For years practitioners used the Web and its language, HTML, as a free format and layout platform for forms and documents (often as paint for software applications).
Are you forgetting the NS4/IE incompatibilities and proprietary tags that plague the web to this day?
The fact is, when W3C came along, the web was seriously broken. Those of us that are so outspoken on standards are generally those that have been on the web long enough to remember how broken it was when non-standard HTML ruled the day.
Re:Will Firefox make it to the systems as default?
on
IBM Backs Firefox In-House
·
· Score: 2, Informative
I love Firefox, but this approach will get you nowhere.
*NO* web browser in existence is 100% compatible W3C standards. Some are more compliant than others, but none of them are fully compliant, and Firefox has its quirks just like any other.
I wouldn't be as worried about the best hackers in the world hitting my box as the best hackers in the world writing an app that every script kiddie in the world could use to attack my box.
As another poster mentioned, SoundStorm is the only chipset that does on the fly encoding, so there's quite a bit of difference there.
If you think the Audigy 2's are true 24-bit 192Khz, you're the one that needs to do their fact checking here. It's not much more than a marketing scheme on Creative's part, as data is only passed as 24-bit through a few pipelines on the card - the vast majority of it is downsampled. You can find plenty on this issue through Google if you'd like. They're good cards, sure, but it sounds as if you've fallen for the hype to me. Granted I own one, but primarily because I managed to pick it up for half the normal *wholesale* price of the time.
I'm not sure what you mean by you can't replace it. That's what PCI slots are for. Sure you can't rip it off the mainboard, but as plenty of other posts have pointed out, it'd be more expensive to disable functionality than just leave it there.
I think the 'avoid onboard video' thing is a bit silly. When NForce2 was new, its onboard video was perfectly acceptable for a low-midrange system. Hell, for anyone that doesn't do much gaming, it's still fine.
Add to that the fact that ATI's RS480 chipset includes X300 equivalent graphics, and it seems pretty silly to say 'onboard video sucks.' The X300's not a screamer by any means, but onboard video that'll run Far Cry at an acceptable framerate is fine by me.
Tell that to anyone with an NForce2 and SoundStorm audio.
Quite frankly, there are people who go out of their way for it as it's the ONLY way to get Dolby Digital encoding on a PC. Just try to tell them that it sounds like crap. You need to update your views on onboard audio.
Agreed on the P6 core being Intel's best work. When Netburst was first introduced, Intel tried to reassure everyone with promises of a heady 5-6GHz within a few years and eventually numbers as high as 8GHz. Well, we all know what happened now - Netburst never broke 4GHz.
On the SMT issue, it's the deep pipeline nature of Netburst that makes it as effective as it is on the P4 in the first place. Netburst was launched with a 20 instruction deep pipeline, Prescott extended that to 31 stages. For comparison, P6 was 14 levels, K7 was 15 levels, and K8 is 17 levels. This means that relative to other chips available at the time, Netburst takes a seriously penalty for a branch mispredict, and the entire pipeline has to be flushed out. SMT helps it make up for this by allowing one thread to continue running while the pipeline is being flushed.
While K8 or the Pentium M would probably benefit some from SMT, it wouldn't be to the level that the Pentium 4 does. Therefore, Intel and AMD are both focusing on other ways of improving performance in these processor designs that should yield better results for the resources invested.
The thing is, 1GHz wasn't much more than a speed boost. The fundamental chips were still the same cores that Intel and AMD were already shipping. Here we've got a major reworking of how the chip works that was rushed, which is a bit worse, IMO. Let's also not forget that while 1.0GHz was no big deal for either Intel or AMD, Intel had major problems with the 1.13GHz PIIIs. Go figure.:-P
Sun's also been shipping 64 bit systems for years. IBM's been shipping multicore POWER servers for several years too.
I meant bragging rights in the AMD/Intel rivalry. While Sun may technically be a competitor, it's in a very different way. This is head-to-head competition between two companies in the mainstream CPU market. Most consumers are barely aware AMD exists, let alone that Sun makes something besides Java.
Intel actually shipped dual core first (by a couple of days) to pick up bragging rights, but as you said, it was rushed and we're seeing the results of that now.
Intel's problem is that they're tied to the inefficient Netburst architechture for the time being, and it's really just hit a complete brick wall. The move back the the P6 architechture of the Pentium M is in the works, but that's going to take time.
In the meantime, AMD has the K8 which absolutely brutalizes Netburst in performance per clockcycle, and which, at this point, also seems to have plenty of headroom left.
The biggest problem Intel has is that, in the processor business, you can only be so agile. Yes, they've hacked in x86-64 and dual core support to Netburst, but their implementations are just that - hacks and bandaids. The result is what we're seeing now - AMD with a highly efficient and what's looking to be scalable (at this point) architechture, while Intel is limping along with Netburst that's really been on its last legs since Prescott was introduced 14 months ago.
The Pentium M shows some promise, and is probably Intel's best processor design to date, but has problems of its own (ie: its relatively weak FPU performance compared to K8 and Netburst both). It's entirely possible that Intel will come up with something in the long run that will straighten them out, but as it stands now, I wouldn't expect it in any time frame less than 18 months from now. The first dual core Xeons will likely be Prescott-based, and suffer all the same fundamental flaws Intel has been fighting for the past few years. By that time, AMD may have already had a chance to entrench itself in data centers, which would be a huge loss for Intel. That's why this is big news.
Re:Good news, even for Sid users.
on
Sarge is Now Frozen
·
· Score: 3, Insightful
No, I understand perfectly what Debian stable is intended for. I just feel it's taken to an unreasonable extreme.
You know why using testing or unstable isn't an acceptable answer? Because when things break you get the copout excuse that 'things change and break' because it's a development branch.
Having a stable distribution does *NOT* mean that you have to run old, borederline obsolete software. It means that once a distribution is released, the software within that release remains stable, and the only real changes it's going receive are security patches. This simply means you don't go pulling the carpet out from under things by making changes like major version upgrades in that release. Which is fine, but it does not mean that you need to go three *years* between releases, and then launching with older software than any other major distribution is currently using.
So the short version of this is that being "STABLE," as you put it in the manner that matters for a server simply means that you have an unchanging baseline to work from. The problem comes when you have multi-year stretches between those baseline points. That's the flaw with the current Debian manner of operation, and the reason people like myself avoid it even though we see some great strongpoints to the distribution.
BTW thanks for making my point exactly about Debian zealots with the "STFU you don't understand how it works, now go run unstable" response. I never have understood what it is about Debian that draws the elitists.
Re:Good news, even for Sid users.
on
Sarge is Now Frozen
·
· Score: 5, Insightful
Congratulations, you've just hit on exactly why most people will never even try Debian, which is a shame. Because when legitimate complaints are raised, the Debian zealots get defensive and give reasons that would excuse the distribution for being somewhat behind the curve, but not in the state it's in.
If they really are releasing Sarge without X.org, this is a perfect example of just this mentality. Debian is the *only* major distribution that hasn't made the switch. Even Slackware's done so, even with that stretch where Pat was absent from development. Let's face it, X.org has been the de facto standard for months now, and the initial release was little more than a fork of the last XFree release before the license change.
So what exactly is the reason for completely ignoring a changing landscape every other distribution has accepted? I just see it as the 'nothing changes' mentality taken to an absurdist extreme. Yes, it makes sense for a stable distribution that's already released, but putting out a new stable branch that, if given Debian's current record, will end up being the standard bearer for a few *years* with major system components that have been replaced for months at launch? Beats me.
The fact of the matter is, like it or not, this is the perception of Debian that's been gained by a good segment of the Linux using populace, and its users aren't helping with comments like "I consider it a great blessing that I am granted access to their *testing* and *unstable* branch."
Which is a shame because from a technical standpoint, Debian has quite possibly the best underpinnings of any of the major distributions. apt is an elegant solution to the package management problem. Debian's configuration tools are generally top notch. Even the new installer's not half bad. But as a whole it's held up by an overly slow development cycle and an elitist attitude amongst its users.
*That's* the reason you see people migrating to Ubuntu en masse. It's all the technical goodness of Debian on a sane development schedule and with a friendly user base. It's what I've wished Debian could be for years, but never seemed to have any interest in becoming. Ah well, C'est La Vie. It just shows what's possible when development becomes focused on getting things done rather than allowing them to languish.
I think the most interesting thing about this is how similar the original page is to today's Google. Guess it goes to show they really hit on the right thing pretty early on as far as the site's interface goes.
So you now have to constantly be on the watchout for security fixes and patches to functionality you might need? The whole point of a system-wide update on any distribution is to alleviate some of the pain in that task from the user.
I haven't used Gentoo in a couple of years, but it's behavior like this that gives it it and its users the reputation they have.
One of the applications of Gentoo that the compile time issue can be somewhat negated in is with rollouts of similar systems, doing the optimized compilation on a single unit.
So why in the world are you wasting 2 days a unit on install? Create a base installation, tar it up, and then copy it to the new machines, tweaking what you need along the way. While optimizing for your machine may make some sense, a C3 is a C3, and wasting literally days duplicating effort is just silly, and certainly doesn't help promote the distribution.
Again, I don't have a problem with Gentoo, I've just moved onto other things - a focus on out of the box usability grabbed me, leading me to Dropline Gnome and Ubuntu since then. But when Gentoo's users do things that are entirely counterproductive when there's a perfectly good way of doing them *within Gentoo's framework*, you start to see where the reputation comes from.
Reread my post. I'm fully aware that AMD was designed for multicore from the start. It's a *much* cleaner design than Intel's. That said, it's *still* not much more than two cores on a single die. AMD and Intel both are on the record saying that a shared cache design is really a better implementation, however it's more complex than either have worked out at this point.
I'll definitely agree with you that AMD's design is the better of the two. The interconnect circuitry on the K8 has been there since it was released in Sept 2003, Intel's is an 11th hour hack. I'm not questioning that.
My whole point was that the dual cores we're seeing now are stopgap measures to be first to market with dual core. Both companies are planning to go to a cleaner shared cache design at a later time, but as I said in my original post, that's still 12-18 months away.
Didn't mean to sound quite so harsh, but it was early and I wasn't fully awake.:-)
The Opterons that are just hitting the market ARE a rushed design, however. Yes, they work, yes they're dual core, and yes they were designed for it from the start. The key, however, is that the chips you can buy now have a seperate cache for each core. The 'elegant' designs that aren't anywhere near ready at this time use a shared cache that's accessed by both cores. That's what I meant by 'rushed' design.
Apple holds quite a few patents on UI elements found in OS X. I think the posted screenshots thing is BS however. Don't inventors have a year after publication to file?
AMD and Intel both rushed dual core to market for bragging rights. Both companies are using a design that's really not much more than two dies on the same wafer with a little interconnect circuitry. I think the *most* optimistic estimate I've heard for a clean, shared cache design is 12-18 months away still. This is very new stuff for both companies.
IBM, on the other hand, has been building dual core for several *years* now with the POWER series now. And not just single core - we're talking eight cores on the same wafer last I'd heard.
So no, the PPC970 hasn't received dual core yet, but claiming that IBM 'can't keep up' from a technological standpoint is absolutely ridiculous, and suggests that you don't really know what you're talking about.
Correction: Most users who are set in their ways and refuse to try anything new hate it.
It's just like the 'geeks' who set every XP box they touch to the classic style start menu simply because it's not what they're used to. It doesn't matter what might be better, it's different and therefore they hate it.
I'd highly recommend you read this article at Ars Technica regarding the Finder and spatialness. It's more than up to Ars' usual high standard, and should give a better idea of what a spatial interface is, and why it can be a good idea, if implemented right.
That's because the Windows 95 approach to being spacial wasn't very good. On the other hand, MacOS = 9 used a spacial finder, and its absence in OS X is a common complaint amongst the old school Mac crowd. Just because the one implementation you're experienced with sucked doesn't mean the whole concept of a spatial filebrowser is bad.
To clarify: it's the Rebel XT. I have the Rebel and the startup time is somewhere in the 3-5 second range.
I do agree on the high speed flash cards (Sandisk Ultra II's are usually easiest to find, if pricy), but as far as write time goes, you're a world better off than I am. The Rebel has a 4 shot buffer vs the 14 shot buffer on the XT. It's one of the things that almost makes me want to upgrade, but I think I'll wait another generation and put the money into some better lenses in the meantime.:-)
Slashdot Mirror
Are you forgetting the NS4/IE incompatibilities and proprietary tags that plague the web to this day?
The fact is, when W3C came along, the web was seriously broken. Those of us that are so outspoken on standards are generally those that have been on the web long enough to remember how broken it was when non-standard HTML ruled the day.
I love Firefox, but this approach will get you nowhere.
*NO* web browser in existence is 100% compatible W3C standards. Some are more compliant than others, but none of them are fully compliant, and Firefox has its quirks just like any other.
I wouldn't be as worried about the best hackers in the world hitting my box as the best hackers in the world writing an app that every script kiddie in the world could use to attack my box.
As another poster mentioned, SoundStorm is the only chipset that does on the fly encoding, so there's quite a bit of difference there.
If you think the Audigy 2's are true 24-bit 192Khz, you're the one that needs to do their fact checking here. It's not much more than a marketing scheme on Creative's part, as data is only passed as 24-bit through a few pipelines on the card - the vast majority of it is downsampled. You can find plenty on this issue through Google if you'd like. They're good cards, sure, but it sounds as if you've fallen for the hype to me. Granted I own one, but primarily because I managed to pick it up for half the normal *wholesale* price of the time.
I'm not sure what you mean by you can't replace it. That's what PCI slots are for. Sure you can't rip it off the mainboard, but as plenty of other posts have pointed out, it'd be more expensive to disable functionality than just leave it there.
I think the 'avoid onboard video' thing is a bit silly. When NForce2 was new, its onboard video was perfectly acceptable for a low-midrange system. Hell, for anyone that doesn't do much gaming, it's still fine.
Add to that the fact that ATI's RS480 chipset includes X300 equivalent graphics, and it seems pretty silly to say 'onboard video sucks.' The X300's not a screamer by any means, but onboard video that'll run Far Cry at an acceptable framerate is fine by me.
Tell that to anyone with an NForce2 and SoundStorm audio.
Quite frankly, there are people who go out of their way for it as it's the ONLY way to get Dolby Digital encoding on a PC. Just try to tell them that it sounds like crap. You need to update your views on onboard audio.
Agreed on the P6 core being Intel's best work. When Netburst was first introduced, Intel tried to reassure everyone with promises of a heady 5-6GHz within a few years and eventually numbers as high as 8GHz. Well, we all know what happened now - Netburst never broke 4GHz.
On the SMT issue, it's the deep pipeline nature of Netburst that makes it as effective as it is on the P4 in the first place. Netburst was launched with a 20 instruction deep pipeline, Prescott extended that to 31 stages. For comparison, P6 was 14 levels, K7 was 15 levels, and K8 is 17 levels. This means that relative to other chips available at the time, Netburst takes a seriously penalty for a branch mispredict, and the entire pipeline has to be flushed out. SMT helps it make up for this by allowing one thread to continue running while the pipeline is being flushed.
While K8 or the Pentium M would probably benefit some from SMT, it wouldn't be to the level that the Pentium 4 does. Therefore, Intel and AMD are both focusing on other ways of improving performance in these processor designs that should yield better results for the resources invested.
The thing is, 1GHz wasn't much more than a speed boost. The fundamental chips were still the same cores that Intel and AMD were already shipping. Here we've got a major reworking of how the chip works that was rushed, which is a bit worse, IMO. Let's also not forget that while 1.0GHz was no big deal for either Intel or AMD, Intel had major problems with the 1.13GHz PIIIs. Go figure. :-P
Sun's also been shipping 64 bit systems for years. IBM's been shipping multicore POWER servers for several years too.
I meant bragging rights in the AMD/Intel rivalry. While Sun may technically be a competitor, it's in a very different way. This is head-to-head competition between two companies in the mainstream CPU market. Most consumers are barely aware AMD exists, let alone that Sun makes something besides Java.
Intel actually shipped dual core first (by a couple of days) to pick up bragging rights, but as you said, it was rushed and we're seeing the results of that now.
Intel's problem is that they're tied to the inefficient Netburst architechture for the time being, and it's really just hit a complete brick wall. The move back the the P6 architechture of the Pentium M is in the works, but that's going to take time.
In the meantime, AMD has the K8 which absolutely brutalizes Netburst in performance per clockcycle, and which, at this point, also seems to have plenty of headroom left.
The biggest problem Intel has is that, in the processor business, you can only be so agile. Yes, they've hacked in x86-64 and dual core support to Netburst, but their implementations are just that - hacks and bandaids. The result is what we're seeing now - AMD with a highly efficient and what's looking to be scalable (at this point) architechture, while Intel is limping along with Netburst that's really been on its last legs since Prescott was introduced 14 months ago.
The Pentium M shows some promise, and is probably Intel's best processor design to date, but has problems of its own (ie: its relatively weak FPU performance compared to K8 and Netburst both). It's entirely possible that Intel will come up with something in the long run that will straighten them out, but as it stands now, I wouldn't expect it in any time frame less than 18 months from now. The first dual core Xeons will likely be Prescott-based, and suffer all the same fundamental flaws Intel has been fighting for the past few years. By that time, AMD may have already had a chance to entrench itself in data centers, which would be a huge loss for Intel. That's why this is big news.
No, I understand perfectly what Debian stable is intended for. I just feel it's taken to an unreasonable extreme.
You know why using testing or unstable isn't an acceptable answer? Because when things break you get the copout excuse that 'things change and break' because it's a development branch.
Having a stable distribution does *NOT* mean that you have to run old, borederline obsolete software. It means that once a distribution is released, the software within that release remains stable, and the only real changes it's going receive are security patches. This simply means you don't go pulling the carpet out from under things by making changes like major version upgrades in that release. Which is fine, but it does not mean that you need to go three *years* between releases, and then launching with older software than any other major distribution is currently using.
So the short version of this is that being "STABLE," as you put it in the manner that matters for a server simply means that you have an unchanging baseline to work from. The problem comes when you have multi-year stretches between those baseline points. That's the flaw with the current Debian manner of operation, and the reason people like myself avoid it even though we see some great strongpoints to the distribution.
BTW thanks for making my point exactly about Debian zealots with the "STFU you don't understand how it works, now go run unstable" response. I never have understood what it is about Debian that draws the elitists.
Congratulations, you've just hit on exactly why most people will never even try Debian, which is a shame. Because when legitimate complaints are raised, the Debian zealots get defensive and give reasons that would excuse the distribution for being somewhat behind the curve, but not in the state it's in.
If they really are releasing Sarge without X.org, this is a perfect example of just this mentality. Debian is the *only* major distribution that hasn't made the switch. Even Slackware's done so, even with that stretch where Pat was absent from development. Let's face it, X.org has been the de facto standard for months now, and the initial release was little more than a fork of the last XFree release before the license change.
So what exactly is the reason for completely ignoring a changing landscape every other distribution has accepted? I just see it as the 'nothing changes' mentality taken to an absurdist extreme. Yes, it makes sense for a stable distribution that's already released, but putting out a new stable branch that, if given Debian's current record, will end up being the standard bearer for a few *years* with major system components that have been replaced for months at launch? Beats me.
The fact of the matter is, like it or not, this is the perception of Debian that's been gained by a good segment of the Linux using populace, and its users aren't helping with comments like "I consider it a great blessing that I am granted access to their *testing* and *unstable* branch."
Which is a shame because from a technical standpoint, Debian has quite possibly the best underpinnings of any of the major distributions. apt is an elegant solution to the package management problem. Debian's configuration tools are generally top notch. Even the new installer's not half bad. But as a whole it's held up by an overly slow development cycle and an elitist attitude amongst its users.
*That's* the reason you see people migrating to Ubuntu en masse. It's all the technical goodness of Debian on a sane development schedule and with a friendly user base. It's what I've wished Debian could be for years, but never seemed to have any interest in becoming. Ah well, C'est La Vie. It just shows what's possible when development becomes focused on getting things done rather than allowing them to languish.
If the Escort is your only transportation in the meantime, absolutely. That's the situation we're in with Hubble.
I think the most interesting thing about this is how similar the original page is to today's Google. Guess it goes to show they really hit on the right thing pretty early on as far as the site's interface goes.
So you now have to constantly be on the watchout for security fixes and patches to functionality you might need? The whole point of a system-wide update on any distribution is to alleviate some of the pain in that task from the user.
I haven't used Gentoo in a couple of years, but it's behavior like this that gives it it and its users the reputation they have.
One of the applications of Gentoo that the compile time issue can be somewhat negated in is with rollouts of similar systems, doing the optimized compilation on a single unit.
So why in the world are you wasting 2 days a unit on install? Create a base installation, tar it up, and then copy it to the new machines, tweaking what you need along the way. While optimizing for your machine may make some sense, a C3 is a C3, and wasting literally days duplicating effort is just silly, and certainly doesn't help promote the distribution.
Again, I don't have a problem with Gentoo, I've just moved onto other things - a focus on out of the box usability grabbed me, leading me to Dropline Gnome and Ubuntu since then. But when Gentoo's users do things that are entirely counterproductive when there's a perfectly good way of doing them *within Gentoo's framework*, you start to see where the reputation comes from.
Reread my post. I'm fully aware that AMD was designed for multicore from the start. It's a *much* cleaner design than Intel's. That said, it's *still* not much more than two cores on a single die. AMD and Intel both are on the record saying that a shared cache design is really a better implementation, however it's more complex than either have worked out at this point.
I'll definitely agree with you that AMD's design is the better of the two. The interconnect circuitry on the K8 has been there since it was released in Sept 2003, Intel's is an 11th hour hack. I'm not questioning that.
My whole point was that the dual cores we're seeing now are stopgap measures to be first to market with dual core. Both companies are planning to go to a cleaner shared cache design at a later time, but as I said in my original post, that's still 12-18 months away.
Didn't mean to sound quite so harsh, but it was early and I wasn't fully awake. :-)
The Opterons that are just hitting the market ARE a rushed design, however. Yes, they work, yes they're dual core, and yes they were designed for it from the start. The key, however, is that the chips you can buy now have a seperate cache for each core. The 'elegant' designs that aren't anywhere near ready at this time use a shared cache that's accessed by both cores. That's what I meant by 'rushed' design.
Apple holds quite a few patents on UI elements found in OS X. I think the posted screenshots thing is BS however. Don't inventors have a year after publication to file?
AMD and Intel both rushed dual core to market for bragging rights. Both companies are using a design that's really not much more than two dies on the same wafer with a little interconnect circuitry. I think the *most* optimistic estimate I've heard for a clean, shared cache design is 12-18 months away still. This is very new stuff for both companies.
IBM, on the other hand, has been building dual core for several *years* now with the POWER series now. And not just single core - we're talking eight cores on the same wafer last I'd heard.
So no, the PPC970 hasn't received dual core yet, but claiming that IBM 'can't keep up' from a technological standpoint is absolutely ridiculous, and suggests that you don't really know what you're talking about.
vi.
;-)
Correction: Most users who are set in their ways and refuse to try anything new hate it.
It's just like the 'geeks' who set every XP box they touch to the classic style start menu simply because it's not what they're used to. It doesn't matter what might be better, it's different and therefore they hate it.
I'd highly recommend you read this article at Ars Technica regarding the Finder and spatialness. It's more than up to Ars' usual high standard, and should give a better idea of what a spatial interface is, and why it can be a good idea, if implemented right.
That's because the Windows 95 approach to being spacial wasn't very good. On the other hand, MacOS = 9 used a spacial finder, and its absence in OS X is a common complaint amongst the old school Mac crowd. Just because the one implementation you're experienced with sucked doesn't mean the whole concept of a spatial filebrowser is bad.
To clarify: it's the Rebel XT. I have the Rebel and the startup time is somewhere in the 3-5 second range.
:-)
I do agree on the high speed flash cards (Sandisk Ultra II's are usually easiest to find, if pricy), but as far as write time goes, you're a world better off than I am. The Rebel has a 4 shot buffer vs the 14 shot buffer on the XT. It's one of the things that almost makes me want to upgrade, but I think I'll wait another generation and put the money into some better lenses in the meantime.