1). I don't love the AWT, but it has its place. If nothing else we need the classes java.awt.Component and java.awt.Container so we can create our own lightweight components without inheriting unnecessary baggage from Swing.
2). Changing names for the sake of changing names is stupid, stupid, stupid. Just this morning I had to go back to using the deprecated methods getValue(), putValue(), etc. for a Session object because most programmers in my company are not using the very latest JDK. In a large company keeping everyone's IDEs up to date is a huge task that nobody has time for. I've seen several examples where deprecated methods were replaced with something else that was different only in name, and the name change was totally arbitrary (replacing "Url" with "URL" in a method name). I have to write code that is used by developers throughout the company and this kind of thing really bugs me.
I've been using eclipse for awhile and I like how SWT works in that application. You'd never know that app was written in Java if nobody told you. However, it does need more memory to run than Visual Age for Linux did. I'm well aware of this because I have developed on both on a Pentium with 64 Meg of memory (an Aptiva, so you can't add more memory). So the UI seems to be only part of the problem with Java.
I have used Swing apps like JEdit on machines with a lot of memory and have no complaints. They look as good as a Windows application and performance while they are running is decent. Startup time is slow on any Java program though.
I have also made my own lightweight components which are much less elaborate than Swing but seem to me to work well enough on a Pentium. Making your own Java ui components is easy and even fun, and I've wondered why more people don't try it.
I tend to agree with this article. The first time I ran X on Red Hat 5.2 I saw this very lame attempt to make Linux look like Windows 95. I was very disappointed. Soon I discovered AfterStep, then WindowMaker, and finally GNUstep, which gave me something really different from Windows but which was obviously well thought out. Scrollbars are on the LEFT of the thing being scrolled, instead of the right (where they get pushed off the edge of the screen). The scrollbar arrows are next to each other, instead of at opposite ends of the scrollbar. Menus are in their own windows, and can be moved around and torn off. Dockapps and the dock are neat too.
The new Apple OS is almost as nice, but they put the scrollbars on the wrong side.
I was frustrated that Java Swing's pluggable look and feels make everything look like Windows 95, so I started my own project to make Java applications with the look and feel of GNUstep possible.
If you are looking for something really different but still well thought out in user interfaces you should really check out WindowMaker and GNUstep.
What's funny about this is that awhile back my favorite niece was working on my computer (running WindowMaker) and asked me why I bothered with Linux since it seemed to her that it was just like Windows.
Of course that's the point of the article. Linux should be different from Windows and superior to it so that Windows users wonder what they might be missing by not running it. A more stable version of what they already have is not all that compelling.
I read a story awhile back where somebody claimed that a malicious programmer shut off the landing lights at an airport using the Internet. Nobody questioned that this had actually happened, or why the hell lights at an airport could be shut off over the Internet at all.
I can see a possible scenario for the Pacemaker though. You need a Pacemaker and two kinds are available. One is not connected to the Internet but is available only to the very wealthy. The other is connected at all times and can be shut down remotely if you don't keep up with the subscription payments for the software.
The idea of charging for binaries doesn't bug me that much. I've always paid for my SuSE distros and consider the money well spent. I also compile and install a lot of stuff by hand (mostly WindowMaker and GNUstep, etc.) The stuff I do by hand is the stuff I really care about, and it always takes longer than I think it will. OTOH, the stuff that comes with the distro generally just works. I can afford to take it for granted. That is worth more than the modest cost of the distro.
Maybe if the distro companies started using this approach (free source, charge for binaries) then we could get business people to understand that the "free" in free software is all about freedom, and that freedom is what drives the real cost savings.
I've also worked on a free software project and would be thrilled to death if anyone wanted to include it in a distro so it could be installed via RPM. I would not feel exploited in the slightest way. Heck, I'm giving the stuff away anyway, why wouldn't I be happy about being chosen for inclusion in a distro?
One thing that never seems to come up in these debates is how truly "closed" closed source is. If you take security through obscurity seriously, you'd have to run Microsoft like it was the Manhatten project. Consultants would have to be carefully screened. Employees would not be allowed access to the Internet. They could send email only within the company (otherwise they could email themselves any source code they'd like). They would have to be searched when entering the building and leaving, so they couldn't smuggle code out on diskettes or printed out on paper. Maybe they would have to work on diskless workstations.
It's pretty obvious that no company in the real world is like this, so we can assume that the Black Hats can see Windows source code. And of course there is the famous "shared source" program that makes Windows code available to colleges, etc.
In the open source and free software world we don't have the false sense of security that those in the Windows world may have. As far as I can tell this should make our stuff more secure in practice.
If there is anything that makes me unhappy with the popularity of IE it is that IE does a very poor job printing web pages. For the home user this is rarely a problem, but for corporate users this is a serious problem. When developing a corporate intranet I want to create reports on the web that a user can view or print out easily. IE fails miserably. Two examples:
1). When printing a table it will split a table row in the middle, printing the top half of the row on one page and the bottom half on the next page. Netscape never splits a row like this.
2). IE does the same thing with images. If you have an image near the bottom of a page it will get split across two pages, something Netscape doesn't do.
While Netscape is better than IE, it isn't enough better to compell people to switch.
Here is my printing wish list. If Microsoft was all that innovative they could have fixed this stuff long ago:
1). Give me a way to force a page feed when printing. This would have no effect when viewing pages online.
2). When printing a table, if the table runs over a page give me an option to repeat the table headings at the top of subsequent pages.
3). How about an option to designate sections of pages to be non-printable? In other words, you could print any page marked up this way and it would come out printer friendly.
4). How about an option to tell the printer to print the page landscape or portrait, so users could simply hit the Print button and get something that fit on the page the way the page designer intended?
The W3C created a much longer list of stuff to address printing web pages and as far as I can tell it has been ignored. I would be happy with the short list above. This list *might* be enough to give corporate users a reason to switch browsers.
I had a friend in college who used to do something like this with vinyl records. He'd buy a record from Kmart or someplace and it would be a record that looked interesting but that he hadn't heard. He'd take it home and listen, and if he liked the music he'd keep it. If he didn't he'd scratch the record with car keys and return it for a refund because it was damaged.
It would make much more sense to return a CD for a refund than it would to sell it to a used CD store, if you were really interested in ripping off somebody.
I don't think the survey is totally worthless but it ignores a few things:
1). Maturity is subjective. For instance, WindowMaker hasn't had a 1.0 release yet but I would consider it mature enough to use every day. It would make more sense to look at the top 500 or so downloaded projects to get a sense of what can be considered mature.
2). Small projects don't generate much discussion because in general it is too much of a hassle. I have to have an account on SourceForge to post to a discussion or subscribe to a mailing list. However, I can simply email the author on a small project and get a better chance of being heard. Only a large project with many developers would make it worthwhile to post to a mailing list, etc.
3). Again, I have my own project and I have received bug fixes, new code, and even artwork from people by email. Generally the people who send this stuff don't want to make a permanent commitment to the project, and why should they? I have sent emails to developers on other projects and gotten responses back and this has led to changes to and improvements in my own project, but none of this discussion gets noted on Sourceforge.
I think there is a lot of worthwhile stuff done by lone developers, and I hope my own project falls into that category. However, the software everyone uses and knows about is probably not developed by lone developers.
Re:Previously predicted...
on
Byte Wars
·
· Score: 5, Interesting
I remember trying to use the Yourdon method on a project. We had to draw a hell of a lot of process diagrams that decomposed into lower and more detailed levels. Nobody did all the charts right, and when we were done with the charts nobody looked at them again. It was just a big time sink. We also spent a fortune on CASE tools to draw the damned things. My job was to convince everyone to use these tools and draw the diagrams.
Basically I'd say the guy has been wrong on everything he's ever written. If I was concerned about what terrorists might do to our IT infrastructure before I suppose I should be less concerned now.
The second column has more credibility with management types and the first wins points from the technical side. Its definitely a win for everyone involved. I think Eclipse definitely will do well with this strategy. WSAD adds needed servlet testing, EJB, etc. functionality that makes it a good product for the office, and students and open source/free software types get a first class IDE to use and improve on for free. I think that Sun has a reasonable chance to succeed with this strategy.
If you like a newsreader that doesn't support yEnc decoding yet and you run an OS that has Java JDK 1.2 or better you can use your newsreader to download the raw articles then decode the articles afterwards using the standalone decoder NiceSTEP yEnc Decoder. It has a nice drag and drop GUI and is available with source code or precompiled from http://nicestep.sourceforge.net
I've been using this in combination with Pan (before it had built in support for yEnc) for a few weeks now and it works great. And contrary to what others may say, downloads ARE significantly shorter with yEnc encoding.
What attracted me to Linux in the first place was that it WOULD run on old hardware. My first Linux install was on a 486/66 with 350 meg hard drive (later added another 500 with a drive bought on eBay) and 16 meg RAM. My main machine at home is an IBM Aptiva from 1995, originally 32 meg of RAM but upgraded to 64 (all it will hold) and under 200 megahertz. I run Windowmaker with a nice theme and use text mode mc for file management, plus I run gnome apps like pan and sylpheed. I also do some development in Java.
At work I have a 486/33 originally purchased as a Novell server and left under somebody's desk. It runs Red Hat 5.2, mostly in text mode but I do occasionally fire up X.
The ability to run on low end computers is one of the best selling points of Linux.
Also, I use WindowMaker because I think it is the best desktop ever. It isn't lack of resources keeping me from using GNOME or KDE. On Windows at home I use LiteStep, and I'd use it at work too if I could.
There was supposed to be a "Mai The Psychic Girl" movie a few years ago. This was based on an excellent Japanese comic book, that was sort of inspired by the movie "Firestarter." The comic book was great. They could have used it as a storyboard for a really great movie. Unfortunately somebody decided that the story needed to be relocated to the U.S.A. and that it should be a MUSICAL. No, I'm not kidding. As far as I know the movie never got made, and there's a good chance they'll never make a Dragonball movie either.
For the kind of stuff that people want DVD writers for (other than backing up data) I have to wonder if VCDs would work better. For instance, I have a big collection of Betamax tapes I bought cheap when all the video stores were dumping them. I find that I can get reasonably good results converting them to VCDs, which can play on most DVD players, especially the cheaper ones. I could do the same thing with my wedding video, videos of my friend's kids, and the student films I made in college and later transferred to videotape (at great expense). It's like they used to say in the old mainframe days: garbage in, garbage out. If your source material isn't that great then doing an analog transfer to DVD won't help. VCD is good enough. On the other hand, if the source material is clean and you filter it digitally a VCD can look very nice.
Unless you have really expensive equipment I don't see how you could make a decent DVD at home. OTOH, you can make a VCD with modest equipment and cheap or even free software.
The window manager for GNUstep is WindowMaker, and it is lightweight to run on anything. I have run it on a 486/33 with 16 meg of RAM. I also use WindowMaker every day on a Pentium I. with 64 (formerly 32) meg of RAM. It is definitely lighter weight than KDE or GNOME, both of which should still be useable on your Pentium.
I have some nieces who love American Girl dolls and paraphernalia. Once or twice a year I go on their website to buy stuff for them. Their mother sends me descriptions and even catalog numbers and prices of the stuff they want and even with that information it takes me upwards of an hour to place a simple order. In the meantime I learn more about those dolls than I really wanted to know. Of course I should just use the catalog, and eventually I'll be smart enough to do that. But shouldn't the web be easier and faster than using a catalog? There are sites where that is true, and Amazon is among the best.
1). Price. Granted, they may not be THAT much more expensive, but I'm the kind of guy that keeps a computer a long time and doesn't buy a new computer until I find one I like that is *already* discontinued. Buying the latest thing every two years is not an option for me.
2). I've invested too much in MS-DOS and Windows software that I still want to run now and then. "Sam and Max Hit The Road" being a good example. I bought it for my 386, and three computers later I still haven't finished with it.
3). No compelling reason to switch. I can manage pictures and create MP3's and CD's on my old Pentium. I can create a VCD on my reconditioned Pentium III. Why would any reasonable person want to create his own DVDs? Even creating a VCD is a very time consuming process (capture the video in real time, then compress it for eight-twelve hours). Who has the equipment to create a professional quality DVD anyway? If it isn't going to be professional quality, why not just make a VCD instead?
4). Mac users seem to be an elitist bunch. They look down on people that actually write software and enjoy tinkering with computers. Like Jeff Goldblum in the ads: "I don't want to be that person!" and "You can have a computer and not be a computer person." Who needs that?
I *have* switched to Linux for most things because there was a lot of software I wanted to try out and because I didn't have to give up my old stuff. All my machines dual boot. I would give OS/X a chance if it ran on Intel hardware. Hell, I was ready to try BeOS and NeXTSTEP and would have if they were still available.
The RAR format is regularly used in multimedia newsgroups to archive files into many small pieces that can later be recombined. Using RAR to compress something costs something, but there is a free UNRAR utility (with source code) that can be used to do the decompression that works on any platform. Plus the download would be easier for guys like me with dial-up!
Our company uses Visual Age for Java and while I have found the repository to be very useable and much easier to learn than most source control tools it has always irked me a bit that a shared repository was only available in the top of the line Enterprise product. For most of us that was the ONLY real difference (the one thing we actually used) that separated the inexpensive Professional edition from the many times more expensive Enterprise version.
I'm not looking forward to the migration but when we're all on CVS (or whatever) I think we'll be a lot better off, and will probably have an opportunity to save some money on development tools.
My one current gripe is that IBM bundles ClearCaseLT and a plugin to use it with WebSphere Application Developer (based on Eclipse) but does not supply any documentation on how to use the two together.
CVS integration on the other hand is very well done.
I've been using this one awhile. I used to use Visual Age for Java for Linux but IBM discontinued that after version 3.0, which left me with no way to write JDK 1.2 apps. Eclipse makes me want to forgive them for that. It has some of the best features of VAJ and has some great new features.
For instance, VAJ was always difficult to use with CVS, but Eclipse support for CVS could not be better. Really outstanding.
Eclipse does need more memory than VAJ did (I run both on an IBM Aptiva with 64 mb of memory and the difference is notable) but given enough memory it runs fine. For those reporting stability problems remember that Eclipse runs under Java and all Java IDEs under Linux are not equal. IBM's tend to work best but they aren't flawless. The IBM JDK does work better than Sun/Blackdown for running Eclipse so try that and see if you don't like Eclipse better.
Eclipse so far lacks a GUI design tool but there seem to be several people at IBM and elsewhere working on one, so we should have several to choose from in time.
I like very much that Eclipse is the base for IBM's commercial offering WebSphere Application Developer (the successor to Visual Age for Java). This means that most plugins written for Windows should also be available on the Linux side and that IBM should be able to offer a Linux version of WSAD without much extra effort (something that probably wasn't true of VAJ.)
I find Eclipse very useable on its own and it has been a great help to my own free software project.
I would take this one step farther. Why not write the GUI front ends in Java? That too is really easy and guarantees that the whole thing is as cross platform as it can be.
I've done this with mkisofs and cdrecord and it worked really well. vcdimager and cdrdao (used to create VCDs to play on your DVD player) would seem to be excellent candidates for this too, because both are command line programs that will run on just about any platform.
Your post reminded me of a notion I had. The famous Apple commercial had the line "Why 1984 won't be like 1984". You could do a similar spot against Windows ME saying "Windows ME: why 2001 won't be like 2001."
Personally, I like the way GNUstep looks. The scrollbars, in particular, are better than any others I've used. I liked it so much that I wrote a library of Java components so I could create Java apps with that look and feel.
Many many people enjoy using Windowmaker and feel it is the most polished and attractive window manager out there. While I use a lot of GNOME apps (with the gtkStep theme) I have no use for the GNOME desktop. Windowmaker does what I want, uses modest resources, and looks good. Give it a chance, it may well grow on you.
Everyone has heard the stories about two kids in a garage that make a new kind of software and become millionaires. Of course it isn't that easy. The software business is tough to break into. Consider the following:
1). Microsoft has such a strong position in the market that people have an irrational fear of using anything else. If I wrote a word processor that was demonstrably better than MS Word I'd have trouble GIVING it away, let alone selling it.
2). Software is easy to copy. Even if I write really good software and sell it cheap, not everyone will buy a legal copy.
3). Software can easily be written by one person working on his kitchen table. Selling software requires a LOT more resources, including employees who will insist on being paid whether the company makes any money or not.
4). If I sell software I am morally obligated to stand behind it to some extent, to provide support. Giving away software with source means that anyone who gets my software can solve his own problems. I can refuse to be liable for what my software does with a clean conscience.
5). If I give my software away with source code I don't get any money for the software. However, I don't lose money either. (http://sourceforge.net provides a method of distributing my software to anyone that wants it without cost to me.) Maybe I can gain a reputation for writing good code or designing good systems that may help me find work.
There is plenty of work writing custom software and that is a surer way of making money than running a software business.
6). Since everyone who uses my program gets my source code, some of them may be motivated to find and fix bugs, add features to my project, etc., all at no cost to me. This has in fact already happened on my own project.
7). I don't have to give anyone guilt trips about paying for my work and not sharing it with their friends. I WANT them to share with their friends! The more the better!
8). Finally, I can benefit by the code and programs of others. If someone else's program has a feature that would also be useful in my own, I can use his code as is or try to improve on it.
Contrast this with Microsoft's business model, which is to convince people to buy basically the same products over and over again every few years, when the products they bought the last time still work and will NEVER wear out? When the new products require a new computer to run? It is amazing they are still in business!
Slashdot Mirror
I agree with a couple of points here:
1). I don't love the AWT, but it has its place. If nothing else we need the classes java.awt.Component and java.awt.Container so we can create our own lightweight components without inheriting unnecessary baggage from Swing.
2). Changing names for the sake of changing names is stupid, stupid, stupid. Just this morning I had to go back to using the deprecated methods getValue(), putValue(), etc. for a Session object because most programmers in my company are not using the very latest JDK. In a large company keeping everyone's IDEs up to date is a huge task that nobody has time for. I've seen several examples where deprecated methods were replaced with something else that was different only in name, and the name change was totally arbitrary (replacing "Url" with "URL" in a method name). I have to write code that is used by developers throughout the company and this kind of thing really bugs me.
I've been using eclipse for awhile and I like how SWT works in that application. You'd never know that app was written in Java if nobody told you. However, it does need more memory to run than Visual Age for Linux did. I'm well aware of this because I have developed on both on a Pentium with 64 Meg of memory (an Aptiva, so you can't add more memory). So the UI seems to be only part of the problem with Java.
I have used Swing apps like JEdit on machines with a lot of memory and have no complaints. They look as good as a Windows application and performance while they are running is decent. Startup time is slow on any Java program though.
I have also made my own lightweight components which are much less elaborate than Swing but seem to me to work well enough on a Pentium. Making your own Java ui components is easy and even fun, and I've wondered why more people don't try it.
I tend to agree with this article. The first time I ran X on Red Hat 5.2 I saw this very lame attempt to make Linux look like Windows 95. I was very disappointed. Soon I discovered AfterStep, then WindowMaker, and finally GNUstep, which gave me something really different from Windows but which was obviously well thought out. Scrollbars are on the LEFT of the thing being scrolled, instead of the right (where they get pushed off the edge of the screen). The scrollbar arrows are next to each other, instead of at opposite ends of the scrollbar. Menus are in their own windows, and can be moved around and torn off. Dockapps and the dock are neat too.
The new Apple OS is almost as nice, but they put the scrollbars on the wrong side.
I was frustrated that Java Swing's pluggable look and feels make everything look like Windows 95, so I started my own project to make Java applications with the look and feel of GNUstep possible.
If you are looking for something really different but still well thought out in user interfaces you should really check out WindowMaker and GNUstep.
What's funny about this is that awhile back my favorite niece was working on my computer (running WindowMaker) and asked me why I bothered with Linux since it seemed to her that it was just like Windows.
Of course that's the point of the article. Linux should be different from Windows and superior to it so that Windows users wonder what they might be missing by not running it. A more stable version of what they already have is not all that compelling.
I read a story awhile back where somebody claimed that a malicious programmer shut off the landing lights at an airport using the Internet. Nobody questioned that this had actually happened, or why the hell lights at an airport could be shut off over the Internet at all.
I can see a possible scenario for the Pacemaker though. You need a Pacemaker and two kinds are available. One is not connected to the Internet but is available only to the very wealthy. The other is connected at all times and can be shut down remotely if you don't keep up with the subscription payments for the software.
The idea of charging for binaries doesn't bug me that much. I've always paid for my SuSE distros and consider the money well spent. I also compile and install a lot of stuff by hand (mostly WindowMaker and GNUstep, etc.) The stuff I do by hand is the stuff I really care about, and it always takes longer than I think it will. OTOH, the stuff that comes with the distro generally just works. I can afford to take it for granted. That is worth more than the modest cost of the distro.
Maybe if the distro companies started using this approach (free source, charge for binaries) then we could get business people to understand that the "free" in free software is all about freedom, and that freedom is what drives the real cost savings.
I've also worked on a free software project and would be thrilled to death if anyone wanted to include it in a distro so it could be installed via RPM. I would not feel exploited in the slightest way. Heck, I'm giving the stuff away anyway, why wouldn't I be happy about being chosen for inclusion in a distro?
One thing that never seems to come up in these debates is how truly "closed" closed source is. If you take security through obscurity seriously, you'd have to run Microsoft like it was the Manhatten project. Consultants would have to be carefully screened. Employees would not be allowed access to the Internet. They could send email only within the company (otherwise they could email themselves any source code they'd like). They would have to be searched when entering the building and leaving, so they couldn't smuggle code out on diskettes or printed out on paper. Maybe they would have to work on diskless workstations.
It's pretty obvious that no company in the real world is like this, so we can assume that the Black Hats can see Windows source code. And of course there is the famous "shared source" program that makes Windows code available to colleges, etc.
In the open source and free software world we don't have the false sense of security that those in the Windows world may have. As far as I can tell this should make our stuff more secure in practice.
If there is anything that makes me unhappy with the popularity of IE it is that IE does a very poor job printing web pages. For the home user this is rarely a problem, but for corporate users this is a serious problem. When developing a corporate intranet I want to create reports on the web that a user can view or print out easily. IE fails miserably. Two examples:
1). When printing a table it will split a table row in the middle, printing the top half of the row on one page and the bottom half on the next page. Netscape never splits a row like this.
2). IE does the same thing with images. If you have an image near the bottom of a page it will get split across two pages, something Netscape doesn't do.
While Netscape is better than IE, it isn't enough better to compell people to switch.
Here is my printing wish list. If Microsoft was all that innovative they could have fixed this stuff long ago:
1). Give me a way to force a page feed when printing. This would have no effect when viewing pages online.
2). When printing a table, if the table runs over a page give me an option to repeat the table headings at the top of subsequent pages.
3). How about an option to designate sections of pages to be non-printable? In other words, you could print any page marked up this way and it would come out printer friendly.
4). How about an option to tell the printer to print the page landscape or portrait, so users could simply hit the Print button and get something that fit on the page the way the page designer intended?
The W3C created a much longer list of stuff to address printing web pages and as far as I can tell it has been ignored. I would be happy with the short list above. This list *might* be enough to give corporate users a reason to switch browsers.
I had a friend in college who used to do something like this with vinyl records. He'd buy a record from Kmart or someplace and it would be a record that looked interesting but that he hadn't heard. He'd take it home and listen, and if he liked the music he'd keep it. If he didn't he'd scratch the record with car keys and return it for a refund because it was damaged.
It would make much more sense to return a CD for a refund than it would to sell it to a used CD store, if you were really interested in ripping off somebody.
I don't think the survey is totally worthless but it ignores a few things:
1). Maturity is subjective. For instance, WindowMaker hasn't had a 1.0 release yet but I would consider it mature enough to use every day. It would make more sense to look at the top 500 or so downloaded projects to get a sense of what can be considered mature.
2). Small projects don't generate much discussion because in general it is too much of a hassle. I have to have an account on SourceForge to post to a discussion or subscribe to a mailing list. However, I can simply email the author on a small project and get a better chance of being heard. Only a large project with many developers would make it worthwhile to post to a mailing list, etc.
3). Again, I have my own project and I have received bug fixes, new code, and even artwork from people by email. Generally the people who send this stuff don't want to make a permanent commitment to the project, and why should they? I have sent emails to developers on other projects and gotten responses back and this has led to changes to and improvements in my own project, but none of this discussion gets noted on Sourceforge.
I think there is a lot of worthwhile stuff done by lone developers, and I hope my own project falls into that category. However, the software everyone uses and knows about is probably not developed by lone developers.
I remember trying to use the Yourdon method on a project. We had to draw a hell of a lot of process diagrams that decomposed into lower and more detailed levels. Nobody did all the charts right, and when we were done with the charts nobody looked at them again. It was just a big time sink. We also spent a fortune on CASE tools to draw the damned things. My job was to convince everyone to use these tools and draw the diagrams.
Basically I'd say the guy has been wrong on everything he's ever written. If I was concerned about what terrorists might do to our IT infrastructure before I suppose I should be less concerned now.
I think one trend we're already seeing and which we'll see more of is the hybrid open source/commercial product. Some examples:
1). Mozilla/Netscape 6
2). NetBeans/Forte
3). Eclipse/WebSphere Application Developer
4). Open Office/Star Office
The second column has more credibility with management types and the first wins points from the technical side. Its definitely a win for everyone involved. I think Eclipse definitely will do well with this strategy. WSAD adds needed servlet testing, EJB, etc. functionality that makes it a good product for the office, and students and open source/free software types get a first class IDE to use and improve on for free. I think that Sun has a reasonable chance to succeed with this strategy.
I've been using this in combination with Pan (before it had built in support for yEnc) for a few weeks now and it works great. And contrary to what others may say, downloads ARE significantly shorter with yEnc encoding.
What attracted me to Linux in the first place was that it WOULD run on old hardware. My first Linux install was on a 486/66 with 350 meg hard drive (later added another 500 with a drive bought on eBay) and 16 meg RAM. My main machine at home is an IBM Aptiva from 1995, originally 32 meg of RAM but upgraded to 64 (all it will hold) and under 200 megahertz. I run Windowmaker with a nice theme and use text mode mc for file management, plus I run gnome apps like pan and sylpheed. I also do some development in Java.
At work I have a 486/33 originally purchased as a Novell server and left under somebody's desk. It runs Red Hat 5.2, mostly in text mode but I do occasionally fire up X.
The ability to run on low end computers is one of the best selling points of Linux.
Also, I use WindowMaker because I think it is the best desktop ever. It isn't lack of resources keeping me from using GNOME or KDE. On Windows at home I use LiteStep, and I'd use it at work too if I could.
There was supposed to be a "Mai The Psychic Girl" movie a few years ago. This was based on an excellent Japanese comic book, that was sort of inspired by the movie "Firestarter." The comic book was great. They could have used it as a storyboard for a really great movie. Unfortunately somebody decided that the story needed to be relocated to the U.S.A. and that it should be a MUSICAL. No, I'm not kidding. As far as I know the movie never got made, and there's a good chance they'll never make a Dragonball movie either.
Unless you have really expensive equipment I don't see how you could make a decent DVD at home. OTOH, you can make a VCD with modest equipment and cheap or even free software.
The window manager for GNUstep is WindowMaker, and it is lightweight to run on anything. I have run it on a 486/33 with 16 meg of RAM. I also use WindowMaker every day on a Pentium I. with 64 (formerly 32) meg of RAM. It is definitely lighter weight than KDE or GNOME, both of which should still be useable on your Pentium.
I have some nieces who love American Girl dolls and paraphernalia. Once or twice a year I go on their website to buy stuff for them. Their mother sends me descriptions and even catalog numbers and prices of the stuff they want and even with that information it takes me upwards of an hour to place a simple order. In the meantime I learn more about those dolls than I really wanted to know. Of course I should just use the catalog, and eventually I'll be smart enough to do that. But shouldn't the web be easier and faster than using a catalog? There are sites where that is true, and Amazon is among the best.
2). I've invested too much in MS-DOS and Windows software that I still want to run now and then. "Sam and Max Hit The Road" being a good example. I bought it for my 386, and three computers later I still haven't finished with it.
3). No compelling reason to switch. I can manage pictures and create MP3's and CD's on my old Pentium. I can create a VCD on my reconditioned Pentium III. Why would any reasonable person want to create his own DVDs? Even creating a VCD is a very time consuming process (capture the video in real time, then compress it for eight-twelve hours). Who has the equipment to create a professional quality DVD anyway? If it isn't going to be professional quality, why not just make a VCD instead?
4). Mac users seem to be an elitist bunch. They look down on people that actually write software and enjoy tinkering with computers. Like Jeff Goldblum in the ads: "I don't want to be that person!" and "You can have a computer and not be a computer person." Who needs that?
I *have* switched to Linux for most things because there was a lot of software I wanted to try out and because I didn't have to give up my old stuff. All my machines dual boot. I would give OS/X a chance if it ran on Intel hardware. Hell, I was ready to try BeOS and NeXTSTEP and would have if they were still available.
The RAR format is regularly used in multimedia newsgroups to archive files into many small pieces that can later be recombined. Using RAR to compress something costs something, but there is a free UNRAR utility (with source code) that can be used to do the decompression that works on any platform. Plus the download would be easier for guys like me with dial-up!
I'm not looking forward to the migration but when we're all on CVS (or whatever) I think we'll be a lot better off, and will probably have an opportunity to save some money on development tools.
My one current gripe is that IBM bundles ClearCaseLT and a plugin to use it with WebSphere Application Developer (based on Eclipse) but does not supply any documentation on how to use the two together.
CVS integration on the other hand is very well done.
For instance, VAJ was always difficult to use with CVS, but Eclipse support for CVS could not be better. Really outstanding.
Eclipse does need more memory than VAJ did (I run both on an IBM Aptiva with 64 mb of memory and the difference is notable) but given enough memory it runs fine. For those reporting stability problems remember that Eclipse runs under Java and all Java IDEs under Linux are not equal. IBM's tend to work best but they aren't flawless. The IBM JDK does work better than Sun/Blackdown for running Eclipse so try that and see if you don't like Eclipse better.
Eclipse so far lacks a GUI design tool but there seem to be several people at IBM and elsewhere working on one, so we should have several to choose from in time.
I like very much that Eclipse is the base for IBM's commercial offering WebSphere Application Developer (the successor to Visual Age for Java). This means that most plugins written for Windows should also be available on the Linux side and that IBM should be able to offer a Linux version of WSAD without much extra effort (something that probably wasn't true of VAJ.)
I find Eclipse very useable on its own and it has been a great help to my own free software project.
I would take this one step farther. Why not write the GUI front ends in Java? That too is really easy and guarantees that the whole thing is as cross platform as it can be.
I've done this with mkisofs and cdrecord and it worked really well. vcdimager and cdrdao (used to create VCDs to play on your DVD player) would seem to be excellent candidates for this too, because both are command line programs that will run on just about any platform.
Your post reminded me of a notion I had. The famous Apple commercial had the line "Why 1984 won't be like 1984". You could do a similar spot against Windows ME saying "Windows ME: why 2001 won't be like 2001."
Many many people enjoy using Windowmaker and feel it is the most polished and attractive window manager out there. While I use a lot of GNOME apps (with the gtkStep theme) I have no use for the GNOME desktop. Windowmaker does what I want, uses modest resources, and looks good. Give it a chance, it may well grow on you.
1). Microsoft has such a strong position in the market that people have an irrational fear of using anything else. If I wrote a word processor that was demonstrably better than MS Word I'd have trouble GIVING it away, let alone selling it.
2). Software is easy to copy. Even if I write really good software and sell it cheap, not everyone will buy a legal copy.
3). Software can easily be written by one person working on his kitchen table. Selling software requires a LOT more resources, including employees who will insist on being paid whether the company makes any money or not.
4). If I sell software I am morally obligated to stand behind it to some extent, to provide support. Giving away software with source means that anyone who gets my software can solve his own problems. I can refuse to be liable for what my software does with a clean conscience.
5). If I give my software away with source code I don't get any money for the software. However, I don't lose money either. (http://sourceforge.net provides a method of distributing my software to anyone that wants it without cost to me.) Maybe I can gain a reputation for writing good code or designing good systems that may help me find work. There is plenty of work writing custom software and that is a surer way of making money than running a software business.
6). Since everyone who uses my program gets my source code, some of them may be motivated to find and fix bugs, add features to my project, etc., all at no cost to me. This has in fact already happened on my own project.
7). I don't have to give anyone guilt trips about paying for my work and not sharing it with their friends. I WANT them to share with their friends! The more the better!
8). Finally, I can benefit by the code and programs of others. If someone else's program has a feature that would also be useful in my own, I can use his code as is or try to improve on it.
Contrast this with Microsoft's business model, which is to convince people to buy basically the same products over and over again every few years, when the products they bought the last time still work and will NEVER wear out? When the new products require a new computer to run? It is amazing they are still in business!