Having tempting video games, tempting websites, and tempting symbolic equation solvers on the same box is a distraction? You mean that spending 30 minutes figuring out the equation solver to solve an equation that could be done in 3 minutes by hand is a bad thing?
For most services, a simple cut and dry config file is all that is really needed. This has worked for many years, and most of the problem can be solved by having good manual pages for each of the files (see OpenBSD for good man pages).
For the few problems that need a more complex solution, then those can be taken on a case-by-case basis. For example, Sun revamped the whole init process with SMF, replacing many separate files and directories with a few command line tools. Of course, there is compatibility for "legacy" apps, but the rc directories and inetd.conf files are much much smaller, now.
Replacing all the/etc/ files with XML equivalents is not a solution, because UNIX has been around for 30 years, will be around for another 30, and XML is not even a teenager, yet. It is also hellish in vi.
When it comes to/etc, KISS really is the best approach, IMO.
Yes, but it is tangentially on topic, becuase XBox might be all Microsoft has in ten years, in addition to their legacy support base. It's getting to a point where it's hard to charge money for an operating system or an office suite, and the only places that can continue doing so are embedded systems, like gaming consoles (no one really ever sees the OS there--it's a package deal).
It would be pretty retarded for applications makers to ignore UNIX/Linux desktops in the next few years. They can use Qt or Java to ease the pain of moving to a different platform, if they like. I think the snowball has begun a-rollin, and it's all downhill from here.
Innovation happens elswhere. It's been said a thousand times, but Microsoft won't know what business it will need to be in until that business is already two years past them and they adopt it.
Sun and IBM are much bigger risk takers, right now, which is amazing for how big and relatively old these companies are. HP took a big risk too, but...the execution is leaving something to be desired.
If IBM takes off with the theories about IBM-branded Linux desktops, and Sun has a compatible Java Desktop System (Linux/Solaris), both of which are much cheaper than Windows and Office...I mean, really, where does that leave Microsoft in ten years?
The future is Linux/UNIX, OpenOffice.org, and Mozilla/Firefox. The cheaper-is-better lessons of history require it. And, IBM and Sun can back up their software with selling real hardware and services (it's called diversification, which Microsoft really doesn't do).
They way that quote reads, I could just grab any Java app and suddenly be able to read it's data in some.NET app? Across the network? Sigh. This is just going to spur another 500 books about this interoperability mechanism, 500,000,000 developers will start talking about it like its the greatest thing ever, and after two years only 5,000 developers will have figured it out to actually use it. I do not welcome our vague marketing information driving mass adoption before the technology is ready overlords!
The sense I get is that they only speak against companies trying to monetize open source as direct competition against them. For example, Sun openly competes against Red Hat's support business, but I've never heard Sun say anything negative about Debian, Gentoo, Slackware, etc. or FreeBSD, etc.
This is fair, IMO, because business is business when real money gets involved. Sun and Red Hat have been jabbing at each other for quite some time now. It's no different than Sun vs. HP or Sun vs. IBM. They are all after the same base of _paying_ customers, so they need to fling mud at eachother every so often. (No company is innocent in the mud-flinging, not even Red Hat)
There are some nonsense statements in the article like "We are concerned about security on an open standard environment like that." WTF? At the beginning of the article it mentions something about the Agility Alliance "help EDS create 'best of breed' solutions". Who cares about EDS? Really?
Perhaps this whole alliance is about getting cozy government contracts somewhere. It kinds smells like it, at least.
Unfortunately, there's this stigma of "prefab housing" being small, low-quality, one-design "housing for the poor".
It all depends on the finishing contractor. Modular housing is built to the same standards as regular housing, but if you have a crappy contractor come in to do the finish work, you might as well have bought a trailer. Just like regular construction, you have to really stay on top of the process (e.g., visit the site, double check things after important milestones, etc.)
Is their code that flaky, or is it truly their monopolistic tendencies?
Yes, and yes.
Re:Microsoft has finally been forced to innovate
on
IE7 Details Emerge
·
· Score: 1
Innovate, innovate, innovate, innovate, innovate, and innovate.
One thing missing from that sentence is Microsoft. Their only real innovation is a business model, which, hopefully, is merely temporary in the IT industry.
Why is per-socket pricing legitimate, but per-core is not?
IMO, the problem is that the cores vary so widely. Given a certain size peice of silicon, companies can put two, four, eight, etc. cores on it, but still be within a certain size and power consumption envelope. And it is probably arguable that state of the art design can really only squeeze so much performance per mm^2 or per watt. That means current state of the art will be more or less even per socket (for licensing arguments), but not per core.
This might be real-time for alot of research, but when it comes to games anything less then 15 fps is a joke.
It probably wouldn't be a huge challenge for them to up their clock rate by 75% or so to get 15FPS. Imagine if they moved it from FPGA into custom silicon with appropriate chip designer voodoo.
Solaris 8 will run well on SPARCstations (sun4m at least) and makes a great server or X terminal. OpenBSD or NetBSD on the little lunchbox SPARCstations (e.g., IPX or LX) makes a great low-power DSL/cable modem firewall, as well.
Slashdot Mirror
I thought those children went on to build computers in their garages and found successful companies, thus not having time to post on Slashdot
Having tempting video games, tempting websites, and tempting symbolic equation solvers on the same box is a distraction? You mean that spending 30 minutes figuring out the equation solver to solve an equation that could be done in 3 minutes by hand is a bad thing?
For most services, a simple cut and dry config file is all that is really needed. This has worked for many years, and most of the problem can be solved by having good manual pages for each of the files (see OpenBSD for good man pages).
For the few problems that need a more complex solution, then those can be taken on a case-by-case basis. For example, Sun revamped the whole init process with SMF, replacing many separate files and directories with a few command line tools. Of course, there is compatibility for "legacy" apps, but the rc directories and inetd.conf files are much much smaller, now.
Replacing all the
When it comes to
Yes, but it is tangentially on topic, becuase XBox might be all Microsoft has in ten years, in addition to their legacy support base. It's getting to a point where it's hard to charge money for an operating system or an office suite, and the only places that can continue doing so are embedded systems, like gaming consoles (no one really ever sees the OS there--it's a package deal).
It would be pretty retarded for applications makers to ignore UNIX/Linux desktops in the next few years. They can use Qt or Java to ease the pain of moving to a different platform, if they like. I think the snowball has begun a-rollin, and it's all downhill from here.
Innovation happens elswhere. It's been said a thousand times, but Microsoft won't know what business it will need to be in until that business is already two years past them and they adopt it.
Sun and IBM are much bigger risk takers, right now, which is amazing for how big and relatively old these companies are. HP took a big risk too, but...the execution is leaving something to be desired.
If IBM takes off with the theories about IBM-branded Linux desktops, and Sun has a compatible Java Desktop System (Linux/Solaris), both of which are much cheaper than Windows and Office...I mean, really, where does that leave Microsoft in ten years?
The future is Linux/UNIX, OpenOffice.org, and Mozilla/Firefox. The cheaper-is-better lessons of history require it. And, IBM and Sun can back up their software with selling real hardware and services (it's called diversification, which Microsoft really doesn't do).
I can wait until next year and pay $20! Even less! HAHAHAHA!
Soooo, guys...are you trying to say something about EDS? It seems it's right on the tip of your tounge. Come on, I really want to know!
They way that quote reads, I could just grab any Java app and suddenly be able to read it's data in some
Ah, so the disease is the cure! Brilliant!
The sense I get is that they only speak against companies trying to monetize open source as direct competition against them. For example, Sun openly competes against Red Hat's support business, but I've never heard Sun say anything negative about Debian, Gentoo, Slackware, etc. or FreeBSD, etc.
This is fair, IMO, because business is business when real money gets involved. Sun and Red Hat have been jabbing at each other for quite some time now. It's no different than Sun vs. HP or Sun vs. IBM. They are all after the same base of _paying_ customers, so they need to fling mud at eachother every so often. (No company is innocent in the mud-flinging, not even Red Hat)
Well, how some sysadmins I've known follow their registration plan, one registration = 10 servers, so I guess the cosmic balance is still intact.
Funny how so many people disagree with you.
www.sunsource.net
www.opensolaris.org
www.openoffice.org
Sun supports Linux, too
What were you saying?
Basically, I think ZDNet has trolled Slashdot.
There are some nonsense statements in the article like "We are concerned about security on an open standard environment like that." WTF? At the beginning of the article it mentions something about the Agility Alliance "help EDS create 'best of breed' solutions". Who cares about EDS? Really?
Perhaps this whole alliance is about getting cozy government contracts somewhere. It kinds smells like it, at least.
Unfortunately, there's this stigma of "prefab housing" being small, low-quality, one-design "housing for the poor".
It all depends on the finishing contractor. Modular housing is built to the same standards as regular housing, but if you have a crappy contractor come in to do the finish work, you might as well have bought a trailer. Just like regular construction, you have to really stay on top of the process (e.g., visit the site, double check things after important milestones, etc.)
Is their code that flaky, or is it truly their monopolistic tendencies?
Yes, and yes.
Innovate, innovate, innovate, innovate, innovate, and innovate.
One thing missing from that sentence is Microsoft. Their only real innovation is a business model, which, hopefully, is merely temporary in the IT industry.
Microsoft and Slashdot are both chronic infections. One gets in your holes, and one overwhelms you with trolls.
If kids can't even figure out how to install Final Fantasy XI, how can they be pirates, too?
Why is per-socket pricing legitimate, but per-core is not?
IMO, the problem is that the cores vary so widely. Given a certain size peice of silicon, companies can put two, four, eight, etc. cores on it, but still be within a certain size and power consumption envelope. And it is probably arguable that state of the art design can really only squeeze so much performance per mm^2 or per watt. That means current state of the art will be more or less even per socket (for licensing arguments), but not per core.
This might be real-time for alot of research, but when it comes to games anything less then 15 fps is a joke.
It probably wouldn't be a huge challenge for them to up their clock rate by 75% or so to get 15FPS. Imagine if they moved it from FPGA into custom silicon with appropriate chip designer voodoo.
They use the same dark magic that they used to make the Internet go faster.
Solaris 8 will run well on SPARCstations (sun4m at least) and makes a great server or X terminal. OpenBSD or NetBSD on the little lunchbox SPARCstations (e.g., IPX or LX) makes a great low-power DSL/cable modem firewall, as well.
Actually, Linux shines in the 1-way to 8-way space, where, coincidentally, Opteron shines as well.
the problem is going to be software availability.
Nope. Niagara is SPARC and will run Solaris. Just like any other Sun server.