were my OS preference? Certainly.
My point is that Sun is a hardware company, who produced Solaris to fully support their hardware. The hardware speaks for itself. CPU2000 zealots out there, who think the Pentium 4 is the Supreme Being, don't see the larger picture, which is that Sun hardware is typically very well-rounded and very well-engineered. They build their hardware from the CPU innards on out to be consistent and robust.
I do bet my reputation on a well-configured network of Sun servers. I could also bet my reputation on a network of Intel-based servers, but I know from experience this is a riskier choice. And this remains fairly independent of whether I am using Solaris or Linux, because the hardware has its own merits.
There is little or nothing Sun's OS can do now that Linux can't.
Careful, here.
Solaris and Linux each satisfy the needs of different markets.
Solaris allows Sun Enterprise and Sun Fire servers to break, be dismantled, and put back together--all while the server remains available to its users.
Solaris scales extremely efficiently with the size of the server. I would not hesitate to put Solaris on a server with hundreds of CPUs. Linux is probably more appropriate for a cluster of smaller servers.
Solaris also fully supports all of the subtle, yet important, RAS features that Sun servers have.
In short, if you have a Sun server, Solaris is the only OS that allows you to use it to its full potential. Linux is extremely useful in a wide variety of applications, but, when I last checked, the SPARC ports still didn't support the same scope of Sun hardware that Solaris does.
Sun will sooner or later have to realise that Linux will dominate the Unix OS market to an ever greater extent in the future.
Be aware that Linux isn't the only non-commercial UNIX OS in the world. There are other kernels, such as GNU HURD, or even other OS models, such as Plan 9, that may suprise us one day. Of course Sun would be negligent to be unaware of Linux, but Linux isn't the one-and-only thing Sun needs to keep tabs on.
I thought the PS2 Linux kit was $200.
If it's an honest mistake, then, okay--just check the numbers better next time.
If you are presenting misinformation as a FUD tatic, then just go home, stick your head in the toilet, and blow bubbles until you pass out.
Does Open Office not even print to PostScript?
If it does, then you should have no problem. RedHat has a very easy default ghostscript setup. Other popular systems should have something similar, or you can just roll your own ghostscript command line. Or you can buy a PostScript printer; they are pretty inexpensive now-a-days.
With ghostscript you can also print PostScript on Windows, so you can keep a consistent system going across platforms.
From what I've heard, high-test medical software runs on mainframe-type machines. You may need to go with something from IBM, just because that is the platform the software is designed for. Medical computing is old enough that its roots really are in the mainframe era, not the relatively new UNIX/Linux era. Keep in mind that you won't need a big mainframe, just a desk-side model.
Also, I don't know if it was MediSoft, but I knew someone who worked with a Windows based billing program at one time--and hated it. It was down more than once every day; the the lost productivity probably made it much much more expensive than just shelling out the cash for a real mainframe. Just think of it, a whole medical lab stopped in its tracks while one person argues with tech support over the phone. Terrible.
If a bank manager decided to leave the vault and the office doors wide open for one night, and the following day everything had been stolen, who is to be held responsible? The bank manager or the thieves?
The answer is all of them.
Unfortunately, liability has yet to be soundly worked out in the software industry, and Microsoft continues to walk free for what should be considered criminal negligence. "How were they negligent", someone asks? Well, marketing software as an idiot-proof point-and-click haven, when it clearly isn't, is simply negligent.
Microsoft really should be taken to court by those damaged by these viruses. Microsoft can't claim they didn't know about security holes, when the holes have been so obvious for years, now. This is no different than an auto company putting prices on passenger's lives to improve margins or tobacco companies continuing to sell cigarettes when they are clearly harmful. Microsoft simply cares more about profit than protecting its customers.
is for the World to begin the arduous and expensive task of removing Microsoft software from their computers.
The first step is to eliminate Outlook for e-mail. There are other options, even Emacs, that really aren't too user unfriendly.
The second step is to eliminate Office for shared documents. There are other options, perhaps Open Office, that will be less prone to viruses and will be more maintainable over time.
The third step is to begin evaluating other operating systems besides Windows. This is harder, because it will be difficult to replace all the software that was useful in Windows. Over time, however, a fairly comprehensive list can be developed, and a plan can be made to make the switch to a non-Windows OS.
The fourth step is to take the plunge and dump Windows entirely. This may be the hardest step, because this is where the most learning needs to take place. But it is just a matter of time before users adapt to the new environment.
This is what I have been doing at home and know it isn't easy to make a full transition. However, I have found adequate replacements for nearly everything and am pretty satisfied with the results.
This doesn't have to be an all-Free-all-the-time solution, either, because there really is a way to mix open and closed software to meet your needs. It just takes research, time, and patience to find that Microsoft really doesn't rule the world at all--they just want us to think they do.
I'm glad I work for the government now....
The grass is always greener....
It must depend on which portion of the goverment you work for. I work indirectly for the goverment and am continually amazed--and frustrated--at how politics and people who think they are in-the-know continually muck up what could be a very nice software project. I am also continually amazed that people are not usually promoted for merit but tenure (interesting that tenure is two letters away from manure!). Sadly, this often happens in the private sector, too. Oh, well.
Actually the speed improvement is due to javac not Ant, and the same result can be achieved in a makefile. The difference, here, is that Ant is intended for Java programs and has built in this optimization already.
For example,
all: Class1 Class2
javac -classpath . `cat ClassList.txt`
rm ClassList.txt
Class1: Class1.java
printf "Class1.java " >> ClassList.txt
Class2: Class2.java
printf "Class2.java " >> ClassList.txt
This is unheard of in the US... Who doesn't finish high-school? Who hasn't used a computer by then?
Perhaps you need to visit the prosperous southern U.S., where every student is equally denied access to a good education? People not finishing high school in the U.S.A. might be more common than you think.
One thing I really don't understand is why Ant reinvents the wheel.
Many people have been using makefiles successfully for years, and, now, part of the Java community comes out and says, "Make sucks. Let's just throw it away (even though it is really mature) and start over from scratch (instead of implementing Make in Java)."
I have found that the common arguments against Make really aren't very strong. The tab issue is really minor. The only issue that bugs me is that parallel builds aren't possible with most implementations. There are other more theoretical issues, such as the dependency graphing, etc., but I don't see Ant taking on these much more complex issues.
My other peeve is that Ant is fairly Java-specific, while Make is really language-independent (I use it for Java, C, C++, release management, testing, etc.).
Is the popularity of Ant really just a case of jumping on the bandwagon?
Companies in the U.S. do this, too. It really boils down to whether theft can cripple the company or whether materials can be resold for large amounts of money. For example, I knew someone who worked at a catalyst manufacturer (chemicals, not routers) and had to go through metal detectors and searches before leaving the factory. The reason: the metals that went into the catalysts were so valuable that a person could become wealthy by smuggling small amounts out over time.
... Microsoft Office is one of the best software packages ever produced.
... do you all want to work as sysadmins on Linux and databaseadmins on MySQL instead of software engineers & technical managers on projects that aim to sell the software you have created?
At first, I thought that this may have been valid at one time or another, but, then, I realized it isn't. A recent example: even with its auto-super-helpful features disabled, Word 2000 still makes "decisions" for me and crashes frequently. An earlier example: I remember when formatting a nice paper was a piece of cake in Lotus Ami Pro (or even LaTeX, for that matter), but when I tried to do the same basic operations, such as foot notes, in Word, I wasted an entire evening trying. I quickly found that the Word user interface is really pretty crappy.
You may have had better experience with Excel, and that is fine. But, as a whole, Microsoft Office is at best average and more likely mediocre. It solves many problems poorly and a very few problems well. I'd argue that it has caused more problems via its unreliable file maintenance (trust me, I've seen files disappear and files that simply go bad during normal use) than it has solved through its encyclopedic set of "features".
What will the economic effects for the software industry be if software such as Office (which is a huge driver for software revenue and profits) is given away for free?
In the long run, the economic effect can only be positive.
For starters, the Market has made it very clear that office productivity suites are desirable. This is evident in the massively wide-spread use of them and the many many different office suites that came and went.
However, Microsoft, over the years, has taken steps to ensure that its office suite is the one used by pretty much everyone. The end result is that Microsoft Office file formats have become a defacto standard communications protocol. One may think that this is a good thing, but I don't.
Currently, this widely used communications protocol is proprietary, which means you really don't know how your information is transmitted. You don't know how to fix it if it breaks, so, if your file becomes corrupted, a text editor won't be able to save you (let's hope your IT dept. keeps good daily backups). If you have a very large number of documents, you don't know how to query them efficiently for various data (how does one search what are effectively random binary files?). How will you access these files in ten years, when there isn't even a guarantee that Microsoft will still be here (this is true of any company)? If Microsoft dissolves and you didn't save the CD-ROM for Office 95...well, those files might just as well be deleted.
My argument, here, is that using the proprietary Microsoft communications protocol in Office is risky. Very risky. For some reason, our society at large has not grasped that our important data is simply not accessible to us without Microsoft getting it for us. From a risk management point of view, this is a terrible position for any company or individual to be in.
Reducing this risk is why I choose do document software, write e-mail, take notes, etc. in plain text or plain-text-based file formats, such as TeX and SGML. This way, if I have the file in hand--but the software that created it is unavailable--it is trivial for me to write my own program to decode it again if all else fails. From a risk management point of view, this is nearly ideal.
I think many people spend more time than they will admit to dealing with this proprietary communications protocol. Dealing with subtle incompatibilties. Dealing with data corruption. Doing everything manually, for cripes sake, when a text-based format would allow automation through scripts.
When--not if--we are finally using open file formats, such as the XML formats with Open Office, we will notice a general improvement in the quality of our communications. These open files allow for flexibility that can be invaluable when large amounts of data need to be processed or when the office suite isn't available and we just need a few tidbits of information. It may not be possible to quantify the impact of these improvements, but they will certainly be good for our society.
How many engineers (and for that sake.. people like managers and support staff) are involved in the MS Office product? Tens of thousands make their living of that product.
Out of the millions of engineers, managers, and support staff in the world, a few thousand displaced is kinda sad, but it really is a small number of people. The people who used to make a living selling Microsoft Office will adjust. What happens to them is no different than what happens to realestate agents or car salesmen when segments of our economy take a dive. Volatility is nothing new. That's why community colleges are successful. They are an integral part of retraining our workforce as the markets evolve.
While I must say that it is very nice to have free software such as operating systems, compilers etc available instead of having to buy (or copy..) expensive software, I think that this is doing more damage than good to the people involved in software development.
Free Software is a real part of the software industry. Yes, it does affect the commercial parts of that industry. However, I, a software engineer, don't mind. I will adapt to the presence of Free Software. All Free Software does is alter what is marketable. I will find something new to produce and sell, and I might be successful at it. This is how the free market worked before and nothing has changed.
We shouldn't try to do something foolish, such as suppressing Free Software in the hopes of creating jobs, when the Market obviously wants something else. This is what the U.S. goverment tries frequently often with debatable results.
I used to think that it was great to be able to set up an entire advanced Internet-system for free...
It still is. The Internet is about community; it is part of our civilisation's infrastructure.
No. As I stated above, there will always be marketable software that remains unwritten. There will always be new problems that need solving, and there will always be people (sometimes called entrepeneurs) who are up to the task.
Unsupported? The collective experience among people who purchase and employ Sun workstations, for example, speaks for itself. In my office, 7-year-old Sun workstations are still useful, are still upgradable, and they are as predictable as night following day. Meanwhile, the 7-year-old PCs...wait a minute...where are they?
The RISC-based hardware by folks like Sun, IBM, SGI, HP, whatever is a real investment. In a hard-working business, their flexibility and durability pays both in time saved and frustration saved. I'm not making this up, believe me!
the Ultra 2s rather suck
Why? They are really well engineered and are solid workhorses. They really aren't expensive, either (unless you buy directly from Sun). Also, an Ultra 2 is a much better workstation than an Ultra 5 or 10, unless you really need a PCI bus.
An Ultra 60, while sounding considerably more impressive than the modest 10, isn't worth much more than a 10 unless it has either the Elite 3D graphics card or dual processors
Ultra 60s come with dual-channel UltraSCSI controllers, 4MB cache per CPU, and dual UPA framebuffer slots. Ultra 10s have IDE disks and 2MB cache per CPU. The Ultra 60 also has the CPUs mounted on their own daughter cards, which makes it considerably more flexible than the Ultra 10. They really are different beasts.
It isn't just the technical shortcomings of the x86 processor. Consider, also, that most PCs are hodge-podge assemblies of cheap components. One reason why many people choose brand-name RISC workstations is that they really are simpler to work with, they break less, they perform consistently, they have longer useful lifetimes, and they have real firmware.
I have been using Sun workstations for a few years, now, and when my PC breaks, I'm replacing it with an Ultra. What about games, someone asks? Why, that's why I have a PlayStation. What about Windows, someone asks? Well, Windows is a toy and has nothing I need, and I need to get real work done.
Take a look at the secondary market of Sun hardware. For less than $1000, you can have an Ultra 2 workstation with SCSI disks and SMP capability. Or you get an older SPARCstation 10 or 20 that still supports SCSI-2 and up to 4 CPUs.
While these computers won't win CPU2000 flame wars, they really are beautiful machines that have full firmware, super-clean layout, and integrated Ethernet and SCSI. Also, you can run Solaris 8, Linux, NetBSD, and OpenBSD on them. They make great personal workstations (I have KDE on a 40MHz SS10--still usable) or great file or web servers. On top of that, they run forever (my SS10 is now 10 years old). Because they're SCSI, you can put big disks into them (9GB, no problem) and connect external tapes, CDROMs and Zip drives to them. Even the old ones support gobs of ram (at least 512MB). If you can figure it out, the SS10s even have integrated ISDN interfaces.
In short, they are a joy to work with.
There are many vendors, so be sure to get several quotes. Some vendors are arrogant and still think they can charge an arm and a leg for old hardware. Don't let them get you down, because you will find a good price if you are persistent. Also, try eBay or other auctions.
Your talking about regulating what a company can sell. Tell me this isn't about regulation again?
...
Others may have argued for regulating Microsoft, I have not. I simply want more options for consumers. The DOJ may find that regulations are a way to achieve this, but there are other emerging market forces, such as GNOME and KDE, that may do this for me.
You do have 10 models to choose from. You have FreeBSD, OS2, Linux, Solaris X86, FreeDos, DOS, CP/M, Netware, Darwin and tons of other OS's to choose from.
The original argument concerned consumer-grade operating systems, such as Windows and MacOS. UNIX and its derivatives, for example, are excellent operating systems, but they are simply not intended for Mr. and Mrs. Average Consumer. For these people, there is still only one dominating choice: Windows. MacOS is still a small player. Other promising consumer-grade options, such as OS/2 and BeOS, were simply crushed by the market dominance of Windows.
I can't take the spark plugs out of my rx-7 and fit them in my tiburon.
The point is that Microsoft wants to own the roads themselves. This is much more fundamental than whether certain components are interchangable, this is an issue of whether different people can even share basic information without Microsoft software intervening.
you have problems if your going to commit suicide because of windows
If Bill Gates is able to fufill his visions, then we will basically be living in an information dictatorship--one that I will certainly be looking for a way out of if it occurs. This doesn't imply suicide; rather, I may just stop using computers and change professions.
Prove to me how windows is a Kludge?
Why is my Windows 2000 installation directory nearly 900MB in size? Why is it comprised of 40 million lines of source code? How many tens of thousands of known bugs are there? How many hundreds of thousands of unknown bugs are there? How many thousands of security holes are there? Why can I not uninstall the software I don't want? What is that in the registry? What's with the multi-rooted file system?
From a software engineering standpoint, this is a kludge, where the complexity is simply not justified. There is no way I would use Windows in an application where someone's life depended on it. It's hard enough to see my family and friends trust their important information with it.
It is wrong for microsoft to give away internet explorer but it is fundamentally right for people to give away an entire OS for free?
Microsoft crossed the line, where they used IE to dominate a market. Others package things or give some things away for free as a matter of survival in a competitive marketplace. There really is a difference between "value added" and "value mandated".
Yes, you can choose what kind of cars you want, where to buy cars, what color you want it and low and behold when you get it home it is still a car no matter who built it.
With Microsoft, they choose what you want and what color it comes in.
But *I* for one don't want the computer industry regulated like the car industry.
This isn't about regulation. It's about competition.
I don't want DellXP, CompaqXP, MSXP, GatewayXP. I don't want a stripped down car either.
Many other people do. Having options makes some decisions harder, but our lives are better as a result. I'd rather have 10 models to choose from than one. Let the companies scramble for my business, and let me put them in their place. This is what happens when the free market is in good health.
Don't let this choice BS get to your head.
Without choice, is my life worth living?
Windows is CHEAP, Affordable and RUNS JUST FINE.
Windows is not cheap, and it is a kludge. It does not run fine. In fact, it's behavior is so inconsistent sometimes that I want to punch my monitor.
You can CHOOSE YOUR OWN GODDAMN ROAD THOUGH.
Not when all roads lead to Microsoft.
Just remember you do get what you pay for, and you don't get something for nothing.
When what I'm buying is selling for its true market value. Operating systems used to be expensive, but the market has spoken. Other companies have accepted this fact. For example, I can get Solaris, RedHat Linux, and OpenBSD media for less than $50 (one of these used to be really expensive).
However, there is so much competition in the auto industry that the quality of new cars has improved greatly over the years. People shopping for cars, now, have a pretty level field to choose from, and they bicker over prices and features. In today's auto market, the consumer has the edge over the salespeople ("You won't come down in price?? Well, I just go across the street.").
... hmmm, that is about it.
How many models of the standard 4-door family mover are there in the U.S.A.: GM has a few, Ford has a few, Chrysler has a few, Toyota, Nissan, Honda, Hyundai, Kia, Daewoo, VW, BMW, Volvo, Saab, Mercedes, and more I can't remember.
How many models of consumer-grade operating systems are there: Microsoft has a few (>85% share), Apple has a couple (<15% share) ,
Consumer to Microsoft salesperson, "You won't come down in price?? Well, okay, who do I make the check out to?"
Also, no one is forced to buy a new car. A technically-inclined person can go scavenge a junk yard and rebuild a classic. The laws work so that he can get by with older technology, too, with just a few restrictions.
The car-road interface has been standardized well enough, that we don't have to worry about suddenly having to drive on rails or fly on tethers. In software, however, Microsoft wants to own the roads and dictate that only Microsoft tires can achieve traction on those roads. They want us to be under their control.
The vm-auto-folder-alist is basically a list of which fields to scan and what to do with classes of entries in those fields. A simple example is:
(setq vm-auto-folder-alist ("Sender:" ("mailing-list@domain" . "mailing-list.saved" ) ("mailing-list2@domain" . "mailing-list2.saved" ) ) ( "From:" ( "user@domain" . "user.saved" ) ( "your-e-mail@your-domain" . "sent_mail.saved" ) )
A more powerful example using regular expressiongs:
(setq vm-auto-folder-alist ("From:" ( "^.*@dot[.]bomb$" . "dot.bomb.saved" ) ) )
This will take every e-mail whose From field matches the expression and save it into the file, dot.bomb.saved.
I think this is by far the most useful and time-saving feature in VM, especially when subscribed to a high-volume mailing list.