The breakup of AT&T, which I personally believe was lobbied-for by IBM, so scared the shit out of everybody in the industry that the gov covered its ass and soon dropped the IBM anti-trust action.
Take the comment above a little different way, and allow content to be 'unrated', which is the default today, and don't criminalize unrated content. Then let browsers default to child-rated content for the kids, and allow adults to select for 'mature' and 'unrated' but not 'smut'. Or, smut-only, late on Fridays. You get the idea. Flexibility.
Should sites that that emit XXX content with child-acceptable ratings be prosecutable? One would think it would be easier to draw up such standards, but each locality has its own standards. But any such efforts to enforce ratings has to deal with the vast body of information already out there, so sensibly dealing with unrated content is necessary.
I have heard that XFS and IRIX paging/scheduling are "Siamese Twins Connected at the Bung", as my late father would say. This does not make any porting easy, and many design decisions need to be made as to what should change to keep the functionality and the performance goals.
Maybe you wouldn't mind being forced to wear a badge at all times that stated your religion. After all, this just empowers people to know something of your biases and culture. I know I'd fight such a measure with all at my disposal, and I don't think this plan is far different.
Perhaps an appropriate symbol for stating opposition to this rating proposal would be the "yellow star-of-David", especially considering the site of this conference (Munich). I am recalling the example of the Danes, when the occupying Nazis made the edict that all Jews must wear the stars, *all* the Danes put one on. (Good story, anyway, and I hope it is true.)
The best thing that happened to IRIX was for the Cray OS QA team to get 'hold of it and whack the bugs out of it. Those folks know that the customer has no sense of humor about crashes losing the last two months of calculations, or whatever. The stability improvement between 6.2 and 6.5 is enormous, and the performance jumped, as well. Yup, there have been five point releases of 6.5 over the past year, but hey! that looks good when it is the Linux 2.2 kernel, but not here?
I personally view the O2 machine as a big mistake by SGI, and it mars the company's reputation. While the O2000 multi-processor boxes really rock, and the O200 machines support big disk and RAM and show quite decent I/O for the apps I run, I am just way unimpressed by the O2. For the work I do, and as in all this discussion, that is a *key* distinction, it is an overpriced, underpowered PC-wannabe. And, it came with the incredibly buggy IRIX 6.3, the version that was incompatible with most everything else IRIX. (Yes, 6.3, the OS that third-party developers love to hate. 25-40 meg patch files? That's an OS replacement!) To me, the O2's niche must be a very small one. Your mileage may vary.
I've seen at least three types of applications for SGI hardware discussed on this thread: Heavy Graphics (the SGI myth author), sys admins for web servers, and data transfer/communications (what I do). The graphics guy is happy with his high-end SGI machines. The web server sysadmin is unhappy with the performance and stability of SGI in an area SGI used to target. I have a mixed opinion, liking the higher-end machines, but not the O2 for data transfer/database/communications.
Thus endeth my attempts to influence direction on this thread.:-)
The newsflash is that the problems with AI are solved: it takes ~18 years to train one, and they get really cranky between 14 and 17 years old.
reboots mean do deterioration "entropy" testing
on
CrackThisBox Updates
·
· Score: 1
The several reboots mean that no "entropy" testing is really happening here -- small leaks, etc. that add up to something special after several days of processing.
It would be nice to see the recent Mindcraft benchmarks run against *BSD. The MS guys wouldn't have to change a thing -- just bring in the *BSD experts and have them tune-up their machines and go....
Like newsgroups, we *could* treat each other on/. like 'family' and flame the hell out of each other, while still putting on a more civilized face for the 'outside world'. This does run the risk of allowing just anybody to troll, or otherwise kick the/. ant pile just for fun. You are correct in making the distinction between a/. posting and direct, personal, email.
Besides, real flaming is an art that has little to do with vulgarity and obscenity. Look back at, say, Maddie Hauseman flaming newbies in alt.fan.dan-quayle for great examples.
Well, for what I have heard, the optimization on Apache largely stopped when a modest machine could completely stuff a T1-capacity line. That is a "practical" goal, which, of course, is not related to benchmarking in general. This isn't a slam!
My experience with this was with CAD/CAM board lay-out and HW modeling software. At least it was a well-known, understood-before-buying requirement that the software would only work for so many users for so many weeks. Then we had to call the vendor to get new codes and bludgeon our Accouting Dept. to promise them money. So this provision doesn't bother me as much as the "no reverse engineering" clause.
I'm not trying to be facetious in this response, but I don't see the words of Claim 20 describing something much different than how USENET news and NNTP senders/recipients communicate, especially when IHAVE/SENDME is employed. There is also the variation of the protocol where the consumer (downstream sites) communicate back up just which newsgroups have been read, and so tell the provider (upstream site) which newsgroups are needed. The objects are less granular than those described by Intermind, but Intermind is only proposing a variation on the theme.
Sprinkling a description of the above system with the words 'metadata' and using the word 'said' liberally achieves something awfully close to the verbage used by Intermind.
In my mind, the PTO was wowed by the application to "Channels" and the Web and the concept of "object".
StarOffice has folded underneath me, but I have never lost data because of it. I am looking forward to the new "filter pack" release that has increased ability to read MS '97 documents. Of course, the '98 versions and '00 versions of Office will do things in new, incompatible ways for no apparent benefit....
As for actual use, the PowerPoint part of StarOffice is REALLY close to the MS software. I also tried to use Word to create a document with several images downloaded from the WWW. Word SUCKED at placing those images in a document. StarOffice "just worked" as one would expect.
I'm thinking that I'd rather pay $40 to OSS for a Creative sound driver, and donate $50 to XFree86 for a graphics driver for a Voodoo card. Both organizations have provided the framework for the implementation of other vendors' hardware drivers.
For myself, it will definitely affect a future buy decision if I can look at for example, Gateway or Dell's component list for a system, and determine either directly or by going to the component vendors' web pages, that drivers are available for Linux. It would affect my decision even more directly if the computer vendor provided drivers on their "System Restoration" CD.
Slashdot Mirror
The breakup of AT&T, which I personally believe was lobbied-for by IBM, so scared the shit out of everybody in the industry that the gov covered its ass and soon dropped the IBM anti-trust action.
Brilliant tactics by IBM, IMHO.
Take the comment above a little different way, and allow content to be 'unrated', which is the default today, and don't criminalize unrated content. Then let browsers default to child-rated content for the kids, and allow adults to select for 'mature' and 'unrated' but not 'smut'. Or, smut-only, late on Fridays. You get the idea. Flexibility.
Should sites that that emit XXX content with child-acceptable ratings be prosecutable? One would think it would be easier to draw up such standards, but each locality has its own standards. But any such efforts to enforce ratings has to deal with the vast body of information already out there, so sensibly dealing with unrated content is necessary.
I have heard that XFS and IRIX paging/scheduling are "Siamese Twins Connected at the Bung", as my late father would say. This does not make any porting easy, and many design decisions need to be made as to what should change to keep the functionality and the performance goals.
Perhaps an appropriate symbol for stating opposition to this rating proposal would be the "yellow star-of-David", especially considering the site of this conference (Munich). I am recalling the example of the Danes, when the occupying Nazis made the edict that all Jews must wear the stars, *all* the Danes put one on. (Good story, anyway, and I hope it is true.)
The best thing that happened to IRIX was for the Cray OS QA team to get 'hold of it and whack the bugs out of it. Those folks know that the customer has no sense of humor about crashes losing the last two months of calculations, or whatever. The stability improvement between 6.2 and 6.5 is enormous, and the performance jumped, as well. Yup, there have been five point releases of 6.5 over the past year, but hey! that looks good when it is the Linux 2.2 kernel, but not here?
:-)
I personally view the O2 machine as a big mistake by SGI, and it mars the company's reputation. While the O2000 multi-processor boxes really rock, and the O200 machines support big disk and RAM and show quite decent I/O for the apps I run, I am just way unimpressed by the O2. For the work I do, and as in all this discussion, that is a *key* distinction, it is an overpriced, underpowered PC-wannabe. And, it came with the incredibly buggy IRIX 6.3, the version that was incompatible with most everything else IRIX. (Yes, 6.3, the OS that third-party developers love to hate. 25-40 meg patch files? That's an OS replacement!) To me, the O2's niche must be a very small one. Your mileage may vary.
I've seen at least three types of applications for SGI hardware discussed on this thread: Heavy Graphics (the SGI myth author), sys admins for web servers, and data transfer/communications (what I do). The graphics guy is happy with his high-end SGI machines. The web server sysadmin is unhappy with the performance and stability of SGI in an area SGI used to target. I have a mixed opinion, liking the higher-end machines, but not the O2 for data transfer/database/communications.
Thus endeth my attempts to influence direction on this thread.
The newsflash is that the problems with AI are
solved: it takes ~18 years to train one, and they get really cranky between 14 and 17 years old.
The several reboots mean that no "entropy" testing is really happening here -- small leaks, etc. that add up to something special after several days of processing.
Long as everyone knows that....
It would be nice to see the recent Mindcraft benchmarks run against *BSD. The MS guys wouldn't have to change a thing -- just bring in the *BSD experts and have them tune-up their machines and go....
Like newsgroups, we *could* treat each other on /. like 'family' and flame the hell out of each other, while still putting on a more civilized face for the 'outside world'. This does run the risk of allowing just anybody to troll, or otherwise kick the /. ant pile just for fun. You are correct in making the distinction between a /. posting and direct, personal, email.
Besides, real flaming is an art that has little to do with vulgarity and obscenity. Look back at, say, Maddie Hauseman flaming newbies in alt.fan.dan-quayle for great examples.
Well, for what I have heard, the optimization on Apache largely stopped when a modest machine could completely stuff a T1-capacity line. That is a "practical" goal, which, of course, is not related to benchmarking in general. This isn't a slam!
My experience with this was with CAD/CAM board lay-out and HW modeling software. At least it was a well-known, understood-before-buying requirement that the software would only work for so many users for so many weeks. Then we had to call the vendor to get new codes and bludgeon our Accouting Dept. to promise them money. So this provision doesn't bother me as much as the "no reverse engineering" clause.
I'm not trying to be facetious in this response, but I don't see
the words of Claim 20 describing something much different than
how USENET news and NNTP senders/recipients communicate, especially
when IHAVE/SENDME is employed. There is also the variation of the
protocol where the consumer (downstream sites) communicate back
up just which newsgroups have been read, and so tell the provider
(upstream site) which newsgroups are needed. The objects are less
granular than those described by Intermind, but Intermind is only
proposing a variation on the theme.
Sprinkling a description of the above system with the words 'metadata'
and using the word 'said' liberally achieves something awfully close
to the verbage used by Intermind.
In my mind, the PTO was wowed by the application to "Channels" and
the Web and the concept of "object".
... But we can send such email if the intent is to entertain?
StarOffice has folded underneath me, but I have
never lost data because of it. I am looking forward to the new "filter pack" release that has increased ability to read MS '97 documents. Of course, the '98 versions and '00 versions of Office will do things in new, incompatible ways for no apparent benefit....
As for actual use, the PowerPoint part of StarOffice is REALLY close to the MS software. I also tried to use Word to create a document with several images downloaded from the WWW. Word SUCKED at placing those images in a document. StarOffice "just worked" as one would expect.
I'm thinking that I'd rather pay $40 to OSS for a Creative sound driver, and donate $50 to XFree86 for a graphics driver for a Voodoo card. Both organizations have provided the framework for the implementation of other vendors' hardware drivers.
For myself, it will definitely affect a future buy decision if I can look at for example, Gateway or Dell's component list for a system, and determine either directly or by going to the component vendors' web pages, that drivers are available for Linux. It would affect my decision even more directly if the computer vendor provided drivers on their "System Restoration" CD.