You're likely a believer in the "free market" and thus likely a fan of economic theory, so I'll put this in terms that should be familiar:
The RIAA currently has a monopoly on a good known as "popular musicians." The vast majority of consumers willingly pay this monopoly to listen to the music that they love. Consumers (like you and me) are charged with variable costs that are extremely inflated due to the lack of competition. The barriers to entry (leverage to put CDs on shelves) and the economies of scale (manufacturing enough to sell to chain stores) are so difficult to overcome in this instance, other labels have a hard time. Note that I said "popular musicians" -- not "good musicians."
That's what gets me. Taking Napster down for "contributory infringement" is like taking down Smith & Wesson for being an accomplice to a murder. A tool is a tool.
Well, this is an interesting issue. IANACL, but as far as I know, copyright refers to duplicating someone else's work. It is questionable whether downloading a copyrighted MP3 is duplication in that:
1) Copying bits is different in that you MUST copy bits to look at them. A data stream and a file in memory or on a hard drive are necessary for viewing, not just duplicating. 2) The recording industry sells copied disks. That is their business, selling copies of their original work. Copying the song verbatim is copyright, but is copying a CD? If you copy a CD, you've effectively got an exact replica of something that they would sell you: a CD. You have effectively "taken" an item from them without giving them money. I think that fits the definition of stealing.
That said, stealing from the mafia is not immoral in my book. The idealist in me says "Well, don't buy from them and hope they get brought up on charges of breaking anti-trust laws and price gouging." But, the realist in me realizes that our government and courts are already in the pocket of the recording industry. Just look at how most legislation and court decisions have been going.
Detachment seems to be a common theme. Rule your company at such a high level that you don't see the details, right?
It really is weird, the trend I'm noticing... bosses get managers to handle certain projects, not caring how they're done as long as they're under budget. I know this isn't the same as replacing "being responsible" with "consequence based thinking", but the trend does seem to be all across the board.
Dear Lord, everyone's talking about replacing the Linux kernel. Why is this even being considered?
Even IF SCO's, uh, "claim" were true (what part of what *NIX is it today?), the related developers and companies can be sued, the code removed and Linux continues. Linux is not controlled or owned by one entity but hundreds. Unless they want a, uh, reverse-class-action lawsuit (do these exist) or make a few hundred cases, Linux is fine.
Quit fretting, the storm will be over soon, immature folks with money are just having a little temper tantrum.
Sometimes I sincerely wonder about RMS. He's got some great ideas that I buy into, he's made some great software and has really pushed software into an area not thought possible...
You can now work your entire IT career using tools that you did not pay for.
What other industry can say that?
That said, you'd think by now he'd know how to make a good argument. Don't preach, don't write your argument sitting on the assumption that you're right and everyone else is wrong. And for God's sakes, don't push semantics.
You wonder why Java fanatics have problems winning people over?
Calm, rational, aggravatingly logical disussion will win people over much faster. I wish RMS, Linux zealots, Java nuts, etc. would pick up on this. Why? Because I absolutely love everything they stand for and preach about... it's not an understatement to say that it's changed my life. I get giddy when I think about it all. But you don't win an argument by pulling someone up the ladder. You win an argument by jumping down to the bottom and climbing with them.
Long story short, I see your point completely -- in some mainly few-and-far-between instances, SWT will need to be upgraded and that requires intervention on your part. You also cannot claim to run your application on more than a few systems.
That said, I still think you're making a mountain out of a mole-hill. For the vast majority of applications that will need a GUI, you're not going to need to implement a version for anything besides Windows/Linux/Mac, and windowing toolkits won't change often enough to pose a frequent problem.
If you want to have the perfect implementation (which there is something to be said for), SWT is not the perfect solution. But considering its native look-and-feel and the speed, SWT is generally the most practical choice.
--- "Also: why IBM did it the way they did instead of pushing SWT via JCP? When MS does something like that (J++) people are unhappy and speak about embracing, extending... but when IBM does the same everyone is prising it happily?"
Well, SWT is open source, and IBM isn't trying to bastardize Java into making it IBM dependent. If they were, it wouldn't be open source and portable/extendable.
--- "What's the future of swing? Will SWT become a part of J2SE and swing will be deprecated? What's the future of SWT? Will it ever become a part of J2SE thus enabling Pure Java SWT apps? Does anyone like this uncertainty?"
I sincerely hope that somehow, whether through two implementations or one, we get the abilities of 1) platform independance and 2) the option of using native look-and-feel (MS open dialog). I don't care which toolset I use -- I just want a realistic implementation.
Re:Like it or not, managers default to commercial
on
What is Open Source?
·
· Score: 1
1) I understand that many companies are on the ball; mine generally is. (Hopefully I should know, I'm one of the guys whose job it is to fix these problems!). However, the corporate attitude is to make money. I don't think anyone can argue with that. Many tech companies, software companies especially, are so concerned with the bottom line these days that fixing problems and keeping the customer smiling isn't a priority, since they can still sell their product. This is by no means applicable to all companies, but it is a large trend.
2) Although you are correct in pointing out that developers in companies still have "their babies" and do use them, I think you're missing the overall point (which I didn't point out clearly). Companies are, again, after money. There are deadlines, there are insane managers who push for unrealistic release dates and then move on to new software instead of fixing what just went out the door. The vast majority of software companies do it for the money, whereas the vast majority of OSS developers do it for the fun and the personal benefit alone.
Just because you do it for money doesn't mean you won't invest yourself in your work; when I was developing at my company just over a year ago, my projects were my passion. I put my all into them. But a corporate environment isn't a place where you often can take the time to do things right.
I don't know about that. My first Linux installation was plain old Red Hat 6.2 a few years ago (I think it had just come out). I used GNOME and didn't know of Ximian at the time (if it existed at the time).
Man, oh MAN was that a horrible looking interface. It was confusing to read a HOW-TO page walking me through some complex operation on the command line.
But, in the same breath, I spent every second of 30-45 minutes playing with the screensavers. Another 15 spent on themes. And I could use it and the system didn't crash -- I just didn't have the problems I had had with Windows.
Granted I was a developer then, but Linux wasn't near as user friendly as it is now. Most of the folks I show Linux to are, well, found gawking at everything you can do with the desktop.
Users these days are used to crufty interfaces, crashes and bad patches. I don't think Linux will have a hard time being adopted. The folks in my life haven't found it too troubling in itself.
As I've pointed out in previous posts in another Java thread, with the momentum of Red Hat and the entire OSS movement behind it, probably.
Much like Microsoft and the Internet revolution, Sun needs to wake up and smell the coffee (bad pun intended). Sun will die alone. Proprietary lock-in is a science at Microsoft and too big of a threat, in this developer's eyes. It won't last forever, but it very well may outlast Sun.
I heard a suggestion that I'd be much happier with.
Include a clause that says that any user of the software who issues a lawsuit against OSS automatically loses the license to use all software with the same clause in their license.
To illustrate this, if this clause were in the LGPL, any company suing an OSS project would be unable to use any LGPL'ed software.
Archaic? Draconian? Too risky you say? Read your EULA lately?
Re:Like it or not, managers default to commercial
on
What is Open Source?
·
· Score: 1
"You're probably not familiar with how this normally works. If you license a product from IBM or MS it's not like all your communications go from the very top down to the bottom and they simply dump a product on your lap and leave you alone. You are generally (always?) given a contact with the team lead(s) and have a way to contact the people who coded the software itself. To say that licensing from a big corporation means your concerns won't be heard unless you are also a big corporation is pretty much dead wrong."
If this is the case, then I stand corrected. Somehow, though, from what I've heard from folks I work with, the things I've seen personally and the things I read about online (in dar intarnet huh huh), companies like MS generally aren't inclinded to fix problems.
--- "Again, open source developers don't have a business relationship with you, so they don't have any real incentive to be attentive to your needs. They don't have to fix anything."
As someone else suggested, throw them some money to fix a bug. Many OSS developers get cash this way. It will almost certainly be cheaper to throw him some cash than to fix it yourself, and will likely be about as expensive as an "incident report" or yearly support contract.;)
--- "There are many developers who may just flat out refuse to alter the way their program works, even if it's minor thing, because "it's their baby"."
How many medium to large corporations have you known of to alter their software at a single customer's request? This practice is generally frowned upon in all but the smallest companies, unless, of course, the company is in the business of selling solutions, which most medium-larged sized companies aren't.
I don't know that I agree with that. In my humble opinion, Gaim is one of the best pieces of software out there. It's flexible as hell, fast, configurable in many ways, and in just the right areas, and has some really useful features (buddy pouncing, multiple accounts on different services, spellchecking, aliasing, etc.). All that and I still feel like it's one of the most user friendly applications for Linux. And Windows.
They list 7 people as "developers" and another 5 as "contributors". How much or how little each of them develops/contributes I don't know, but I get the feeling that this isn't a "major project".
Maybe our opinions differ, maybe this is just an isolated example, but I've worked with enough tiny projects that have helped me quite a bit to feel comfortable saying that there is a wealth of projects under the top few that rock the house.
As for the 'dead projects', we can go into the ratio of dead/alive closed source companies compared to that of OSS projects. We can even bring in CS shareware if you like.:)
Please feel free to correct me if I'm missing something.:)
The SWT libraries for all systems can be sent with an application and invoked dynamically. One.jar file is included in the classpath and one directory containing libraries is invoked dynamically in the call to the VM. This is about as trivial as you can get.
--- "go crazy with support, cause now it's yours - app developer's - responsibility to maintain your packages - upgrade for any future changes in GUI subsystems of any of your target OSes (YOU will have to repackege your software with the new versions of SWT, because Sun is not going to distrbute them)."
I will admit that changes in the GUI widget set can force you to send updates to the *SWT libraries*, but odd are that this will happen far less often than updates to your own product. This "you'll have to repackage your whole application" is pretty ridiculous. To update the version repository of your software, just copy in a new.jar and handful of library files. For existing customers? The same thing.
You argument has some merit, but I really don't see this as anything but a trivial obstacle. Maybe I misread or missed something altogether in your post?
Cheers
Re:Like it or not, managers default to commercial
on
What is Open Source?
·
· Score: 2, Insightful
"you could spend a couple days trying to track down and fix a problem that is preventing you from finishing your real work or you could simply send a problem description off to the guys who originally wrote the code (and who you're already paying for), and they'll probably be able to fix the problem far quicker than you ever could."
This point does have merit, but I suppose you're talking about problems with small to mid sized companies. Big ones (Like Microsoft, IBM, Novell, etc.) aren't going to fix your problems any time soon; they have so many customers, unless you're a big-shot (who can spend the $$$ to fix it yourself), they won't care. Have you had the pleasure of paying money for an incident to report a bug to MS?:)
--- "What you do know is that you don't have the ability to really push open source developers to fix problems for you, and that's a major drawback."
No, but truth be told, most OSS developers aren't working on their projects because they're told to or are being paid. They do it because their projects are their babies.
Typically, problems reported are fixed in relatively short order because: 1) you've likely, in the process of debugging your own software or tools, have isolated the problem to a specific area and pattern and 2) OSS folks almost always use the software they write themselves. That's usually why they write it in the first place.
Your argument has some weight, but I don't think it pans out in the large scheme of things.
[Apple Switch] My name is Dalcius, and I'm a Software Analyst (debugger) for a mid-sized company that deals with semi-large ones. [/Apple Switch]
I think I can get a reply out by looking at your last paragraph:
"I should probably clarify something I said before then. I'm concerned you'll feel that I'm arguing justification and not property. My argument is actually a blurring of the two. The labelling of 'theft' by the music industry is not a reflection of actual monetary losses, but a defense aimed at preserving their lucrative business model. What's happening here is the RIAA is screwing it's customers, and they're defending themselves by violating the terms of service. Nobody's been given the chance to pay for what they've downloaded. They can enjoy the song for years, at some point along the way they should pay for it. It's hard to call it theft when the door to even up is still open."
If I understand you correctly, your basis for this whole argument is this: 1) The major labels effectively act as one company 2) That "company" holds a monopoly on most music 3) That "company" charges overwhelmingly inflated prices for its goods. 4) There is enough demand for these expensive goods because they can't be bought any cheaper (see #2). 5) Thus, the consumer is being strongarmed into paying too much money for a product. 6) In fair retaliation, customers violate the ownership of the major labels in order to sample the music, sometimes not buying a CD that they would have if those tracks were short demos, expired after a week, etc.
A staunch capitalist would reply that consumers aren't forced into buying the music, thus they are not being strongarmed. In theory, I would agree, however in practice, with the current barriers to entry for startup labels and with the fact that nearly *everyone* needs music (as much as they need things to read and TV to watch -- it's a part of our daily lives), I do agree that things should change.
However, that said: 1) I do not have a moral problem sampling their music or even downloading and keeping entire CDs without paying for them. 2) However, that is, in my mind, still stealing: you benefit from their work without rewarding them, one of the conditions of their labor. 3) Referring to #1, the "right" way to do things would be to bring up the labels on charges of anti-trust violations and of price gouging.
This, however, assumes that the public isn't apathetic and that politicians aren't bought and paid for.
Summary: You got me to take things I'd already known and really put together the big picture: these folks are as bad as Microsoft (the similarities are stunning, no?).
That was fun; thanks for the new point of view and for listening to mine. =)
Based on their original (albeit vague) allegations, how can SCO claim that they didn't know that they distributed their IP under the GPL when IBM can claim the same ignorance?
"If the Linux kernel is a derivate work of SCO's (or more likely, certain PARTS of the kernel are derivative works), then SCO can simply claim that ALL copies of the Linux kernel are infringing copies, and ALL public distributions of the code are copyright violations."
The Linux Kernel is a work owned by many people; the copyrights belong to no one person. To declare that, "Well, since this code was added to the kernel, the entire kernel is now in violation and belongs to us" is ridiculous. SCO may sue for damages of the code used (IF it exists, which is sketchy at best), and then it will be removed (IF they ever explain what code that is -- if not, any judge with two brain cells is going to force them to or drop the suit). End of story.
--- "The GPL simply falls by the wayside, since the copyright holder (SCO) NEVER INTENDED to distribute the code. It doesn't matter that they were already distributing it, they can argue that they were duped into distributing it because they didn't know what was in it."
Another idiocy, except I can't put it past the courts to not blame SCO on this one. Still, the Linux code that SCO distributed still belongs to the copyright owners who have deemed that the code is not to be distributed unless all attached code is GPLed. SCO, willingly or not, either published their code under the GPL, or violated the IP of hundreds of developers.
"the link you gave really backs up my point... everyone is saying C# is java and C++ done better."
I feel like I'm nitpicking at this point, but C# and.NET are tied together, as are most MS tools. You use one and you're tied into their whole system. I think the collective "C#/.NET" package is a competitor for Java on the detail level, not the conceptual (for as I understand it, the concepts are much the same); that was my entire point. Maybe I should have been clearer in that I mean the whole.NET/C# package.
--- "but what everyone seems to be looking for is a better programming environment, a COMPLETELY new environment; learning from all OO implementations and fixing the broken bits."
I agree; a language just isn't a language any more, it's the entire solution. This is Java/CORBA/what-have-you vs.NET/C#/etc.
--- "this can only be achieved by a new start, not a patching of an older one, or by making an alternative compiler/runtime download (think blackdown and IBM)."
If the change is so significant that the base needs to be heavily modified, yes, but I don't think that's the case. C#/.NET, AFACT, don't handle things much differently than Java/etc., they just add things in addition. Sure, some concepts would have to change, but Java's done that before.
--- "i think it is good that redhat want to up the support into GNU java projects, but i still think this will have little impact on the real executive world of programming, where OSS is still a scary prospect."
Well, first off, this will likely be LGPL or somesuch, as most development libraries are. I doubt that using this open form of Java will require programs to be OSS.
That said, I'd like to note that the company I working for is practically tripping over itself to use OSS as our foundation, primarily because it saves us an ungodly amount of time and money.
--- "Java does have one very good thing going for it at the moment however which i think is much more likely at keeping itself as top-dog: Swing, the GUI replacement for the heavily bloated AWT."
From here: "Java AWT provides low-level widgets such as lists, text fields, and buttons, but no high-level widgets such as trees or rich text. AWT widgets are implemented directly with native widgets on all underlying window systems. Building a UI using AWT alone means programming to the least common denominator of all OS window systems."
There is more, but basically, AWT is the fast, quick and dirty widget toolkit. It's lightweight, it just isn't flexible at all. Bloat is not a word commonly associated with AWT.
Swing, on the other hand, is generally regarded as a slow piece of trash.;) The default theme looks horrible and it's generally regarded as mind-numbingly slow (it's getting much faster). I wouldn't call Swing Sun's ace in the hole. Bloat and Swing are often synonymous.
--- "redhat are only going to finance coders to work on GNU/Linux and thats the point of the article.... but redhat are not going to finance a *BSD port is my point."
Well, I'm speaking out of my arse here, and I'm too lazy to do quick research on this one, but I'd assume that BSD is already supported by most of the big-time OSS Java projects. Regardless, I doubt Red Hat would directly fund *BSD support, although they might not have a choice if they fund the GCJ developers.
--- "one thing is for sure though... we will see a lot more GNU java programs!"
I sincerely hope so. Either that or a new language that can take on MS, if it isn't too late (this thought scares me). Proprietary lock-in is something Microsoft has honed to a science. Their software is looking better by the year, but I can't say I'll ever really invest my time in MS: lock-in is too much of a turn-off and there are so many other good solutions.
That is correct. One of the core points of SWT is that it uses system widgets, which means that GUI resources (fonts, colors, windows, etc.) are all stored in OS memory. Most OSes out there don't have garbage collection, thus SWT does not. You could use finalize statements in the SWT implementation so that when an SWT Java object gets garbage collected the finalize code runs to clean things up, but there is no guarantee as to when the object will be collected -- thus threading problems and disposal ordering problems can occur, not to mention the trash that's left lying about the OS for an undetermined period of time.
1) Freeing resources isn't as bad as some folks make it sound. It is something you have to consider, but many objects will recursively free themselves. Free a Shell (think JFrame, IIRC) and all elements within free themselves. Good OOP habits make this more or less a non-issue unless you're careless or a habitual Java programmer who relies on garbage collection in his sleep.
2) Providing an SWT library and including it in the java call that loads the program is pretty trivial in my opinion. It's no more difficult that making sure all your necessary.jars are included.
--- "At the very least, with an open source Swing..."
I have the distinct feeling that Swing is on its way out the door. It's getting faster and there are enough themes for it, but if people click "Open" and don't get a Windows open dialog, you're going to start having problems. It's just the breaks. Native OS widgets are pretty much the only way to go for most applications. Sorry if I sound like a zealot; I honestly think this is a major point.
Well, as for Red Hat, there are a good many implementations of Java out there, some of which have a lot of promise. I don't see speed to be an issue, to be frank, especially with the resources (internal and external to the company) Red Hat can direct towards the task.
Second, in regards to Sun, I do seriously doubt that they'll want to switch codebases or open up their current codebase. However, given the benefits of OSS, and the changing market... who knows. Microsoft is known for staying alive by adapting, first to the internet, etc. etc. Sun might just wake up and smell the coffee (bad pun intended).
More likely Sun won't do this; if they do support Red Hat in an open project, there's a change code can be adopted from the new project to Sun's java. I'm no expert on projects like GCJ, but if the code is modular enough, I'm sure that parts of it can be adapted without a horrible bit of fuss. Even going past code, ideas can be taken from the programmers going nuts on the open project and incorporated into Sun's codebase.
Maybe we'll see a combination of the two; maybe Sun will help Red Hat in interfacing the two codebases without actually giving Red Hat a verbatim copy of the source. This is all speculation, obviously.
--- "...but i really don't see how this challenges.NET.
Java and.NET are fighting it out at a conceptual level, not at a "my runtime kit is faster than yours" level. "extra features" which the OSS community might add are miniscule compared to that"
If I understand what you're saying, I disagree. The majority of the talk I've heard about.NET is not from MS zealots, but from (often Ex-) Java programmers who are swooning over.NET's features (and those of related languages). Example speaking for C#. Hopefully I'm not nitpicking.
It seems to me that many of the solutions of.NET (and involved languages) and Java (and related solutions [CORBA]) or J2EE are much the same conceptually. Most of the talk I hear from folks moving between one or the other (mostly Java folks to.NET) seems to be tied into the ease-of-use and the features of a language as well as the overall speed (lots of folks complain about the Java VM).
I'm personally not heading a development team, so this is all second hand. Do you have a differing experience that you can elaborate on? Am I not grasping the conceptual differences between the two solutions?
--- "...and lets face it, they will be geared more toward *NIX use than anything else."
I don't know; projects like Mono seem to have a good bit of steam, and in general, OSS folks like to adhear to standards, especially the rabit Java nuts;). Somehow I don't see this becoming much of a problem.
--- "plus we are forgetting this is redhat taking the helm, not like the mozilla project at all... expect GNU/Linux support to be fantastic, don't expect even *BSD support to exist;never mind all the other platforms out there."
If this is truly an OSS project, Red Hat likely won't have a huge influence on where it gets ported to.
I'm admittedly not a Java/.NET expert: I'd love to hear any corrections you might have.:)
"Mp3s are very attractive in that case because you could download music you'd never thought of before and see how you like it. That's not stealing, that's sampling. The RIAA takes 0 loss but gains free visibility and consideration for that."
I wasn't think of it like that. Still, I think you'll agree, this is taking material against the wishes of the seller who says "You want to listen? You buy the whole thing." They might not take a loss for your download if you would have never bought it in the first place, but it still is taking something without permission, which I would equate to stealing. That said, I think sampling is perfectly fine.
--- Quoting something earlier in your post: "However, to do something that actually equates to stealing, you have to do more than download an Mp3. You have to download all the Mp3s in the album, since the RIAA does not sell individual songs much anymore."
This again falls under the terms of their sale. You can't take a few skittles out of a bag, not pay for the bag and not call it theft. Past that, this falls back on the "stealing == monetary loss" discussion above.
It all rolls back into, "How much control of a product should a corporation have?" Off-hand, I'd say all of it.
This is a very interesting conversation, I'd like to hear your thoughts on the above question -- I haven't yet made up my mind on this one (a rare event indeed!:P).
--- "I ended up going to EB and buying a used copy [of GTA3 to replace my scratched one] for $25. Too much hassle."
It's odd how some companies work. You don't own the software you buy, you only have a license to use it -- which can be revoked at any time for any reason at whim of the owner. However, if you break the CD, you have to buy another licence in many cases.
I wonder if you could pay a minimal fee to get a replacement disk? Many companies offer this.
--- "Sorry I babbled so much here."
Dear Lord, go on! It's rare that you find anyone out there with enough of a mind to discuss these things at length. Most folks roll over.
--- "Just so long as you don't feel I'm just some geek who wants to defend his music downloading habit."
Personally, I have no moral problem stealing from companies like Microsoft or the recording industry. I've bought my share of overpriced crap software and overpriced crap CDs -- I have no qualm about evening the odds. I'm not concerned over justification, I'm concerned that folks don't realize what the concept of "property" is. It has special relevance to me as I'd like to go into business for myself.
I'd greatly appreciate any thoughts. If you want to continue this off of/., my email account is listed (spam-armored of course) in my profile. You might make the email subject noticable, if you send me one, the account listed is a spam account.
Slashdot Mirror
You're likely a believer in the "free market" and thus likely a fan of economic theory, so I'll put this in terms that should be familiar:
The RIAA currently has a monopoly on a good known as "popular musicians." The vast majority of consumers willingly pay this monopoly to listen to the music that they love. Consumers (like you and me) are charged with variable costs that are extremely inflated due to the lack of competition. The barriers to entry (leverage to put CDs on shelves) and the economies of scale (manufacturing enough to sell to chain stores) are so difficult to overcome in this instance, other labels have a hard time.
Note that I said "popular musicians" -- not "good musicians."
This is why anti-trust laws were born.
That's what gets me. Taking Napster down for "contributory infringement" is like taking down Smith & Wesson for being an accomplice to a murder. A tool is a tool.
Well, this is an interesting issue. IANACL, but as far as I know, copyright refers to duplicating someone else's work. It is questionable whether downloading a copyrighted MP3 is duplication in that:
1) Copying bits is different in that you MUST copy bits to look at them. A data stream and a file in memory or on a hard drive are necessary for viewing, not just duplicating.
2) The recording industry sells copied disks. That is their business, selling copies of their original work. Copying the song verbatim is copyright, but is copying a CD? If you copy a CD, you've effectively got an exact replica of something that they would sell you: a CD. You have effectively "taken" an item from them without giving them money. I think that fits the definition of stealing.
That said, stealing from the mafia is not immoral in my book. The idealist in me says "Well, don't buy from them and hope they get brought up on charges of breaking anti-trust laws and price gouging." But, the realist in me realizes that our government and courts are already in the pocket of the recording industry. Just look at how most legislation and court decisions have been going.
I can't say that I have.
Am I missing something?
Detachment seems to be a common theme. Rule your company at such a high level that you don't see the details, right?
It really is weird, the trend I'm noticing... bosses get managers to handle certain projects, not caring how they're done as long as they're under budget. I know this isn't the same as replacing "being responsible" with "consequence based thinking", but the trend does seem to be all across the board.
Dear Lord, everyone's talking about replacing the Linux kernel. Why is this even being considered?
Even IF SCO's, uh, "claim" were true (what part of what *NIX is it today?), the related developers and companies can be sued, the code removed and Linux continues. Linux is not controlled or owned by one entity but hundreds. Unless they want a, uh, reverse-class-action lawsuit (do these exist) or make a few hundred cases, Linux is fine.
Quit fretting, the storm will be over soon, immature folks with money are just having a little temper tantrum.
It should be noted that the parent post isn't an argument, per se, but a rant. This is intended for Slashdot, not ZDNet.
:p
Sorry about that
Sometimes I sincerely wonder about RMS. He's got some great ideas that I buy into, he's made some great software and has really pushed software into an area not thought possible...
You can now work your entire IT career using tools that you did not pay for.
What other industry can say that?
That said, you'd think by now he'd know how to make a good argument. Don't preach, don't write your argument sitting on the assumption that you're right and everyone else is wrong. And for God's sakes, don't push semantics.
You wonder why Java fanatics have problems winning people over?
Calm, rational, aggravatingly logical disussion will win people over much faster. I wish RMS, Linux zealots, Java nuts, etc. would pick up on this. Why? Because I absolutely love everything they stand for and preach about... it's not an understatement to say that it's changed my life. I get giddy when I think about it all. But you don't win an argument by pulling someone up the ladder. You win an argument by jumping down to the bottom and climbing with them.
Long story short, I see your point completely -- in some mainly few-and-far-between instances, SWT will need to be upgraded and that requires intervention on your part. You also cannot claim to run your application on more than a few systems.
That said, I still think you're making a mountain out of a mole-hill. For the vast majority of applications that will need a GUI, you're not going to need to implement a version for anything besides Windows/Linux/Mac, and windowing toolkits won't change often enough to pose a frequent problem.
If you want to have the perfect implementation (which there is something to be said for), SWT is not the perfect solution. But considering its native look-and-feel and the speed, SWT is generally the most practical choice.
---
"Also: why IBM did it the way they did instead of pushing SWT via JCP?
When MS does something like that (J++) people are unhappy and speak about embracing, extending... but when IBM does the same everyone is prising it happily?"
Well, SWT is open source, and IBM isn't trying to bastardize Java into making it IBM dependent. If they were, it wouldn't be open source and portable/extendable.
---
"What's the future of swing? Will SWT become a part of J2SE and swing will be deprecated? What's the future of SWT? Will it ever become a part of J2SE thus enabling Pure Java SWT apps? Does anyone like this uncertainty?"
I sincerely hope that somehow, whether through two implementations or one, we get the abilities of 1) platform independance and 2) the option of using native look-and-feel (MS open dialog). I don't care which toolset I use -- I just want a realistic implementation.
1) I understand that many companies are on the ball; mine generally is. (Hopefully I should know, I'm one of the guys whose job it is to fix these problems!). However, the corporate attitude is to make money. I don't think anyone can argue with that. Many tech companies, software companies especially, are so concerned with the bottom line these days that fixing problems and keeping the customer smiling isn't a priority, since they can still sell their product. This is by no means applicable to all companies, but it is a large trend.
2) Although you are correct in pointing out that developers in companies still have "their babies" and do use them, I think you're missing the overall point (which I didn't point out clearly). Companies are, again, after money. There are deadlines, there are insane managers who push for unrealistic release dates and then move on to new software instead of fixing what just went out the door. The vast majority of software companies do it for the money, whereas the vast majority of OSS developers do it for the fun and the personal benefit alone.
Just because you do it for money doesn't mean you won't invest yourself in your work; when I was developing at my company just over a year ago, my projects were my passion. I put my all into them. But a corporate environment isn't a place where you often can take the time to do things right.
I don't know about that. My first Linux installation was plain old Red Hat 6.2 a few years ago (I think it had just come out). I used GNOME and didn't know of Ximian at the time (if it existed at the time).
Man, oh MAN was that a horrible looking interface. It was confusing to read a HOW-TO page walking me through some complex operation on the command line.
But, in the same breath, I spent every second of 30-45 minutes playing with the screensavers. Another 15 spent on themes. And I could use it and the system didn't crash -- I just didn't have the problems I had had with Windows.
Granted I was a developer then, but Linux wasn't near as user friendly as it is now. Most of the folks I show Linux to are, well, found gawking at everything you can do with the desktop.
Users these days are used to crufty interfaces, crashes and bad patches. I don't think Linux will have a hard time being adopted. The folks in my life haven't found it too troubling in itself.
As I've pointed out in previous posts in another Java thread, with the momentum of Red Hat and the entire OSS movement behind it, probably.
Much like Microsoft and the Internet revolution, Sun needs to wake up and smell the coffee (bad pun intended). Sun will die alone. Proprietary lock-in is a science at Microsoft and too big of a threat, in this developer's eyes. It won't last forever, but it very well may outlast Sun.
I heard a suggestion that I'd be much happier with.
Include a clause that says that any user of the software who issues a lawsuit against OSS automatically loses the license to use all software with the same clause in their license.
To illustrate this, if this clause were in the LGPL, any company suing an OSS project would be unable to use any LGPL'ed software.
Archaic? Draconian? Too risky you say?
Read your EULA lately?
"You're probably not familiar with how this normally works. If you license a product from IBM or MS it's not like all your communications go from the very top down to the bottom and they simply dump a product on your lap and leave you alone. You are generally (always?) given a contact with the team lead(s) and have a way to contact the people who coded the software itself. To say that licensing from a big corporation means your concerns won't be heard unless you are also a big corporation is pretty much dead wrong."
;)
If this is the case, then I stand corrected. Somehow, though, from what I've heard from folks I work with, the things I've seen personally and the things I read about online (in dar intarnet huh huh), companies like MS generally aren't inclinded to fix problems.
---
"Again, open source developers don't have a business relationship with you, so they don't have any real incentive to be attentive to your needs. They don't have to fix anything."
As someone else suggested, throw them some money to fix a bug. Many OSS developers get cash this way. It will almost certainly be cheaper to throw him some cash than to fix it yourself, and will likely be about as expensive as an "incident report" or yearly support contract.
---
"There are many developers who may just flat out refuse to alter the way their program works, even if it's minor thing, because "it's their baby"."
How many medium to large corporations have you known of to alter their software at a single customer's request? This practice is generally frowned upon in all but the smallest companies, unless, of course, the company is in the business of selling solutions, which most medium-larged sized companies aren't.
I don't know that I agree with that. In my humble opinion, Gaim is one of the best pieces of software out there. It's flexible as hell, fast, configurable in many ways, and in just the right areas, and has some really useful features (buddy pouncing, multiple accounts on different services, spellchecking, aliasing, etc.). All that and I still feel like it's one of the most user friendly applications for Linux. And Windows.
:)
:)
They list 7 people as "developers" and another 5 as "contributors". How much or how little each of them develops/contributes I don't know, but I get the feeling that this isn't a "major project".
Maybe our opinions differ, maybe this is just an isolated example, but I've worked with enough tiny projects that have helped me quite a bit to feel comfortable saying that there is a wealth of projects under the top few that rock the house.
As for the 'dead projects', we can go into the ratio of dead/alive closed source companies compared to that of OSS projects. We can even bring in CS shareware if you like.
Please feel free to correct me if I'm missing something.
Cheers
The SWT libraries for all systems can be sent with an application and invoked dynamically. One .jar file is included in the classpath and one directory containing libraries is invoked dynamically in the call to the VM. This is about as trivial as you can get.
.jar and handful of library files. For existing customers? The same thing.
---
"go crazy with support, cause now it's yours - app developer's - responsibility to maintain your packages - upgrade for any future changes in GUI subsystems of any of your target OSes (YOU will have to repackege your software with the new versions of SWT, because Sun is not going to distrbute them)."
I will admit that changes in the GUI widget set can force you to send updates to the *SWT libraries*, but odd are that this will happen far less often than updates to your own product. This "you'll have to repackage your whole application" is pretty ridiculous. To update the version repository of your software, just copy in a new
You argument has some merit, but I really don't see this as anything but a trivial obstacle. Maybe I misread or missed something altogether in your post?
Cheers
"you could spend a couple days trying to track down and fix a problem that is preventing you from finishing your real work or you could simply send a problem description off to the guys who originally wrote the code (and who you're already paying for), and they'll probably be able to fix the problem far quicker than you ever could."
:)
This point does have merit, but I suppose you're talking about problems with small to mid sized companies. Big ones (Like Microsoft, IBM, Novell, etc.) aren't going to fix your problems any time soon; they have so many customers, unless you're a big-shot (who can spend the $$$ to fix it yourself), they won't care. Have you had the pleasure of paying money for an incident to report a bug to MS?
---
"What you do know is that you don't have the ability to really push open source developers to fix problems for you, and that's a major drawback."
No, but truth be told, most OSS developers aren't working on their projects because they're told to or are being paid. They do it because their projects are their babies.
Typically, problems reported are fixed in relatively short order because:
1) you've likely, in the process of debugging your own software or tools, have isolated the problem to a specific area and pattern and
2) OSS folks almost always use the software they write themselves. That's usually why they write it in the first place.
Your argument has some weight, but I don't think it pans out in the large scheme of things.
[Apple Switch] My name is Dalcius, and I'm a Software Analyst (debugger) for a mid-sized company that deals with semi-large ones. [/Apple Switch]
I think I can get a reply out by looking at your last paragraph:
"I should probably clarify something I said before then. I'm concerned you'll feel that I'm arguing justification and not property. My argument is actually a blurring of the two. The labelling of 'theft' by the music industry is not a reflection of actual monetary losses, but a defense aimed at preserving their lucrative business model. What's happening here is the RIAA is screwing it's customers, and they're defending themselves by violating the terms of service. Nobody's been given the chance to pay for what they've downloaded. They can enjoy the song for years, at some point along the way they should pay for it. It's hard to call it theft when the door to even up is still open."
If I understand you correctly, your basis for this whole argument is this:
1) The major labels effectively act as one company
2) That "company" holds a monopoly on most music
3) That "company" charges overwhelmingly inflated prices for its goods.
4) There is enough demand for these expensive goods because they can't be bought any cheaper (see #2).
5) Thus, the consumer is being strongarmed into paying too much money for a product.
6) In fair retaliation, customers violate the ownership of the major labels in order to sample the music, sometimes not buying a CD that they would have if those tracks were short demos, expired after a week, etc.
A staunch capitalist would reply that consumers aren't forced into buying the music, thus they are not being strongarmed. In theory, I would agree, however in practice, with the current barriers to entry for startup labels and with the fact that nearly *everyone* needs music (as much as they need things to read and TV to watch -- it's a part of our daily lives), I do agree that things should change.
However, that said:
1) I do not have a moral problem sampling their music or even downloading and keeping entire CDs without paying for them.
2) However, that is, in my mind, still stealing: you benefit from their work without rewarding them, one of the conditions of their labor.
3) Referring to #1, the "right" way to do things would be to bring up the labels on charges of anti-trust violations and of price gouging.
This, however, assumes that the public isn't apathetic and that politicians aren't bought and paid for.
Summary: You got me to take things I'd already known and really put together the big picture: these folks are as bad as Microsoft (the similarities are stunning, no?).
That was fun; thanks for the new point of view and for listening to mine. =)
Another interesting point:
Based on their original (albeit vague) allegations, how can SCO claim that they didn't know that they distributed their IP under the GPL when IBM can claim the same ignorance?
This is ridiculous.
"If the Linux kernel is a derivate work of SCO's (or more likely, certain PARTS of the kernel are derivative works), then SCO can simply claim that ALL copies of the Linux kernel are infringing copies, and ALL public distributions of the code are copyright violations."
The Linux Kernel is a work owned by many people; the copyrights belong to no one person. To declare that, "Well, since this code was added to the kernel, the entire kernel is now in violation and belongs to us" is ridiculous. SCO may sue for damages of the code used (IF it exists, which is sketchy at best), and then it will be removed (IF they ever explain what code that is -- if not, any judge with two brain cells is going to force them to or drop the suit). End of story.
---
"The GPL simply falls by the wayside, since the copyright holder (SCO) NEVER INTENDED to distribute the code. It doesn't matter that they were already distributing it, they can argue that they were duped into distributing it because they didn't know what was in it."
Another idiocy, except I can't put it past the courts to not blame SCO on this one. Still, the Linux code that SCO distributed still belongs to the copyright owners who have deemed that the code is not to be distributed unless all attached code is GPLed. SCO, willingly or not, either published their code under the GPL, or violated the IP of hundreds of developers.
"the link you gave really backs up my point... everyone is saying C# is java and C++ done better."
.NET are tied together, as are most MS tools. You use one and you're tied into their whole system. I think the collective "C#/.NET" package is a competitor for Java on the detail level, not the conceptual (for as I understand it, the concepts are much the same); that was my entire point. Maybe I should have been clearer in that I mean the whole .NET/C# package.
.NET/C#/etc.
;) The default theme looks horrible and it's generally regarded as mind-numbingly slow (it's getting much faster). I wouldn't call Swing Sun's ace in the hole. Bloat and Swing are often synonymous.
... but redhat are not going to finance a *BSD port is my point."
I feel like I'm nitpicking at this point, but C# and
---
"but what everyone seems to be looking for is a better programming environment, a COMPLETELY new environment; learning from all OO implementations and fixing the broken bits."
I agree; a language just isn't a language any more, it's the entire solution. This is Java/CORBA/what-have-you vs
---
"this can only be achieved by a new start, not a patching of an older one, or by making an alternative compiler/runtime download (think blackdown and IBM)."
If the change is so significant that the base needs to be heavily modified, yes, but I don't think that's the case. C#/.NET, AFACT, don't handle things much differently than Java/etc., they just add things in addition. Sure, some concepts would have to change, but Java's done that before.
---
"i think it is good that redhat want to up the support into GNU java projects, but i still think this will have little impact on the real executive world of programming, where OSS is still a scary prospect."
Well, first off, this will likely be LGPL or somesuch, as most development libraries are. I doubt that using this open form of Java will require programs to be OSS.
That said, I'd like to note that the company I working for is practically tripping over itself to use OSS as our foundation, primarily because it saves us an ungodly amount of time and money.
---
"Java does have one very good thing going for it at the moment however which i think is much more likely at keeping itself as top-dog: Swing, the GUI replacement for the heavily bloated AWT."
From here:
"Java AWT provides low-level widgets such as lists, text fields, and buttons, but no high-level widgets such as trees or rich text. AWT widgets are implemented directly with native widgets on all underlying window systems. Building a UI using AWT alone means programming to the least common denominator of all OS window systems."
There is more, but basically, AWT is the fast, quick and dirty widget toolkit. It's lightweight, it just isn't flexible at all. Bloat is not a word commonly associated with AWT.
Swing, on the other hand, is generally regarded as a slow piece of trash.
---
"redhat are only going to finance coders to work on GNU/Linux and thats the point of the article.
Well, I'm speaking out of my arse here, and I'm too lazy to do quick research on this one, but I'd assume that BSD is already supported by most of the big-time OSS Java projects. Regardless, I doubt Red Hat would directly fund *BSD support, although they might not have a choice if they fund the GCJ developers.
---
"one thing is for sure though... we will see a lot more GNU java programs!"
I sincerely hope so. Either that or a new language that can take on MS, if it isn't too late (this thought scares me). Proprietary lock-in is something Microsoft has honed to a science. Their software is looking better by the year, but I can't say I'll ever really invest my time in MS: lock-in is too much of a turn-off and there are so many other good solutions.
That is correct. One of the core points of SWT is that it uses system widgets, which means that GUI resources (fonts, colors, windows, etc.) are all stored in OS memory. Most OSes out there don't have garbage collection, thus SWT does not. You could use finalize statements in the SWT implementation so that when an SWT Java object gets garbage collected the finalize code runs to clean things up, but there is no guarantee as to when the object will be collected -- thus threading problems and disposal ordering problems can occur, not to mention the trash that's left lying about the OS for an undetermined period of time.
This is all described in detail here.
Agreed. Two things to point out:
.jars are included.
1) Freeing resources isn't as bad as some folks make it sound. It is something you have to consider, but many objects will recursively free themselves. Free a Shell (think JFrame, IIRC) and all elements within free themselves. Good OOP habits make this more or less a non-issue unless you're careless or a habitual Java programmer who relies on garbage collection in his sleep.
2) Providing an SWT library and including it in the java call that loads the program is pretty trivial in my opinion. It's no more difficult that making sure all your necessary
---
"At the very least, with an open source Swing..."
I have the distinct feeling that Swing is on its way out the door. It's getting faster and there are enough themes for it, but if people click "Open" and don't get a Windows open dialog, you're going to start having problems. It's just the breaks. Native OS widgets are pretty much the only way to go for most applications. Sorry if I sound like a zealot; I honestly think this is a major point.
Well, as for Red Hat, there are a good many implementations of Java out there, some of which have a lot of promise. I don't see speed to be an issue, to be frank, especially with the resources (internal and external to the company) Red Hat can direct towards the task.
.NET.
.NET are fighting it out at a conceptual level, not at a "my runtime kit is faster than yours" level. "extra features" which the OSS community might add are miniscule compared to that"
.NET is not from MS zealots, but from (often Ex-) Java programmers who are swooning over .NET's features (and those of related languages). Example speaking for C#. Hopefully I'm not nitpicking.
.NET (and involved languages) and Java (and related solutions [CORBA]) or J2EE are much the same conceptually. Most of the talk I hear from folks moving between one or the other (mostly Java folks to .NET) seems to be tied into the ease-of-use and the features of a language as well as the overall speed (lots of folks complain about the Java VM).
;). Somehow I don't see this becoming much of a problem.
;never mind all the other platforms out there."
:)
Second, in regards to Sun, I do seriously doubt that they'll want to switch codebases or open up their current codebase. However, given the benefits of OSS, and the changing market... who knows. Microsoft is known for staying alive by adapting, first to the internet, etc. etc. Sun might just wake up and smell the coffee (bad pun intended).
More likely Sun won't do this; if they do support Red Hat in an open project, there's a change code can be adopted from the new project to Sun's java. I'm no expert on projects like GCJ, but if the code is modular enough, I'm sure that parts of it can be adapted without a horrible bit of fuss. Even going past code, ideas can be taken from the programmers going nuts on the open project and incorporated into Sun's codebase.
Maybe we'll see a combination of the two; maybe Sun will help Red Hat in interfacing the two codebases without actually giving Red Hat a verbatim copy of the source. This is all speculation, obviously.
---
"...but i really don't see how this challenges
Java and
If I understand what you're saying, I disagree. The majority of the talk I've heard about
It seems to me that many of the solutions of
I'm personally not heading a development team, so this is all second hand. Do you have a differing experience that you can elaborate on? Am I not grasping the conceptual differences between the two solutions?
---
"...and lets face it, they will be geared more toward *NIX use than anything else."
I don't know; projects like Mono seem to have a good bit of steam, and in general, OSS folks like to adhear to standards, especially the rabit Java nuts
---
"plus we are forgetting this is redhat taking the helm, not like the mozilla project at all... expect GNU/Linux support to be fantastic, don't expect even *BSD support to exist
If this is truly an OSS project, Red Hat likely won't have a huge influence on where it gets ported to.
I'm admittedly not a Java/.NET expert: I'd love to hear any corrections you might have.
Cheers
"Mp3s are very attractive in that case because you could download music you'd never thought of before and see how you like it. That's not stealing, that's sampling. The RIAA takes 0 loss but gains free visibility and consideration for that."
:P).
/., my email account is listed (spam-armored of course) in my profile. You might make the email subject noticable, if you send me one, the account listed is a spam account.
I wasn't think of it like that. Still, I think you'll agree, this is taking material against the wishes of the seller who says "You want to listen? You buy the whole thing." They might not take a loss for your download if you would have never bought it in the first place, but it still is taking something without permission, which I would equate to stealing. That said, I think sampling is perfectly fine.
---
Quoting something earlier in your post:
"However, to do something that actually equates to stealing, you have to do more than download an Mp3. You have to download all the Mp3s in the album, since the RIAA does not sell individual songs much anymore."
This again falls under the terms of their sale. You can't take a few skittles out of a bag, not pay for the bag and not call it theft. Past that, this falls back on the "stealing == monetary loss" discussion above.
It all rolls back into, "How much control of a product should a corporation have?" Off-hand, I'd say all of it.
This is a very interesting conversation, I'd like to hear your thoughts on the above question -- I haven't yet made up my mind on this one (a rare event indeed!
---
"I ended up going to EB and buying a used copy [of GTA3 to replace my scratched one] for $25. Too much hassle."
It's odd how some companies work. You don't own the software you buy, you only have a license to use it -- which can be revoked at any time for any reason at whim of the owner. However, if you break the CD, you have to buy another licence in many cases.
I wonder if you could pay a minimal fee to get a replacement disk? Many companies offer this.
---
"Sorry I babbled so much here."
Dear Lord, go on! It's rare that you find anyone out there with enough of a mind to discuss these things at length. Most folks roll over.
---
"Just so long as you don't feel I'm just some geek who wants to defend his music downloading habit."
Personally, I have no moral problem stealing from companies like Microsoft or the recording industry. I've bought my share of overpriced crap software and overpriced crap CDs -- I have no qualm about evening the odds. I'm not concerned over justification, I'm concerned that folks don't realize what the concept of "property" is. It has special relevance to me as I'd like to go into business for myself.
I'd greatly appreciate any thoughts. If you want to continue this off of
Cheers