... someone will finally add native support for UNIX-domain sockets to Java (for those platforms that support them). (I know there's third-party JNI thingies that provide this capability, but why isn't it native?)
I've never understood the notion of sacrificing the ability to use facilities provided by the operating system in the name of portability. I'm sure the same kinds of situations exist under Windows.
WTF, Java designers? Maybe I just want to write some software, and I don't care whether it's portable or not.
Now maybe that it's GPLed, people can finally start changing it so that we can finally start using Java and OS facilities at the same time .
And under the GPL I can pretty well use Linux any way I want, especially if I'm not distributing anything.
Right, but if you're a business distributing Linux and making money off that and also, say, selling support services, and if any of the OSS you're distributing and making money on happens to infringe on someone else's software patents, then you and your customers can get your asses sued off.
Micro Soft is promising not to sue Suse users' asses off. That's what this is about.
A 320kbps MP3 with Q=1 and 320kbps mp3 with Q=9 are WILDLY different...
And "lame '--vbr-new -h --preset standard" will save a butt-load of disk (or iPod) space, and will sound great for about 99% of the non-audiophile-nerdboys.
Not only that, but it's is how all the MP3s available from Emusic.com are encoded (non-DRM, too!). Highly recommended.
Clearly certs aren't enough to maintain a company of RedHat's size...
Maybe so, but having the certs is the ONLY way that Linux of any kind would be approved to deploy on certain networks. No certs, no deployment. Pick a different OS platform. Period. As far as having and maintaining these certs go, RHEL is the only game in town.
We've been waiting for years for someone to get a Linux distribution, any Linux distribution, approved for a particular DoD network that we deploy our systems on. We can't do it ourselves because the process is too onerous and expensive for us to take on. Nonetheless, we'd LOVE to deploy our systems on this network using a Linux platform, but until something gets approved we're stuck with Solaris. Solaris-8 on SPARC, that is. It's the only Unix OS approved for this network. We don't make the rules, but if we want to stay in business, we have to follow them.
That RHEL4 is Common Criteria Scheme certified at CAPP/EAL4+ is just the first step toward getting an OS platform certified to deploy on this network. Without these certs, there'd be no way. If Red Hat goes out of business, then there will be networks out there that will never, ever see a Linux distribution.
In fact, we're aware of a particular program that's done a lot of the heavy lifting to get RHEL-4 certified for deploying on this network. The certifying authority is close to making a decesion. Now, I suspect this authority will just say no. Thanks a lot, Larry. You've torpedoed several avenues of innovation.
If Oracle would guarantee that they'll maintain the certs, etc, then maybe there'd be a chance. But I don't trust Oracle.
They say that you can't be a great writer unless and until you've done close readings of great works of literature.
Similarly, it's extremely elightening to do close readings of source code of software projects you consider "great". Pick a "great" (or even not-so-great) software project for which you have access to the source code. Download the source. Then, do a close reading of the source. Understand how the software works, control flow, data organization, class heirarchies, etc., even the build system.
You might decide that in the end that it's crap, but you'll never know until you read it. But it does provide a good method for learning a lot about how software is put together. Many (but not all) open-source projects are very close to doing best practices due to the collective nature of the development.
... seamless integration with iTunes (not to be confused with ITMS, which is a topic for a different article)
iTunes is a pretty good juke box package, and it provides for uploading to a player (and synchronizing, too) from inside the application. I won't touch any portable media player unless I can manage it (ie upload) via iTunes.
With computers becoming so small and easy to distribute over a wireless network, do we really need data centers to house computers or are the computers going to be placed where they are really needed?
Yes, but where do you store the data produced by these small sensors (which in reality is what he's calling "small computers")?
Surely, you won't want to rely on the remote sensors for data storage. What happens to your data when your smart drill-bit is destroyed?
Those of us that do a lot of high-speed, high-volume data collection -- and who are stuck with SCSI -- want to know: Where are the 15000 RPM SATA hard drives? 10K is too slow.
So, by filing this suit, Microsoft admits that someone stole the source code. If that's the case, how can they ever guarantee that the same thief(s) didn't also plant code in their repository as well?
Have you ever wanted to ask the people behind the Vista UI exactly what they were thinking when they did things like Flip 3D or the windows that turn black when maximized?
Remember Computer science isnt about the plumbing...
This is true only when Computer Science doesn't involve software running on computer hardware. When Computer Science does involve software running on computer hardware, only the foolish fail to remember that it is, in fact, all about the plumbing.
It doesn't matter how abstracted the S/W is from the H/W -- at the end of the day there's memory, busses, clocks, devices, (in short: H/W resources) beneath the S/W. Everyone should try to remember that when they code. That is, unless they're writing software that doesn't run on computer hardware. No, you don't need to know everything about low-level arcana, but the more about it you do know, the more informed you will be when you write code. And you'll write better code for real-world purposes.
...so I can have my phone and MP3 player with me in one package.
Um..., why?
When I'm strapped in to the iPod I have the perfect excuse for not answering the phone -- couldn't hear it ringing. If your player is your phone, you lose that excuse.
Compaired to who? Last I checked Lockheed makes the best radar systems in the world. Last I checked Lockheed makes the ONLY anti-ballistic missle defense systems in the world, not just land based by sea based.
Big deal. Given that there are only five "big" defense contractors left, it doesn't mean much when you say that any one of 'em are the only ones that make anything.
Besides, we all know that big programs such as these are never awarded based on their technical merit. How could they be? The gov't folks that evaluate the proposals are as dumb as rocks. It's all about shmoozing. And Lockheed has a lot of damn good shmoozers on staff.
Somethings missing here...
on
Bob Saget 2.0
·
· Score: 1
a kid doing a cannonball onto his dad's groin, a baby running into a church pew, a dog peeing on a wedding dress, and a kid clocking his dad in the nuts with a helmet.
Slashdot Mirror
One character == one byte, dammit!
Don't let them change that.
Close to reference to metal ( eg, metal& )?
What kind of initiative is that?
Doesn't run Java.
I knew it was a bad idea to let the unwashed masses start using computers.
I've never understood the notion of sacrificing the ability to use facilities provided by the operating system in the name of portability. I'm sure the same kinds of situations exist under Windows.
WTF, Java designers? Maybe I just want to write some software, and I don't care whether it's portable or not.
Now maybe that it's GPLed, people can finally start changing it so that we can finally start using Java and OS facilities at the same time .
Right, but if you're a business distributing Linux and making money off that and also, say, selling support services, and if any of the OSS you're distributing and making money on happens to infringe on someone else's software patents, then you and your customers can get your asses sued off.
Micro Soft is promising not to sue Suse users' asses off. That's what this is about.
And "lame '--vbr-new -h --preset standard" will save a butt-load of disk (or iPod) space, and will sound great for about 99% of the non-audiophile-nerdboys. Not only that, but it's is how all the MP3s available from Emusic.com are encoded (non-DRM, too!). Highly recommended.
Maybe so, but having the certs is the ONLY way that Linux of any kind would be approved to deploy on certain networks. No certs, no deployment. Pick a different OS platform. Period. As far as having and maintaining these certs go, RHEL is the only game in town.
We've been waiting for years for someone to get a Linux distribution, any Linux distribution, approved for a particular DoD network that we deploy our systems on. We can't do it ourselves because the process is too onerous and expensive for us to take on. Nonetheless, we'd LOVE to deploy our systems on this network using a Linux platform, but until something gets approved we're stuck with Solaris. Solaris-8 on SPARC, that is. It's the only Unix OS approved for this network. We don't make the rules, but if we want to stay in business, we have to follow them.
That RHEL4 is Common Criteria Scheme certified at CAPP/EAL4+ is just the first step toward getting an OS platform certified to deploy on this network. Without these certs, there'd be no way. If Red Hat goes out of business, then there will be networks out there that will never, ever see a Linux distribution.
In fact, we're aware of a particular program that's done a lot of the heavy lifting to get RHEL-4 certified for deploying on this network. The certifying authority is close to making a decesion. Now, I suspect this authority will just say no. Thanks a lot, Larry. You've torpedoed several avenues of innovation.
If Oracle would guarantee that they'll maintain the certs, etc, then maybe there'd be a chance. But I don't trust Oracle.
They say that you can't be a great writer unless and until you've done close readings of great works of literature.
Similarly, it's extremely elightening to do close readings of source code of software projects you consider "great". Pick a "great" (or even not-so-great) software project for which you have access to the source code. Download the source. Then, do a close reading of the source. Understand how the software works, control flow, data organization, class heirarchies, etc., even the build system.
You might decide that in the end that it's crap, but you'll never know until you read it. But it does provide a good method for learning a lot about how software is put together. Many (but not all) open-source projects are very close to doing best practices due to the collective nature of the development.
8. F.ckin' awesome beaches.
Cool. So Dell's finally shipping OSX on their laptops?
Otherwise, what's the point?
... seamless integration with iTunes (not to be confused with ITMS, which is a topic for a different article)
iTunes is a pretty good juke box package, and it provides for uploading to a player (and synchronizing, too) from inside the application. I won't touch any portable media player unless I can manage it (ie upload) via iTunes.
Yes, but where do you store the data produced by these small sensors (which in reality is what he's calling "small computers")?
Surely, you won't want to rely on the remote sensors for data storage. What happens to your data when your smart drill-bit is destroyed?
In the order that I thought of 'em:
- Dominoes, particularly "42"
- Dart games like "501", etc.
- Hackey Sack
- Ultimate Frisbee
- Sunday Afternoon Softball With Plenty of Beer
- Gin Rummy
- Spades
- Kings-on-the-Corner
- Chess
- Hide the sausage
Those of us that do a lot of high-speed, high-volume data collection -- and who are stuck with SCSI -- want to know:
Where are the 15000 RPM SATA hard drives? 10K is too slow.
So, by filing this suit, Microsoft admits that someone stole the source code. If that's the case, how can they ever guarantee that the same thief(s) didn't also plant code in their repository as well?
Data mining? How does having 80 cores improve I/O?
no apostrophe.
My eyes... the goggles do nothing!
No.
This is true only when Computer Science doesn't involve software running on computer hardware. When Computer Science does involve software running on computer hardware, only the foolish fail to remember that it is, in fact, all about the plumbing.
It doesn't matter how abstracted the S/W is from the H/W -- at the end of the day there's memory, busses, clocks, devices, (in short: H/W resources) beneath the S/W. Everyone should try to remember that when they code. That is, unless they're writing software that doesn't run on computer hardware. No, you don't need to know everything about low-level arcana, but the more about it you do know, the more informed you will be when you write code. And you'll write better code for real-world purposes.
I wish the Java wonks would remember that.
You sound like an absolute idiot when you use profanity.
Geeze, when did Slashdot acquire so many prudes?
...so I can have my phone and MP3 player with me in one package.
Um..., why?
When I'm strapped in to the iPod I have the perfect excuse for not answering the phone -- couldn't hear it ringing.
If your player is your phone, you lose that excuse.
Big deal. Given that there are only five "big" defense contractors left, it doesn't mean much when you say that any one of 'em are the only ones that make anything.
Besides, we all know that big programs such as these are never awarded based on their technical merit. How could they be? The gov't folks that evaluate the proposals are as dumb as rocks. It's all about shmoozing. And Lockheed has a lot of damn good shmoozers on staff.
Links?