I work for Adaptec on a next generation storage management software project. The project is in Java, with some C/C++ for hardware interfacing. That being said, I'm 25 and the youngest engineer on the project by eight years. My team lead is old enough to be my father, and my manager is even older! So no, age does not matter as much as you'd like to think. As long as you are willing to grow, learn, and take on new challeges (or even management!), you will never be obsolete.
"I shoulda never sent a penguin out to do a daemon's work."
Some examples: - Milk is good for you (says who? This should be an urban legend. For hundreds of thousands of years (at least) humans lived without drinking milk (and this means milk from another animal, and drinking milk as an adult, and drinking pasteurized milk), why is it now milk is so popular? Give you a hint: "Got Milk?"
Ok, this is silly. It's like saying that humans lived for hundreds of thousands of years without antibiotics, so therefore antibiotics are not good for you, and instead are a marketing vehicle for evil pharmaceutical companies to take your money. Come on! Yeah, way off topic, but this crap irks me.
"I shoulda never sent a penguin out to do a daemon's work."
I'm guessing that part of it is the Java It most likely is Java. Under NT, Java maps large chunks of memory, even when it has no need for it. The actual memory usage is much lower than the reported memory usage because of this. I know of a few Java apps that implement a native Win32 call to unmap this memory and let the JVM page-fault it back in as needed, thereby greatly reducing the reported footprint.
"I shoulda never sent a penguin out to do a daemon's work."
If you are looking to do your project in Java, a great packaging tool that you should look at is InstallAnywhere by ZeroG. It provides the same functionality as InstallShield, but is java based. It also allows you to bundle the JVM of your choice with your product, to allow for seamless installation and execution. IA is used as the installer for JBuilder (although no JVM is bundled), and I'm using it for an upcoming product at my work (Adaptec) (no, it's not a Linux version of EasyCD either =-). It supports Windows, MacOS, and Solaris out of the box, and will support virtually any java enabled Unix with trivial changes that ZeroG tech support is happy to provide. I know I sound like a sales rep for ZeroG here, but it has really made my life a lot easier considerig the nature of the project I'm on.
"I shoulda never sent a penguin out to do a daemon's work."
Man this takes me back. I remember ten years ago wandering around the BBS scene in the Ann Arbor area and discovering M-Net. At that time I was using an Atari 800XL, so I didn't even know what a DOS prompt was, let alone a shell prompt ("what the heck is this '$' and why can't I do anything with it?"). I totally forgot about about it once I went to U of Mich and discovered the internet. Man... noce to know that it is still around.
"I shoulda never sent a penguin out to do a daemon's work."
Hate to nit-pick here. I know that the switchover to 4.4Lite was done for 2.0. But there was a second release of 4.4lite called 4.4lite2. This was integrated in for 3.0. If you don't believe me, I can show you the CVS tags.
"I shoulda never sent a penguin out to do a daemon's work."
The 4.0 development branch spun off of the 3.x tree back last January '99. I don't consider 14 months to be a incredibly long development cycle. Going from 2.2 to 3.0 took a long time (a few years IIRC), but it also involved the codebase switch to 4.4Lite2, along with i386 SMP and Alpha introduction. None of those were trivial tasks!
"I shoulda never sent a penguin out to do a daemon's work."
All eye rolling and snickering aside, I think that the owner of this patent deserves a Slashdot interview. That way we can start to determine whether: A. This is a revolutionary invention that puts us one step closer to Star Trek. B. This invention is completely misunderstood and does something less-than-revolutionary, but still interesting. C. This invention has been translated to lawyer-speak and is really just a toaster. D. This guy is total quack with too much time and money.
"I shoulda never sent a penguin out to do a daemon's work."
"Approved open source license"? Come on now! Since when does Sun owe you, me, or anyone else anything? Sun can release their products under whatever license they feel like. I'm thrilled that they're releasing anything. Do you have any idea how much energy it takes to change corporate inertia and convince the managers, directors, and board members that releasing source is good? I'm impressed that they do anything at all. And I'm even more impressed that they don't get pissed off by people whinning about the steps that they take. Stand down your GPL jihad and give them a break!
"I shoulda never sent a penguin out to do a daemon's work."
Oh come on now. This is just silly. Just cause Jordan didn't mention Jesus Monroy Jr. (who, btw, is hightly suspected to be an out of control PERL script with NNTP access), Terry Lambert, et al, doesn't mean that he is also busy wiping all evidence of due credit to them from the archives. And if you hold a lot of sympathy for the Jolitz's, then you obviously were not involved in the project way back then. Jeez. And so what if BSDi and FreeBSD codebases merge to the dissatifaction of the developers and users. Just take a pre-merged version and turn it into "ReallyFreeBSD".
This merger announcement has me shaken. The FreeBSD trademark has long been held by Walnut Creek CDROM, with some rather arbitrary restrictions on its use to describe derivative works that improve things such as the install process, or the overall "new user" experience. I notice that Jordan Hubbard's FreeBSD History at FreeBSD History is a bit revisionist. It casually mentions Bill Jolitz, but fails to mention John Sokol, who posted the original 386BSD code to the net, Jesus Monroy, who had a lot to do with the move from 386BSD 0.0 to 386BSD 0.1, or Terry Lambert, who wrote the original FAQ, patch kit, and patch kit production software, and then handed it off (as far as I can tell from the archives, the phrase "the patch kits last 3 coordinators: Nate Williams, Rod Grimes and myself" means "everyone ever involved in administering or creating the patch kit, except Terry"). It also fails to mention that Bill Jolitz had originally lent official support a "386BSD 0.5" interim release, based on the patch kit work, or that "the patchkit swelled unconfortably every day" is a paraphrase of Lynne Jolitz's complaints, which resulted in Bill Jolitz's "rude" withdrawl of support. This seems antithetical to the BSD credo of "credit when due", and bodes ill for a future where it appears the source tree will have to be closed down for a while, while proprietary bits are sorted out of BSD/OS for integration. I am unsure of what will occur as a result of this merger, and the idea of the merger itself makes me rather anxious about the historically centralized control of the project by a few people who initially checked in a large amount of code, but later became a barrier to progress, and the centralized control of the "BSD" related trademarks. People who argue over such issues have long suggested that it would be possible for a commercial entity to "hijack" a BSD project; I have always publically dismissed this, but it seems that perhaps I was wrong. If so, I have a lot of crow to eat, which would be made ever more bitter for it being the result of failed idealism on my part. I hope that statements are issued soon, clarifying what this merger is going to mean, the process by which it is going to be accomplished, the status of the use of BSD/OS binary-only drivers in a future FreeBSD, the permissable public use of the FreeBSD trademark going forward (preferrably, a trust will be established), and so on. I feel like Tim O'Reilly, decrying the Amazon patents. 8-(
"Pioneers get arrows in the back. Settlers get the land. I'm happy letting Linux be the pioneers."
I looks like you can access the devclass documents now. I just went to http://www.usb.org/developers/devclass.html and a login is no longer required. Only the "Approved Class specifications" docs are available though... the Draft documents are not there. Hmmmm....
"Pioneers get arrows in the back. Settlers get the land. I'm happy letting Linux be the pioneers."
oh hell yeah....anyone that was ever lucky enuf to own a an old TI-85 computer (for those that weren't, it was something like the commie 64-128)...
Ummm... the TI-85 is a graphing calculator (and a crappy one at that.... the HP series kicks its' ass). You're thinking of the TI-99/4(A). Excellent starter computer... even if it did have a severely brain damaged 16-bit CPU. I still fire it up in emulation once in a while to play Parsec. And the Speech Synthesizer module was every kid's dream!
Pioneers get arrows in the back. Settlers get the land. I'm happy letting Linux be the pioneers.
If you would have read the entire article, you would have seen that Linus eventually reversed his position and agreed that the optimization (used in FreeBSD) was safe and correct. Man, I really want to turn on the flame switch when I read uneducated posts like yours. You make a perfect example of the reason that Linux is slowly gaining a black eye in the enterprise world.
Regardless of who owns J++, they still have to keep within Sun's guidelines to call it Java. The Java licence doesn't pick on Microsoft; everyone has to play by the rules. Personally, I'm looking forward to seeing J++ wither away.
Slashdot Mirror
I work for Adaptec on a next generation storage management software project. The project is in Java, with some C/C++ for hardware interfacing. That being said, I'm 25 and the youngest engineer on the project by eight years. My team lead is old enough to be my father, and my manager is even older! So no, age does not matter as much as you'd like to think. As long as you are willing to grow, learn, and take on new challeges (or even management!), you will never be obsolete.
"I shoulda never sent a penguin out to do a daemon's work."
Some examples: - Milk is good for you (says who? This should be an urban legend. For hundreds of thousands of years (at least) humans lived without drinking milk (and this means milk from another animal, and drinking milk as an adult, and drinking pasteurized milk), why is it now milk is so popular? Give you a hint: "Got Milk?"
Ok, this is silly. It's like saying that humans lived for hundreds of thousands of years without antibiotics, so therefore antibiotics are not good for you, and instead are a marketing vehicle for evil pharmaceutical companies to take your money. Come on! Yeah, way off topic, but this crap irks me.
"I shoulda never sent a penguin out to do a daemon's work."
I'm guessing that part of it is the Java
It most likely is Java. Under NT, Java maps large chunks of memory, even when it has no need for it. The actual memory usage is much lower than the reported memory usage because of this. I know of a few Java apps that implement a native Win32 call to unmap this memory and let the JVM page-fault it back in as needed, thereby greatly reducing the reported footprint.
"I shoulda never sent a penguin out to do a daemon's work."
If you are looking to do your project in Java, a great packaging tool that you should look at is InstallAnywhere by ZeroG. It provides the same functionality as InstallShield, but is java based. It also allows you to bundle the JVM of your choice with your product, to allow for seamless installation and execution. IA is used as the installer for JBuilder (although no JVM is bundled), and I'm using it for an upcoming product at my work (Adaptec) (no, it's not a Linux version of EasyCD either =-). It supports Windows, MacOS, and Solaris out of the box, and will support virtually any java enabled Unix with trivial changes that ZeroG tech support is happy to provide.
I know I sound like a sales rep for ZeroG here, but it has really made my life a lot easier considerig the nature of the project I'm on.
"I shoulda never sent a penguin out to do a daemon's work."
Naw... I used my last point on him... After that posting is fair game
"I shoulda never sent a penguin out to do a daemon's work."
Go out to Borders, pick up the book on GTK/Gnome development, and read the FAQ chapter. There you will see why I moderated you down
"I shoulda never sent a penguin out to do a daemon's work."
To quote from The Princess Bride: "You keep on saying that word. I do not think it means what you think it means."
"I shoulda never sent a penguin out to do a daemon's work."
BSD 3.1
That's BSD/OS 3.1, a derivative of BSD 4.4. See the latest article about BSDi and FreeBSD.
When was the first version of BSD launched at Berkeley?
Looking up my Unix family tree, it shows 1BSD being in late '76.
"I shoulda never sent a penguin out to do a daemon's work."
Man this takes me back. I remember ten years ago wandering around the BBS scene in the Ann Arbor area and discovering M-Net. At that time I was using an Atari 800XL, so I didn't even know what a DOS prompt was, let alone a shell prompt ("what the heck is this '$' and why can't I do anything with it?"). I totally forgot about about it once I went to U of Mich and discovered the internet. Man... noce to know that it is still around.
"I shoulda never sent a penguin out to do a daemon's work."
Hate to nit-pick here. I know that the switchover to 4.4Lite was done for 2.0. But there was a second release of 4.4lite called 4.4lite2. This was integrated in for 3.0. If you don't believe me, I can show you the CVS tags.
"I shoulda never sent a penguin out to do a daemon's work."
The 4.0 development branch spun off of the 3.x tree back last January '99. I don't consider 14 months to be a incredibly long development cycle. Going from 2.2 to 3.0 took a long time (a few years IIRC), but it also involved the codebase switch to 4.4Lite2, along with i386 SMP and Alpha introduction. None of those were trivial tasks!
"I shoulda never sent a penguin out to do a daemon's work."
All eye rolling and snickering aside, I think that the owner of this patent deserves a Slashdot interview. That way we can start to determine whether:
A. This is a revolutionary invention that puts us one step closer to Star Trek.
B. This invention is completely misunderstood and does something less-than-revolutionary, but still interesting.
C. This invention has been translated to lawyer-speak and is really just a toaster.
D. This guy is total quack with too much time and money.
"I shoulda never sent a penguin out to do a daemon's work."
"Approved open source license"? Come on now! Since when does Sun owe you, me, or anyone else anything? Sun can release their products under whatever license they feel like. I'm thrilled that they're releasing anything. Do you have any idea how much energy it takes to change corporate inertia and convince the managers, directors, and board members that releasing source is good? I'm impressed that they do anything at all. And I'm even more impressed that they don't get pissed off by people whinning about the steps that they take. Stand down your GPL jihad and give them a break!
"I shoulda never sent a penguin out to do a daemon's work."
Oh come on now. This is just silly. Just cause Jordan didn't mention Jesus Monroy Jr. (who, btw, is hightly suspected to be an out of control PERL script with NNTP access), Terry Lambert, et al, doesn't mean that he is also busy wiping all evidence of due credit to them from the archives. And if you hold a lot of sympathy for the Jolitz's, then you obviously were not involved in the project way back then. Jeez. And so what if BSDi and FreeBSD codebases merge to the dissatifaction of the developers and users. Just take a pre-merged version and turn it into "ReallyFreeBSD".
This merger announcement has me shaken. The FreeBSD trademark has long been held by Walnut Creek CDROM, with some rather arbitrary restrictions on its use to describe derivative works that improve things such as the install process, or the overall "new user" experience. I notice that Jordan Hubbard's FreeBSD History at FreeBSD History is a bit revisionist. It casually mentions Bill Jolitz, but fails to mention John Sokol, who posted the original 386BSD code to the net, Jesus Monroy, who had a lot to do with the move from 386BSD 0.0 to 386BSD 0.1, or Terry Lambert, who wrote the original FAQ, patch kit, and patch kit production software, and then handed it off (as far as I can tell from the archives, the phrase "the patch kits last 3 coordinators: Nate Williams, Rod Grimes and myself" means "everyone ever involved in administering or creating the patch kit, except Terry"). It also fails to mention that Bill Jolitz had originally lent official support a "386BSD 0.5" interim release, based on the patch kit work, or that "the patchkit swelled unconfortably every day" is a paraphrase of Lynne Jolitz's complaints, which resulted in Bill Jolitz's "rude" withdrawl of support. This seems antithetical to the BSD credo of "credit when due", and bodes ill for a future where it appears the source tree will have to be closed down for a while, while proprietary bits are sorted out of BSD/OS for integration. I am unsure of what will occur as a result of this merger, and the idea of the merger itself makes me rather anxious about the historically centralized control of the project by a few people who initially checked in a large amount of code, but later became a barrier to progress, and the centralized control of the "BSD" related trademarks. People who argue over such issues have long suggested that it would be possible for a commercial entity to "hijack" a BSD project; I have always publically dismissed this, but it seems that perhaps I was wrong. If so, I have a lot of crow to eat, which would be made ever more bitter for it being the result of failed idealism on my part. I hope that statements are issued soon, clarifying what this merger is going to mean, the process by which it is going to be accomplished, the status of the use of BSD/OS binary-only drivers in a future FreeBSD, the permissable public use of the FreeBSD trademark going forward (preferrably, a trust will be established), and so on. I feel like Tim O'Reilly, decrying the Amazon patents. 8-(
"Pioneers get arrows in the back. Settlers get the land. I'm happy letting Linux be the pioneers."
I looks like you can access the devclass documents now. I just went to http://www.usb.org/developers/devclass.html and a login is no longer required. Only the "Approved Class specifications" docs are available though... the Draft documents are not there. Hmmmm....
"Pioneers get arrows in the back. Settlers get the land. I'm happy letting Linux be the pioneers."
After reading both the Time and CNN online stories, I have one big question.... what news agency is not parented by Time/Warner/AOL/Beelzebub now?
Pioneers get arrows in the back. Settlers get the land. I'm happy letting Linux be the pioneers.
oh hell yeah....anyone that was ever lucky enuf to own a an old TI-85 computer (for those that weren't, it was something like the commie 64-128)... Ummm... the TI-85 is a graphing calculator (and a crappy one at that.... the HP series kicks its' ass). You're thinking of the TI-99/4(A). Excellent starter computer... even if it did have a severely brain damaged 16-bit CPU. I still fire it up in emulation once in a while to play Parsec. And the Speech Synthesizer module was every kid's dream!
Pioneers get arrows in the back. Settlers get the land. I'm happy letting Linux be the pioneers.
Yeah, it went in around early December. Only on -current though (which is where experimental stuff is supposed to go).
If you would have read the entire article, you would have seen that Linus eventually reversed his position and agreed that the optimization (used in FreeBSD) was safe and correct. Man, I really want to turn on the flame switch when I read uneducated posts like yours. You make a perfect example of the reason that Linux is slowly gaining a black eye in the enterprise world.
Gattica. Great movie. Touches on many of these concerns.
Regardless of who owns J++, they still have to keep within Sun's guidelines to call it Java. The Java licence doesn't pick on Microsoft; everyone has to play by the rules. Personally, I'm looking forward to seeing J++ wither away.