At 1600x1200 and X at 100 dpi, readably-sized anti-alised TrueType fonts in KDE/Konq look MUCH better than the bitmapped fonts or even scaled Type 1 fonts do. I once thought as you do -- that AA wasn't a big deal and as a Linux user, I could well do without it. But now, it's a real hardship when I have to use other non-AA browsers.
On item #4... maybe I'm missing something (you're the second person to say that Mozilla does easy ads for some time now), but my 07/27 nightly does not have this feature, as far as I can see -- submenus do not have an add bookmark function at the top. What am I doing wrong?
Now, if they'll just get AA fonts (being worked on, apparently) and if 0.9.3 is introducing easy-add bookmarks (as another poster says), then Mozilla may finally be catching up to Konqueror! Probably I'll still use Konq, though. I have Mozilla 0.9.2 installed right now, and it's way slower...
Still, it's nice to know that the Linux browsers are making it easy to get rid of these damn popups. Some people here are complaining about another Slashdot rant on popups, but they drive me NUTS. Nothing pisses me off more than to close a maximized browser window, find 20 popups below it, and realize that people have been sucking my bandwidth for the last n hours and making money off me in the meantime.
Things to do with Konqueror that you can't do with the others, at least in Linux:
1. Disable pop-unders and pop-overs. 2. View the Web in anti-aliased fonts.
3. Open n related pages in one split window.
4. Add bookmarks to specific folders/submenus.
5. Enjoy browsing the Web.
If only Mozilla had come up with something so nice, more people would be using it now...
I haven't had any experience with a TiVO, but a lot of consumer-level (and that's all I can afford) MPEG-2 encoding hardware for PC produces truly lousy quality video compared to what the software encoders can do.
Still, I'm sure you're right, and I have faith that the market for this sort of thing will improve. But I was really just trying to make a point about the old "we don't need more speed" argument...
You ALWAYS need more speed. If someone can't figure out what to do with more CPU power, they don't have much imagination.
With my 1GHz Athlon, it takes ~10 hours to encode 1 hour of MPEG-2 DVD video (720x480 MPEG-2 7M-9M/s).
A 10GHz CPU with linear scaling from my current 1GHz CPU would reduce the time from about 10 hours to about 1 hour. So, if I could just get a 100GHz CPU, I would finally be able to encode 2 hours of MJPEG DV or VHS capture in just about 12 minutes.
I'm a fairly proficient Unix/Linux admin, and I was fighting script kiddies left and right on my home machine for several years (I got rooted twice over three years). I was running my main Linux box with masquerading and filtering for a couple of other PCs and my laptop, at first on ISDN and then on cable. The only reason I didn't install a dedicated firewall at home all that time was because it felt cumbersome, like it would take up extra space and electricity and just be overkill for the small "home" network sitting behind it.
But finally I just got tired of being scanned all the time and seeing people always trying things, so (not wanting to shell out $$$ for a commercial firewall/router), I got some spare parts: a 486DX4/100 board, 16MB ram, a floppy, and two 3Com 3c509 cards. Basically, spare parts.
I bolted the parts all into a cardboard box (it works, just find a stiff box, poke holes in it with a screw driver, and use washers with your screws). Then, I put Freesco (which is Linux-based) on a floppy disk and put the box between my local network and the outside world.
It's been running for a year now and I haven't even thought about it since. Not a single outsider has even come close to touching my PCs -- the Freesco 486-cardboard-box firewall/router has worked very well and I have yet to have to manually reboot it.
Many of us haven't bought MS products in years, but we remain a minority because of the status quo in the current market and business worlds.
I think more than just voting negatively with our dollars (i.e. not supporting companies whose practices we don't like), it's time to vote positively with our dollars by giving them to the EFF, FSF, and so on. Let's put that money to work for the good guys.
This comment doesn't apply to the Model 100 machines, but rather to 300 baud modems in general.
If the line was good, you were in good shape at 300 baud -- you couldn't type that fast at all. On the other hand, if the line was bad, you'd get lots of resent blocks (over and over and over and over) using Xmodem or (god forbid) Ymodem, and then sometimes typing a document over again would be MUCH faster.
What would drive you nuts was when you were sending a 100k or 200k document and after hours of Ymodem you were on block 157 of 158 and the carrier would be dropped. You'd pretty much eat the furniture.
Zmodem was supposed to fix this with some kind of resume function, but it never worked right for me and Zmodem in general always seemed a little flaky...
Then the 9.6k and 14.4k modems with error correction came out and all of these concerns went away...
I wrote my first two books on a Tandy TRS-80 M100.
It was a beautiful device -- instant on, instant off (much more instant than, say, Windows CE) with no moving or hot parts and a basic, no-frills set of applications (the text editor was what I used most). The characters were large and easy-to-read and the keyboard was actually full-size and felt more or less like any old keyboard on any old computer. The whole thing was no thicker than a textbook (and much lighter) and I used it continuously and transferred the data to my PC using a null modem cable.
Then in 1995, it was stolen right out from under my nose at a busy public library -- the power was on, I was in mid-sentence and stopped to turn around and grab a reference from the table behind me. When I turned back, someone (one of about twenty people nearby) had taken it, and by the time I got around to saying "hey, somebody just stole my computer!" the selection of people in the vicinity had completely changed and it was long gone. Old, maybe -- but obviously still in demand.
After that happened, I decided it was time to update my mobile computing, so I got a nice high-end 486 color notebook computer made by AST and ran Linux on it. These days I'm running a ThinkPad 760XD with Linux.
But I'm nowhere near as productive, in sheer page count or imagination, as I was on my M100. I've begun to price them on eBay once again...
Don't know what you're talking about, but my Slackware 8 CD contains XFree86 4.1.0 -- all of it -- and I am using it with accelerated 3D support (NVidia drivers) on my GeForce2 card.
Perhaps you've taken a wrong turn somewhere?
Whoever told you that you had to compile your own X server was likely playing with you because you seem to have taken a wrong turn somewhere. Honestly, XF86_SVGA is the one you want, and combine it with whichever one of the loadable drivers you prefer (XFree86 4 and later uses a loadable driver module system donated by MetroLink, IIRC).
If you have an older card (many S3 cards, for example), you aren't supported under XFree86 anymore (as of release 4) and will have to stick to an XFree86 3.x distribution or downgrade before you can get accelerated graphics.
Unfortunately, many of the suggestions in studies like this one (and they're valid) involve the removal of "adjustments" and configurability because (and this is true) it is easy for novices to screw their desktops up beyond their ability to recover.
This then creates a big conflict: the technical users who want configurability vs. studies like this one which show that for end users, it must be taken out.
It's not a matter of THINK vs. KNOW about safety levels. Accidents like three-mile-island weren't caused by THINKing a technology was safe when it wasn't -- we KNEW all along that the technology was only safe *IF* certain conditions were met and maintained. We also KNEW that if those conditions were not maintained, fission would lead to a chain reaction and great danger to all around.
It (like drilling for oil, driving your car or eating Fugu) was a calculated risk offset by some expected gain (in this case, energy).
Fusion is much safer than fission because unlike fission it is not self-sustaining at the scale humans are talking about. To compare: if containment in a fission reactor (i.e. three-mile-island) is removed, MORE energy will be generated and the amount of energy being generated in the system will INCREASE as a function of time without intervention. We always knew this.
In fusion on the other hand, if containment is removed, LESS energy will be generated (because it is the "containment" which causes the reaction in the first place by adding pressure) and the amount of energy being generated in the system will DECREASE as a function of time without intervention.
In the case of fusion, the amount of time we are talking about is also very small. If you have a catastrophic failure and lose your containment in a fusion reactor of this type, you may at most have a small burst of energy (perhaps enough to knock down a building or two) for a very brief moment, but because the pressure of containment (which just failed) is required to sustain the reaction and produce energy, you will almost instantly lose all energy in the system and the little bit of gas (fairly boring gasses, nothing so dangerous as the materials used in fission) will dissipate and join the rest of the gasses in the atmosphere. And because the materials involved are different, there will be no radiation left behind for future generations at all if an "accident" does occur.
We have understood both technologies since the dawn of the "nuclear age". Fusion has always been preferable -- higher potential energy returns with less danger. We've always even known how to make a fusion reactor in concept -- but building something accurate enough to actually make power requires sophisticated tools that we are only now beginning to have. Fission we figured we could do "now" and *MAKE* safe (note the "IF" statement above regarding conditions).
In short, "containment" in a fission reactor is required to keep the reaction from running away, while "containment" in this type of fusion reactor is required to keep the reaction happening in the first place.
In the nuclear future (and it almost has to be that way), you definitely want your kids' power coming from fusion rather than fission.
I love this idea. I am very interested in projects like this which aim to give information tools to those in cultures who don't yet have access to these sorts of things that we do.
However, there is one drawback: at a price tag of $1000 US, even I couldn't afford it right now, and I make more in a month than many entire villages in the third world do.
Why do we need 10,000 RPM? What about 7,200 or even 5,400? I've seen 5.25" drives at both of these rotational speeds.
The current DiamondMax 80MB drive is a 5,400 RPM drive and it still gives transfer rates good enough for full motion NTSC (720x480) video at 2:1 compression. And I get 30+MB/sec sustained out of my 7200RPM IDE drives that are (now) a generation old.
With such high data density, I don't see the speed issue, especially since even PCs are now starting to use RAID-0 or RAID-5 on a regular basis.
How many platters are in the current 3.5"x1" ATA 80GB drives? Seems like just by going back to the 5.25" full-height format we could hit 250-300GB with today's technology. I don't see that as too big a sacrifice.
A 4-bay RAID case for FH 5.25" drives is about the size of a mini-tower. That's 1200GB in a PC-size format. That's not too bad... And a desktop mid-tower could probably hold two drives. So I think it wouldn't be too hard to keep home users in 1000GB or so without needing lots of extra space.
Isn't 1000GB on a desktop system enough to hold us for 5 years or so?
Re:What anthropologists say when they don't know
on
Pillars Underwater
·
· Score: 5
"Religious significance" may be what "television anthropologists" cite, but speaking as an Anthropologist of sorts, I'd have to say that it's not always the best explanation for things.
Also understand that pop culture can sometimes reinforce things like "religious significance" as an explanation. For example, witness the number of people running around these days with the hazy idea that in early human history, we all worshipped some universal "mother goddess" across all cultures -- very politically correct, and very popular... Seems to give people, especially women, a real warm fuzzy. The evidence people have heard/seen? So-called "fertility figurines" from any number of cultures as seen [of course] on TV documentaries. And of course, it's all nonsense.
This view is certainly not the accepted one within the academic community, for the most part, and certainly not in the universal sense. For all we know, these figurines are the Barbie Dolls of ancient children in one culture and wig-holders in another. But you won't sell product if you spend an hour saying "we don't know" on television.
What I'm getting at is that when you say "judging by the way most anthropologists and archaeologists classify..." it appears that you're mostly looking at the wrong anthropologists and archaeologists. Read the Anthropology journals instead -- the cable channels are in it for the ratings.
I was about to bitch about this because I just barely went to -pre9 on several machines, but from the ChangeLog it looks like '-pre9' became 'final' so I can just let those stay in place.
Or is the ChangeLog not yet complete? (Say, ChangeLog-pre1?)
Try a high-quality (i.e. name-brand) SCSI drive. My experience has been that nearly all EIDE DVD and CD drives sound like airplanes taking off and won't read about half of my CDs/DVDs, while all of the SCSI models I've tried (Pioneer, Toshiba) have been silent, vibration-free and read everything without so much as a hiccup.
Of course, the SCSI models have cost more than twice as much as the EIDE models, but then... I guess you get what you pay for!
Re:Slackware is the best Linux distribution. Perio
on
Slackware 8.0 Released
·
· Score: 1
YOUR SYSTEM, THOUGH LACKING A PACKAGE MANAGER, STILL HAS DEPENDENCIES!!!
Yes, and when you don't have a package manager to f*ck with you can resolve them YOURSELF, using HUMAN intelligence! You can install MULTIPLE VERSIONS of software in MULTIPLE LOCATIONS and then using the shell/scripting/administrative capabilities that made Unix so great to begin with, you can automatically select between them on a per-application, as-needed basis.
With so-called "package management" if you install a new version, the old one's got to go. No good. No good. Every time I've used Debian or an RPM-based system, I end up with EVERY LAST DAMN THING in the system in the/usr/local tree because I've compiled it myself as way to get around the versioning and dependencies in package management.
Of course then you're screwed because eventually your package database doesn't match what you're really using on a day-to-day basis. Then you buy a commercial application, it detects that you're using a package-based system, and installs something based on what it sees in your package database -- which then doesn't work because you gave up on keeping a current package database long ago.
Package management gets in the way of a competent system administrator's day.
Re:The Neverending wave of criticism of slackware.
on
Slackware 8.0 Released
·
· Score: 2
Slackware is what Debian should have been.
Debian is the Red Hat of source-friendly distributions, if you will. Too much cruft. Dependencies should be up to the administrator, not up to the maintainers, because the maintainers will eventually get something wrong. All of the source is included with Slackware as well, only as nice vanilla.tgz files like they ought to be, butt naked like when they were born.
I tried both Debian 2.0 and Debian 2.2 and the complexity level and rigidness level are on part with that of RPM-based distributions, only not as polished.
Maybe it's better said like this: Slackware is the old-school sysadmin's distro!
Slackware is the best Linux distribution. Period.
on
Slackware 8.0 Released
·
· Score: 5
I used Slackware for the first time around '93 because it was the simplest to download and install at the time (using a whole stack of floppies). Red Hat hadn't happened yet and X was very cool just being X, no need for integrated API's like KDE and Gnome.
I left Slackware around 4.0 because I wanted the new "glibc" goodness and I couldn't figure out why Patrick wasn't including it. Unfortunately, I soon found out as I fought Red Hat 5.0 for several months. Since then on my personal box I've used Red Hat 5.1, OpenLinux 1.3 (gave up on "glibc" for a while thanks to Red Hat), Debian 2.0, OpenLinux 2.4, Debian 2.2 and Red Hat 7.1. Last night I downloaded Slack 8.0 (the hard way, ISOs weren't out yet) and re-installed it again to replace Red Hat 7.1.
Slackware is as great as EVER! Very stable, nice BSD-like setup with lovely non-spaghetti init scripts, a *working* KDE 2.1.x/Gnome 1.4 setup, X 4.1.0 without having to kludge it in by hand (what's with Red Hat mixing 4.x with the 3.x servers in the same installation?!), and all of the great stuff that nobody else seems to bother to include (Netatalk! XView! fortune!)
And it doesn't try to configure EVERY LAST DAMN THING during the install process. Red Hat 7.1 takes six years to install because it tries to configure things in "anaconda" -- but it's wasted time because you just have to go back and "fix" all of those automatic configs anyway to get them like you like them. Slackware installs it and then leaves it alone so that you can get it right the first time. For example I get to run "X -configure" or "xf86config" myself rather than having to fight some default setup that didn't really work anyway.
And even the installer itself is sheer heaven -- same simple stuff. Boot in, run fdisk then setup, where you make your own tagfiles to select on a package-by-package basis what goes in and what doesn't. Then, you start it off and it runs, no graphics, no having to click-click-click... In short, no shit. Once you save your tagfiles, it's an unintended install, and on as many machines as you want, even on machines without XFree86 or VGA-compatible graphics hardware.
Way to go, Patrick! Slackware still is and will always be the Linux for Unix users. No showstoppers, packages without the strictness of.rpm or.deb, "technical" purity and simplicity! I don't know why I ever left, and I won't leave again as long as it's available. I'm going to go and buy the CD now to support the effort.
Slashdot Mirror
Have to disagree on #'s 2 and 4.
At 1600x1200 and X at 100 dpi, readably-sized anti-alised TrueType fonts in KDE/Konq look MUCH better than the bitmapped fonts or even scaled Type 1 fonts do. I once thought as you do -- that AA wasn't a big deal and as a Linux user, I could well do without it. But now, it's a real hardship when I have to use other non-AA browsers.
On item #4... maybe I'm missing something (you're the second person to say that Mozilla does easy ads for some time now), but my 07/27 nightly does not have this feature, as far as I can see -- submenus do not have an add bookmark function at the top. What am I doing wrong?
Coolness! Thanks! No more pop-ups in Mozilla!
Now, if they'll just get AA fonts (being worked on, apparently) and if 0.9.3 is introducing easy-add bookmarks (as another poster says), then Mozilla may finally be catching up to Konqueror! Probably I'll still use Konq, though. I have Mozilla 0.9.2 installed right now, and it's way slower...
Still, it's nice to know that the Linux browsers are making it easy to get rid of these damn popups. Some people here are complaining about another Slashdot rant on popups, but they drive me NUTS. Nothing pisses me off more than to close a maximized browser window, find 20 popups below it, and realize that people have been sucking my bandwidth for the last n hours and making money off me in the meantime.
Things to do with Konqueror that you can't do with the others, at least in Linux:
1. Disable pop-unders and pop-overs.
2. View the Web in anti-aliased fonts.
3. Open n related pages in one split window.
4. Add bookmarks to specific folders/submenus.
5. Enjoy browsing the Web.
If only Mozilla had come up with something so nice, more people would be using it now...
I haven't had any experience with a TiVO, but a lot of consumer-level (and that's all I can afford) MPEG-2 encoding hardware for PC produces truly lousy quality video compared to what the software encoders can do.
Still, I'm sure you're right, and I have faith that the market for this sort of thing will improve. But I was really just trying to make a point about the old "we don't need more speed" argument...
You ALWAYS need more speed. If someone can't figure out what to do with more CPU power, they don't have much imagination.
Encoding DVDs.
With my 1GHz Athlon, it takes ~10 hours to encode 1 hour of MPEG-2 DVD video (720x480 MPEG-2 7M-9M/s).
A 10GHz CPU with linear scaling from my current 1GHz CPU would reduce the time from about 10 hours to about 1 hour. So, if I could just get a 100GHz CPU, I would finally be able to encode 2 hours of MJPEG DV or VHS capture in just about 12 minutes.
That would be nice.
Sever timed out while trying to access www.atheos.cx.
Sorry, AtheOS, we just zeroed your uptime counter.
Well I meant to get a proper case, but when I got to Auntie Wainwright's, the cardboard box was all I could afford...
I'm a fairly proficient Unix/Linux admin, and I was fighting script kiddies left and right on my home machine for several years (I got rooted twice over three years). I was running my main Linux box with masquerading and filtering for a couple of other PCs and my laptop, at first on ISDN and then on cable. The only reason I didn't install a dedicated firewall at home all that time was because it felt cumbersome, like it would take up extra space and electricity and just be overkill for the small "home" network sitting behind it.
But finally I just got tired of being scanned all the time and seeing people always trying things, so (not wanting to shell out $$$ for a commercial firewall/router), I got some spare parts: a 486DX4/100 board, 16MB ram, a floppy, and two 3Com 3c509 cards. Basically, spare parts.
I bolted the parts all into a cardboard box (it works, just find a stiff box, poke holes in it with a screw driver, and use washers with your screws). Then, I put Freesco (which is Linux-based) on a floppy disk and put the box between my local network and the outside world.
It's been running for a year now and I haven't even thought about it since. Not a single outsider has even come close to touching my PCs -- the Freesco 486-cardboard-box firewall/router has worked very well and I have yet to have to manually reboot it.
Many of us haven't bought MS products in years, but we remain a minority because of the status quo in the current market and business worlds.
I think more than just voting negatively with our dollars (i.e. not supporting companies whose practices we don't like), it's time to vote positively with our dollars by giving them to the EFF, FSF, and so on. Let's put that money to work for the good guys.
This comment doesn't apply to the Model 100 machines, but rather to 300 baud modems in general.
If the line was good, you were in good shape at 300 baud -- you couldn't type that fast at all. On the other hand, if the line was bad, you'd get lots of resent blocks (over and over and over and over) using Xmodem or (god forbid) Ymodem, and then sometimes typing a document over again would be MUCH faster.
What would drive you nuts was when you were sending a 100k or 200k document and after hours of Ymodem you were on block 157 of 158 and the carrier would be dropped. You'd pretty much eat the furniture.
Zmodem was supposed to fix this with some kind of resume function, but it never worked right for me and Zmodem in general always seemed a little flaky...
Then the 9.6k and 14.4k modems with error correction came out and all of these concerns went away...
I wrote my first two books on a Tandy TRS-80 M100.
It was a beautiful device -- instant on, instant off (much more instant than, say, Windows CE) with no moving or hot parts and a basic, no-frills set of applications (the text editor was what I used most). The characters were large and easy-to-read and the keyboard was actually full-size and felt more or less like any old keyboard on any old computer. The whole thing was no thicker than a textbook (and much lighter) and I used it continuously and transferred the data to my PC using a null modem cable.
Then in 1995, it was stolen right out from under my nose at a busy public library -- the power was on, I was in mid-sentence and stopped to turn around and grab a reference from the table behind me. When I turned back, someone (one of about twenty people nearby) had taken it, and by the time I got around to saying "hey, somebody just stole my computer!" the selection of people in the vicinity had completely changed and it was long gone. Old, maybe -- but obviously still in demand.
After that happened, I decided it was time to update my mobile computing, so I got a nice high-end 486 color notebook computer made by AST and ran Linux on it. These days I'm running a ThinkPad 760XD with Linux.
But I'm nowhere near as productive, in sheer page count or imagination, as I was on my M100. I've begun to price them on eBay once again...
Don't know what you're talking about, but my Slackware 8 CD contains XFree86 4.1.0 -- all of it -- and I am using it with accelerated 3D support (NVidia drivers) on my GeForce2 card.
Perhaps you've taken a wrong turn somewhere?
Whoever told you that you had to compile your own X server was likely playing with you because you seem to have taken a wrong turn somewhere. Honestly, XF86_SVGA is the one you want, and combine it with whichever one of the loadable drivers you prefer (XFree86 4 and later uses a loadable driver module system donated by MetroLink, IIRC).
If you have an older card (many S3 cards, for example), you aren't supported under XFree86 anymore (as of release 4) and will have to stick to an XFree86 3.x distribution or downgrade before you can get accelerated graphics.
Unfortunately, many of the suggestions in studies like this one (and they're valid) involve the removal of "adjustments" and configurability because (and this is true) it is easy for novices to screw their desktops up beyond their ability to recover.
This then creates a big conflict: the technical users who want configurability vs. studies like this one which show that for end users, it must be taken out.
It's not a matter of THINK vs. KNOW about safety levels. Accidents like three-mile-island weren't caused by THINKing a technology was safe when it wasn't -- we KNEW all along that the technology was only safe *IF* certain conditions were met and maintained. We also KNEW that if those conditions were not maintained, fission would lead to a chain reaction and great danger to all around.
It (like drilling for oil, driving your car or eating Fugu) was a calculated risk offset by some expected gain (in this case, energy).
Fusion is much safer than fission because unlike fission it is not self-sustaining at the scale humans are talking about. To compare: if containment in a fission reactor (i.e. three-mile-island) is removed, MORE energy will be generated and the amount of energy being generated in the system will INCREASE as a function of time without intervention. We always knew this.
In fusion on the other hand, if containment is removed, LESS energy will be generated (because it is the "containment" which causes the reaction in the first place by adding pressure) and the amount of energy being generated in the system will DECREASE as a function of time without intervention.
In the case of fusion, the amount of time we are talking about is also very small. If you have a catastrophic failure and lose your containment in a fusion reactor of this type, you may at most have a small burst of energy (perhaps enough to knock down a building or two) for a very brief moment, but because the pressure of containment (which just failed) is required to sustain the reaction and produce energy, you will almost instantly lose all energy in the system and the little bit of gas (fairly boring gasses, nothing so dangerous as the materials used in fission) will dissipate and join the rest of the gasses in the atmosphere. And because the materials involved are different, there will be no radiation left behind for future generations at all if an "accident" does occur.
We have understood both technologies since the dawn of the "nuclear age". Fusion has always been preferable -- higher potential energy returns with less danger. We've always even known how to make a fusion reactor in concept -- but building something accurate enough to actually make power requires sophisticated tools that we are only now beginning to have. Fission we figured we could do "now" and *MAKE* safe (note the "IF" statement above regarding conditions).
In short, "containment" in a fission reactor is required to keep the reaction from running away, while "containment" in this type of fusion reactor is required to keep the reaction happening in the first place.
In the nuclear future (and it almost has to be that way), you definitely want your kids' power coming from fusion rather than fission.
I love this idea. I am very interested in projects like this which aim to give information tools to those in cultures who don't yet have access to these sorts of things that we do.
However, there is one drawback: at a price tag of $1000 US, even I couldn't afford it right now, and I make more in a month than many entire villages in the third world do.
Why do we need 10,000 RPM? What about 7,200 or even 5,400? I've seen 5.25" drives at both of these rotational speeds.
The current DiamondMax 80MB drive is a 5,400 RPM drive and it still gives transfer rates good enough for full motion NTSC (720x480) video at 2:1 compression. And I get 30+MB/sec sustained out of my 7200RPM IDE drives that are (now) a generation old.
With such high data density, I don't see the speed issue, especially since even PCs are now starting to use RAID-0 or RAID-5 on a regular basis.
How many platters are in the current 3.5"x1" ATA 80GB drives? Seems like just by going back to the 5.25" full-height format we could hit 250-300GB with today's technology. I don't see that as too big a sacrifice.
A 4-bay RAID case for FH 5.25" drives is about the size of a mini-tower. That's 1200GB in a PC-size format. That's not too bad... And a desktop mid-tower could probably hold two drives. So I think it wouldn't be too hard to keep home users in 1000GB or so without needing lots of extra space.
Isn't 1000GB on a desktop system enough to hold us for 5 years or so?
"Religious significance" may be what "television anthropologists" cite, but speaking as an Anthropologist of sorts, I'd have to say that it's not always the best explanation for things.
Also understand that pop culture can sometimes reinforce things like "religious significance" as an explanation. For example, witness the number of people running around these days with the hazy idea that in early human history, we all worshipped some universal "mother goddess" across all cultures -- very politically correct, and very popular... Seems to give people, especially women, a real warm fuzzy. The evidence people have heard/seen? So-called "fertility figurines" from any number of cultures as seen [of course] on TV documentaries. And of course, it's all nonsense.
This view is certainly not the accepted one within the academic community, for the most part, and certainly not in the universal sense. For all we know, these figurines are the Barbie Dolls of ancient children in one culture and wig-holders in another. But you won't sell product if you spend an hour saying "we don't know" on television.
What I'm getting at is that when you say "judging by the way most anthropologists and archaeologists classify..." it appears that you're mostly looking at the wrong anthropologists and archaeologists. Read the Anthropology journals instead -- the cable channels are in it for the ratings.
I was about to bitch about this because I just barely went to -pre9 on several machines, but from the ChangeLog it looks like '-pre9' became 'final' so I can just let those stay in place.
Or is the ChangeLog not yet complete? (Say, ChangeLog-pre1?)
Try a high-quality (i.e. name-brand) SCSI drive. My experience has been that nearly all EIDE DVD and CD drives sound like airplanes taking off and won't read about half of my CDs/DVDs, while all of the SCSI models I've tried (Pioneer, Toshiba) have been silent, vibration-free and read everything without so much as a hiccup.
Of course, the SCSI models have cost more than twice as much as the EIDE models, but then... I guess you get what you pay for!
Ha! Sorry...
"unattended install"
Though unintended might be interesting...
Um, Slackware 8 comes with KDE 2.1.2, Gnome 1.4 and XFree86 4.1.0. Whose GUI is modern, now?
Yes, and when you don't have a package manager to f*ck with you can resolve them YOURSELF, using HUMAN intelligence! You can install MULTIPLE VERSIONS of software in MULTIPLE LOCATIONS and then using the shell/scripting/administrative capabilities that made Unix so great to begin with, you can automatically select between them on a per-application, as-needed basis.
With so-called "package management" if you install a new version, the old one's got to go. No good. No good. Every time I've used Debian or an RPM-based system, I end up with EVERY LAST DAMN THING in the system in the
Of course then you're screwed because eventually your package database doesn't match what you're really using on a day-to-day basis. Then you buy a commercial application, it detects that you're using a package-based system, and installs something based on what it sees in your package database -- which then doesn't work because you gave up on keeping a current package database long ago.
Package management gets in the way of a competent system administrator's day.
Slackware is what Debian should have been.
.tgz files like they ought to be, butt naked like when they were born.
Debian is the Red Hat of source-friendly distributions, if you will. Too much cruft. Dependencies should be up to the administrator, not up to the maintainers, because the maintainers will eventually get something wrong. All of the source is included with Slackware as well, only as nice vanilla
I tried both Debian 2.0 and Debian 2.2 and the complexity level and rigidness level are on part with that of RPM-based distributions, only not as polished.
Maybe it's better said like this: Slackware is the old-school sysadmin's distro!
I used Slackware for the first time around '93 because it was the simplest to download and install at the time (using a whole stack of floppies). Red Hat hadn't happened yet and X was very cool just being X, no need for integrated API's like KDE and Gnome.
.rpm or .deb, "technical" purity and simplicity! I don't know why I ever left, and I won't leave again as long as it's available. I'm going to go and buy the CD now to support the effort.
I left Slackware around 4.0 because I wanted the new "glibc" goodness and I couldn't figure out why Patrick wasn't including it. Unfortunately, I soon found out as I fought Red Hat 5.0 for several months. Since then on my personal box I've used Red Hat 5.1, OpenLinux 1.3 (gave up on "glibc" for a while thanks to Red Hat), Debian 2.0, OpenLinux 2.4, Debian 2.2 and Red Hat 7.1. Last night I downloaded Slack 8.0 (the hard way, ISOs weren't out yet) and re-installed it again to replace Red Hat 7.1.
Slackware is as great as EVER! Very stable, nice BSD-like setup with lovely non-spaghetti init scripts, a *working* KDE 2.1.x/Gnome 1.4 setup, X 4.1.0 without having to kludge it in by hand (what's with Red Hat mixing 4.x with the 3.x servers in the same installation?!), and all of the great stuff that nobody else seems to bother to include (Netatalk! XView! fortune!)
And it doesn't try to configure EVERY LAST DAMN THING during the install process. Red Hat 7.1 takes six years to install because it tries to configure things in "anaconda" -- but it's wasted time because you just have to go back and "fix" all of those automatic configs anyway to get them like you like them. Slackware installs it and then leaves it alone so that you can get it right the first time. For example I get to run "X -configure" or "xf86config" myself rather than having to fight some default setup that didn't really work anyway.
And even the installer itself is sheer heaven -- same simple stuff. Boot in, run fdisk then setup, where you make your own tagfiles to select on a package-by-package basis what goes in and what doesn't. Then, you start it off and it runs, no graphics, no having to click-click-click... In short, no shit. Once you save your tagfiles, it's an unintended install, and on as many machines as you want, even on machines without XFree86 or VGA-compatible graphics hardware.
Way to go, Patrick! Slackware still is and will always be the Linux for Unix users. No showstoppers, packages without the strictness of
Slackware: old school feel, new school gear.