I think this message I received from Suretrade might explain things for you... It's not about alternate browsers, it's about insecure browsers.
============================
We have been informed by Netscape and VeriSign? that the Digital Certificates** contained in Netscape Navigator? and Netscape? Communicator browsers with version numbers of 4.08 or below are about to expire. In order to maintain the highest level of security, many secure sites, require a current certificate when establishing a connection.
A browser upgrade to Netscape Navigator or Netscape Communicator 4.5 or higher, or conversion to Microsoft? Internet Explorer 4.01 or higher (Apple Macintosh? users must use version 4.5 or higher) will automatically update this certificate. If you do not upgrade to Netscape Navigator or Netscape Communicator 4.5 or higher, or convert to Microsoft Internet Explorer by December 31, 1999, you may no longer be able to connect to any secure site on the Internet, including Suretrade. We will not support earlier versions of these browsers after this date.
PLEASE NOTE THAT THIS IS NOT A YEAR 2000 ISSUE.
To ensure uninterrupted service to Suretrade and other secure web sites please follow the directions below.
========================
(instructions for upgrading followed....)
Netscape 4.05 is not compliant, and it's my understanding that Mozilla M12 doesn't yet have a valid digital security certificate authentication in place yet. I could be wrong.
I think this particular problem is someone getting all freaked out over nothing. Note that the Lynx connection with SSL works fine.
IS THIS REALLY POOR WEB DESIGN?
They should be refusing accounts from these soon-to-be-useless-to-connect-to-secure-sites browsers unless they want a whole bunch of tech support calls on or soon after Jan 1, 2000!
I dumped Earthstink as well, after only about a month and a half of pure hell. Earthlink has been the Satan of online services. Before that, I thought I knew what bad internet service was like from hopping my way through about a dozen of them, but I'd only experienced poor to mediocre service before Earthlink. Earthlink showed me new depths of frustration, and almost led to me to, um,...replace my monitor unnecessarily.
Not only would I get busy signals all the time, the only reliable thing about Earthlink was that I could count on the connection to go comatose after somewhere between 5 and 90 minutes into the dialup. Usually much closer to the 5 minute mark.
When I say comatose, I mean dead as a doornail, in all ways, except that connection would still be there. It wouldn't pass any meaningful data (not even a ping), but dammit, once comatose, it wouldn't drop your call, either!
It was always either that, or a random disconnect in the 15-40 minute range. Never both. And their news server sucked, too. This was less a year ago.
Then I switched to Mindspring, with access in all the cities I travel to.
Since then, I have never (not even once) heard a busy signal, and I've not had any connections go comatose on me, either. My connections stay connected. Every time I leave it up overnight when I go to bed, it's there when I get up, happily churning off data in the morning.
Further, the news server has been quite excellent. This is coming from an avid newsgroup user. If it weren't for Usenet, I probably would have went straight back to Compuserve when I first got on the net about 5 years ago.
Now I realize that no service is great on all POPs, but for the "rust belt" area that I travel, it really has been this good. I've never needed tech support other than to ask a pointy question or two of a senior tech (whom I always engage before asking, so as not to confuse the front-line droids), so I don't know about that part.
Now this. The worst internet service I've ever had is merging with the only one that I've ever been satisfied with...
The only bright spot? I'll have cable modem service here in less than 3 months, and could get by with NetZero, or a minimalist account with AOL (eeeww) or something like that for travel.
Here's hoping that Mindspring/Earthlink doesn't suck for all of those out there who don't have many options. But incompetence is almost always like a contagious disease, and I'm pretty sure that Earthlink will have Mindspring totally screwed up within 6 months.
I know that there are many sites out there cataloging bizarre inventions...
But recent developments, patents like this one and other similar ones we've all read about lately make me wonder...
Is anyone out there cataloging the truly stupid or genuinely obvious "intellectual property" patents that has been granted, either recently, or over the span of time?
Posting restrictions: A good writer would take about five minutes to write up a good post, with nessisary proof backing up the claim. A "Me Too!" post is under one minute. So does a cheapshot flame. I propose a one message per IP/user per 4 minutes time, with adjustments baised on Karma on the time limit (less time to wait for more Karma, more time for less). Do you really care about a subject enough to write a virtual essay, which would be moderated high, or some short statement which really doesn't add more to the topic and stays scored at one or zero? I'd say the most well written ones are ones in which some time is gotten into it, and it shows with the proof used to back it up -- and it takes time to gather that proof. The system is very tuned to how people write!!!
Now this is just damned silly. Who is to say I'm not a newspaper editor, accustomed to writing and re-writing on the fly, therefore skilled at making useful points in a hurry?
Further, I just happen to carefully regard what I'm planning to say long before I hit the reply link. I'm not always sure my arguments are valid before loading the posting page, so I wait and think it over before clicking that link. Many many times more often than not, I'll pass on the opportunity to post because upon further consideration, I find that I don't have anything really worthy of putting up. But when I do click the link, I'm nearly certain about what I plan to say, and can usually type it up in a minute or two, maximum, unless I'm feeling leisurely.
Lastly, time-based bonuses for longer delays between clicking the post link and actually posting will just ensure that those who can't type get bonus points. Why should anyone get penalized just because they can type over 80 wpm?
In a way, moderators regulate other moderators already. If a post is marked down unfairly, other moderators will come along and mark it back up. (This assumes that there are an adequate number of moderator points floating around, which IMHO is not currently the case.) This is already a form of "meta-moderation".
I agree; I think there should be many more moderators with many more points. And a five point scale is hardly enough to get a good "community vote" on what is a good post or not. That's what we really need, a concensus opinion from the most informed, most interested participants of the/. community.
If I were to moderate and see a good, but not great post with a 3, I wouldn't "plus one" it, but let someone else if they felt it was worth it. But if I saw a post scored 15 of 25, I might feel it needs a little adjustment up or down. And with a larger scale to vote on it, I wouldn't feel bad about moderating a moderately good post downward, as long as I wasn't penalizing someone.
There are a lot of posts that are redundant as well, but aren't bad in a way that deserves moderating downward if they result in the poster having karma taken away because they aren't unique enough to please whoever has moderation duties at the time. That's nearly as bad a trouble as flamers when it comes to trying to read/. in a hurry.
I'd take a whole forum of comments deemed worthy by a large group of informed participants over the choices of a few "chosen elites" any time.
All things considered, I think that things are going pretty well, so far.
If you have ever taken apart a hard drive (some of use have) you'd see that the arms are much too long to fit more than one in there without whacking each other. And if you start to say "make the arms smaller" stop and count to five, if the arm cant reach the inside of the disk then it's going to have a rough time reading the boot sector. And with more than one head you're going to have lots of fragmentation.
I have taken apart several drives, and know that what I'm proposing is entirely possible with few changes to the current technologies in today's hard drives.
Maybe you haven't been thinking about this long enough, or perhaps you're locked into the same "that's how it is" pattern that HD engineers are in now. There is more than one way to fit multiple actuator arms on a drive platter.
This first is obvious. Mount the arms in opposite corners of the drive chassis. This would increase the length of the chassis a little bit, but if Quantum has no trouble selling 5.25" "Bigfoot" drives that don't perform any better than standard 3.5" drives, I don't see how they could have trouble selling 3.5" drives that are an inch longer if they perform significantly better.
Keep in mind that the drive actuator has a fixed pivot point, which only allows the heads to travel in an arc across the surface. This arc can only sweep across a fixed portion of the total surface area. No drive head carrier arm known to me could ever sweep even close to 180 degrees of the 360 degree circle, always leaving room for at least one additional actuator arm. In the drives I've disassembled, there is more than enough unswept disk surface space to fit at least two more arms. With a synchronous controller, and/or arms made with an arc cut out of their inner span, there is enough platter space for at least four arms, perhaps as many as six, without a redesign of current platters.
The second solution is just as simple, but would decrease the storage capacity of the drive. I see it as just another tradeoff to be faced in industrial design. Significantly increase the spindle diameter of the drive. This would allow multiple shorter arms without the potential for them to knock into each other.
Sketch it out on paper with a compass. You'll see that what I'm proposing makes perfect sense.
Further, regarding fragmentation, if the drive arms are ran from a synchronous controller, and the heads are fed a data stream that is composed of striped bits of the data being written, fragmentation is no more of an issue than it already is. (Think a moment about a two armed drive. The arms are synchronized, so when head 0 on arm 0 is at point A, head 0 on arm 1 is always exactly 180 degrees away, at the exact same distance from the spindle. One file divided into two data streams gets written simultaneously in two locations, just as it gets read back later. The file is intentionally fragmented into two equal parts, but because these parts are always accessed simultaneously, no detrimental fragmentation is induced.)
I cannot think of any other computer subsystem that is so ready, so ripe for parallelization as the hard drives, yet no one seems to be thinking of parallel computing here.
As stated before, all rights reserved, and manufacturers are still invited to contact me at mturley@mindspring.com for reasonable licensing terms for use of these ideas.:)
I'm shocked that no hard drive manufacturer has thought of this before, but I've had two simple solutions for years that will solve the bottleneck quickly and easily.
Instead of one carrier arm with read/write heads, why not just two, three, or four with a simple multiplexing controller?
The positioning commands would be synchronized, while data bits would be fed in alternate, optimally sized data packets to the heads.
And if they're not already, the heads themselves, on alternate sides of the platters should be running with a split data stream, too. I think that they're not now, otherwise, multiple platter drives would have a much higher sustained data transfer rate than single platter drives.
The old way of thinking has the formatting tracks, as viewed from above with theoretical x-ray vision, laid directly above and below each other on opposite faces of the drive. This is probably thought of as a challenge to maintain, but this is completely unnecessary.
One set of tracks will be offset by a slight amount under my scheme, but that would be completely harmless, as the second head would always be fed and read by the controller with the exact same delay that was incurred by the drive while formatting. Thus the data remains in sync automatically, althought the formatting tracks are offset.
Simple, cheap, effective, and very quick to develop and implement. If the controller were 100% efficient and the data chunks even, dual multiplexed controller arms would double the performance (and half the access time), and multiplexed heads would then again double the data transfer rate. Ta-da! 3ms or less access, and a theoretical potential to quadruple the sustained data transfer rate. All without increasing the platter speed any more.
With further research, a bit of time, and some new logic circuitry, I'm sure that the drive head carrier arms could later be ran asynchronously, therefore reducing even further the random access time.
All rights reserved, manufacturers are invited to contact me at mturley@mindspring.com for reasonable licensing terms.:)
The comments about reinvention of the wheel regarding Linux's position vis a vis Microsoft weren't too bright.
Windows, for all its merits, is fundamentally flawed, and in desperate need of repair. Programmers outside of Microsoft who do not want to wait for MS to eventually work out the problems CAN NOT repair the current "wheel" of Windows without access to the source code. Microsoft won't release it.
The current wheel, though useful for now, is inefficient. Repairs to the wheel are inefficient, late, and expensive.
Reinvention of the wheel, in this case, is the only option. Just as peer reviewed medical journals insure that scientific research is valid, peer review of open source code is making Linux a far better program for all uses far more quickly than Microsoft will be able to move the target of Windows. ("Target" meaning "standard of usefulness for the average user" rather than an objective to be reached or interface to be copied.)
It's only in this time, when MS has a significant head start, that Linux may look like a fool's venture. But now that momentum has built, it will soon become very evident that secret source code operating systems are well along the same path that the dinosaurs followed years ago.
Windows compromises security and reliability for a pretty, easy user interface. Linux makes no such compromises, and is rapidly moving toward a pretty, easy user interface.
While I use both Windows and Linux today, I can already easily see the time when I'll no longer have use for the Windows compromise that the columnist seems so willing to condemn us to for all time.
The article says "The free OS comes in a number of relatively compatible flavours known as "distributions". Debian and Independence (www.independence.seul.org) are two popular distributions."
This kind of baffled me, because Debian was the last of the "Big Four" that I heard of, and this is the first I've ever heard of Independence, and I just researched the heck out of distributions about 9 months ago. (When I finally stopped dinking with other people's machines and set up my own.)
I won't guess which one is the most popular, but wonder why the author chose to put these at the front of the article, as if these were the ones that people who are unfamiliar with Linux should choose.
Michael. (And no, I won't call it GNU/Linux. Circular logic [GNU's not Unix] in the name, even if technically proper, would only create confusion among the pointy-haired and promote an image of dorkiness for Linux and those who recommend it.)
I wouldn't be the least surprised if Micro$haft and CC/Divx were in cahoots. Anyone know where you could get info like that? Has M$ dumped money into CC/Divx? Or have they just given FUD classes to CC drones?
Just so you know, DIVX was invented by and is owned by Circuit City (66%) and an L.A. law firm (34%), I can't remember the name of the firm, but one of the partners' last name starts with a "Z". Circuit City is blowing huge wads trying to get this crap accepted (over $100 million last year) and has publicly announced intentions to spend over $400 million more over the coming few years. They're also trying to find another partner to invest in and promote the format.
DIVX is invented/owned by the slimiest of electronics retailers and a group of L.A. lawyers. Gee, there's a real news flash.
This reminds me of first year WebTV sales. It actually would have been cheaper for WebTV to wrap the boxes in ten dollar bills and give them away than to promote and sell them like they did. I expect CC will find that millions in promotion cannot make a stupid idea like DIVX popular.
I don't think 3Com can compete over the long term with dozens of other companies trying to create their own WinCE PDAs.
Note items 3 and 4.
MS's handheld tactics so far:
Write crappy, bug ridden OS while hyping the vaporware until the gullible press is in a frenzy.
Spend millions convincing hardware companies to invest even more millions developing hardware for it.
Dictate exact hardware requirements to manufacturers, preventing them from innovating on their own, so all the resulting units are essentially identical, generic commodities.
Make as much money as possible licensing the OS while the competition between hardware vendors cuts their margins to pennies per unit.
Revise crappy OS and hardware requirements, try to steal the name and market of the very innovator that created the market.
Don't bother to think about what works and why it works, just more and more questionable "features" in there until the hardware screams for another upgrade.
Repeat as necessary, using monopoly profits and the resulting desperation of the hardware vendors, the press, and MS toadies to whip them in line.
Notes:
MS doesn't care about the hardware competition, and the vendors can't afford to cut prices because they have too much of a vested interest right now.
I tracked the inventory of the WinCE 1 units we received. We had to sell them each an average of SIX times before the damned things stayed sold.
My own tracking of the WinCE 2 units reveals that they aren't much better at "staying out the door."
To hell with that, let us/.ers in on the venture capital financing round! Rob gets new hardware now and all of us (including Rob) get rich later!
VC is where you start. It's also where it's at if you know you've got a success on your hands. When the IPO lemming buyers get their hands on it, the investment grows an order of magnitude or three.
Think about it Rob, a small farm of multiprocessor Alpha boxes, one of those cool new SGIs, a Thinkpad, a Palm V, and 50,000 BTUs of pre-paid heat!
No more space heaters for him, we'll get him the BFH-9000.
Look into the new NLX or LPX format motherboards and cases. They're available now, and they're great for just this type of use. You may have to add your own NIC, though, but I've found usable NICs for this purpose for as little as $10.
This is the route I'm going with for my automotive MP3 player, assuming I don't run across a cheapo used integrated board soon. I don't know why I haven't had much luck finding one yet, they used to be dirt common, and nearly as cheap.
You can find them on the internet shopping agent sites, but be sure that the chips used for the integrated features are supported by whatever OS you use.
I've been extremely happy with IC-direct at www.ic-direct.com, but their server is dead right now. The case/mobo combo will add up to about $140-200 plus shipping.
According to Iomega's web page (Doesn't anyone check these anymore?) the minimum operating temperature for a Zip drive is even higher than than a hard drive. The specs say that the Zip drive's minimum operating temperature is 10C, which would limit its use to 2 1/2 seasons for most US users, worse than the 3 seasons a hard drive affords.
Any other ideas? I'd like to hear them, as I'm in the process of building a 3 season MPEG player for my car that I'd like to use in the winter, too. I'll probably just run the HDD below it's operating specs and see how long it takes to toast itself, after all, decent size drives can be had for $99 in case of a "learning experience."
The problem is not the LCD, car stereos have those already. The problem is that the MP3 files are read from a hard drive.
I have yet to see a hard drive that doesn't list 5C or thereabouts as the minimum operating temperature. Furthermore, it would be useless to take it inside every day because in addition to the 5C minimum operating temperature, hard drive specs also state a limit of maximum temperature change per hour. The maximum I have found is 10C temperature change per hour, so going from a warm house to a cold car is begging for failure.
So without some other memory technology to store the MP3 files, there is no practical way for this to work in the winter in areas where it gets really cold.
Is it worth the price to use it 3 seasons of the year?
Slashdot Mirror
============================
We have been informed by Netscape and VeriSign? that the Digital Certificates** contained in Netscape Navigator? and Netscape? Communicator browsers with version numbers of 4.08 or below are about to expire. In order to maintain the highest level of security, many secure sites, require a current certificate when establishing a connection.
A browser upgrade to Netscape Navigator or Netscape Communicator 4.5 or higher, or conversion to Microsoft? Internet Explorer 4.01 or higher (Apple Macintosh? users must use version 4.5 or higher) will automatically update this certificate. If you do not upgrade to Netscape Navigator or Netscape Communicator 4.5 or higher, or convert to Microsoft Internet Explorer by December 31, 1999, you may no longer be able to connect to any secure site on the Internet, including Suretrade. We will not support earlier versions of these browsers after this date.
PLEASE NOTE THAT THIS IS NOT A YEAR 2000 ISSUE.
To ensure uninterrupted service to Suretrade and other secure web sites please follow the directions below.
========================
(instructions for upgrading followed....)
Netscape 4.05 is not compliant, and it's my understanding that Mozilla M12 doesn't yet have a valid digital security certificate authentication in place yet. I could be wrong.
I think this particular problem is someone getting all freaked out over nothing. Note that the Lynx connection with SSL works fine.
IS THIS REALLY POOR WEB DESIGN?
They should be refusing accounts from these soon-to-be-useless-to-connect-to-secure-sites browsers unless they want a whole bunch of tech support calls on or soon after Jan 1, 2000!
I'm in the same boat here.
...replace my monitor unnecessarily.
I dumped Earthstink as well, after only about a month and a half of pure hell. Earthlink has been the Satan of online services. Before that, I thought I knew what bad internet service was like from hopping my way through about a dozen of them, but I'd only experienced poor to mediocre service before Earthlink. Earthlink showed me new depths of frustration, and almost led to me to, um,
Not only would I get busy signals all the time, the only reliable thing about Earthlink was that I could count on the connection to go comatose after somewhere between 5 and 90 minutes into the dialup. Usually much closer to the 5 minute mark.
When I say comatose, I mean dead as a doornail, in all ways, except that connection would still be there. It wouldn't pass any meaningful data (not even a ping), but dammit, once comatose, it wouldn't drop your call, either!
It was always either that, or a random disconnect in the 15-40 minute range. Never both. And their news server sucked, too. This was less a year ago.
Then I switched to Mindspring, with access in all the cities I travel to.
Since then, I have never (not even once) heard a busy signal, and I've not had any connections go comatose on me, either. My connections stay connected. Every time I leave it up overnight when I go to bed, it's there when I get up, happily churning off data in the morning.
Further, the news server has been quite excellent. This is coming from an avid newsgroup user. If it weren't for Usenet, I probably would have went straight back to Compuserve when I first got on the net about 5 years ago.
Now I realize that no service is great on all POPs, but for the "rust belt" area that I travel, it really has been this good. I've never needed tech support other than to ask a pointy question or two of a senior tech (whom I always engage before asking, so as not to confuse the front-line droids), so I don't know about that part.
Now this. The worst internet service I've ever had is merging with the only one that I've ever been satisfied with...
The only bright spot? I'll have cable modem service here in less than 3 months, and could get by with NetZero, or a minimalist account with AOL (eeeww) or something like that for travel.
Here's hoping that Mindspring/Earthlink doesn't suck for all of those out there who don't have many options. But incompetence is almost always like a contagious disease, and I'm pretty sure that Earthlink will have Mindspring totally screwed up within 6 months.
Best regards,
Michael.
I know that there are many sites out there cataloging bizarre inventions...
But recent developments, patents like this one and other similar ones we've all read about lately make me wonder...
Is anyone out there cataloging the truly stupid or genuinely obvious "intellectual property" patents that has been granted, either recently, or over the span of time?
Posting restrictions: A good writer would take about five minutes to write up a good post, with nessisary proof backing up the claim. A "Me Too!" post is under one minute. So does a cheapshot flame. I propose a one message per IP/user per 4 minutes time, with adjustments baised on Karma on the time limit (less time to wait for more Karma, more time for less). Do you really care about a subject enough to write a virtual essay, which would be moderated high, or some short statement which really doesn't add more to the topic and stays scored at one or zero? I'd say the most well written ones are ones in which some time is gotten into it, and it shows with the proof used to back it up -- and it takes time to gather that proof. The system is very tuned to how people write!!!
Now this is just damned silly. Who is to say I'm not a newspaper editor, accustomed to writing and re-writing on the fly, therefore skilled at making useful points in a hurry?
Further, I just happen to carefully regard what I'm planning to say long before I hit the reply link. I'm not always sure my arguments are valid before loading the posting page, so I wait and think it over before clicking that link. Many many times more often than not, I'll pass on the opportunity to post because upon further consideration, I find that I don't have anything really worthy of putting up. But when I do click the link, I'm nearly certain about what I plan to say, and can usually type it up in a minute or two, maximum, unless I'm feeling leisurely.
Lastly, time-based bonuses for longer delays between clicking the post link and actually posting will just ensure that those who can't type get bonus points. Why should anyone get penalized just because they can type over 80 wpm?
In a way, moderators regulate other moderators already. If a post is marked down unfairly, other moderators will come along and mark it back up. (This assumes that there are an adequate number of moderator points floating around, which IMHO is not currently the case.) This is already a form of "meta-moderation".
/. community.
/. in a hurry.
I agree; I think there should be many more moderators with many more points. And a five point scale is hardly enough to get a good "community vote" on what is a good post or not. That's what we really need, a concensus opinion from the most informed, most interested participants of the
If I were to moderate and see a good, but not great post with a 3, I wouldn't "plus one" it, but let someone else if they felt it was worth it. But if I saw a post scored 15 of 25, I might feel it needs a little adjustment up or down. And with a larger scale to vote on it, I wouldn't feel bad about moderating a moderately good post downward, as long as I wasn't penalizing someone.
There are a lot of posts that are redundant as well, but aren't bad in a way that deserves moderating downward if they result in the poster having karma taken away because they aren't unique enough to please whoever has moderation duties at the time. That's nearly as bad a trouble as flamers when it comes to trying to read
I'd take a whole forum of comments deemed worthy by a large group of informed participants over the choices of a few "chosen elites" any time.
All things considered, I think that things are going pretty well, so far.
If you have ever taken apart a hard drive (some of use have) you'd see that the arms are much too long to fit more than one in there without whacking each other. And if you start to say "make the arms smaller" stop and count to five, if the arm cant reach the inside of the disk then it's going to have a rough time reading the boot sector. And with more than one head you're going to have lots of fragmentation.
:)
I have taken apart several drives, and know that what I'm proposing is entirely possible with few changes to the current technologies in today's hard drives.
Maybe you haven't been thinking about this long enough, or perhaps you're locked into the same "that's how it is" pattern that HD engineers are in now. There is more than one way to fit multiple actuator arms on a drive platter.
This first is obvious. Mount the arms in opposite corners of the drive chassis. This would increase the length of the chassis a little bit, but if Quantum has no trouble selling 5.25" "Bigfoot" drives that don't perform any better than standard 3.5" drives, I don't see how they could have trouble selling 3.5" drives that are an inch longer if they perform significantly better.
Keep in mind that the drive actuator has a fixed pivot point, which only allows the heads to travel in an arc across the surface. This arc can only sweep across a fixed portion of the total surface area. No drive head carrier arm known to me could ever sweep even close to 180 degrees of the 360 degree circle, always leaving room for at least one additional actuator arm. In the drives I've disassembled, there is more than enough unswept disk surface space to fit at least two more arms. With a synchronous controller, and/or arms made with an arc cut out of their inner span, there is enough platter space for at least four arms, perhaps as many as six, without a redesign of current platters.
The second solution is just as simple, but would decrease the storage capacity of the drive. I see it as just another tradeoff to be faced in industrial design. Significantly increase the spindle diameter of the drive. This would allow multiple shorter arms without the potential for them to knock into each other.
Sketch it out on paper with a compass. You'll see that what I'm proposing makes perfect sense.
Further, regarding fragmentation, if the drive arms are ran from a synchronous controller, and the heads are fed a data stream that is composed of striped bits of the data being written, fragmentation is no more of an issue than it already is. (Think a moment about a two armed drive. The arms are synchronized, so when head 0 on arm 0 is at point A, head 0 on arm 1 is always exactly 180 degrees away, at the exact same distance from the spindle. One file divided into two data streams gets written simultaneously in two locations, just as it gets read back later. The file is intentionally fragmented into two equal parts, but because these parts are always accessed simultaneously, no detrimental fragmentation is induced.)
I cannot think of any other computer subsystem that is so ready, so ripe for parallelization as the hard drives, yet no one seems to be thinking of parallel computing here.
As stated before, all rights reserved, and manufacturers are still invited to contact me at mturley@mindspring.com for reasonable licensing terms for use of these ideas.
I'm shocked that no hard drive manufacturer has thought of this before, but I've had two simple solutions for years that will solve the bottleneck quickly and easily.
:)
Instead of one carrier arm with read/write heads, why not just two, three, or four with a simple multiplexing controller?
The positioning commands would be synchronized, while data bits would be fed in alternate, optimally sized data packets to the heads.
And if they're not already, the heads themselves, on alternate sides of the platters should be running with a split data stream, too. I think that they're not now, otherwise, multiple platter drives would have a much higher sustained data transfer rate than single platter drives.
The old way of thinking has the formatting tracks, as viewed from above with theoretical x-ray vision, laid directly above and below each other on opposite faces of the drive. This is probably thought of as a challenge to maintain, but this is completely unnecessary.
One set of tracks will be offset by a slight amount under my scheme, but that would be completely harmless, as the second head would always be fed and read by the controller with the exact same delay that was incurred by the drive while formatting. Thus the data remains in sync automatically, althought the formatting tracks are offset.
Simple, cheap, effective, and very quick to develop and implement. If the controller were 100% efficient and the data chunks even, dual multiplexed controller arms would double the performance (and half the access time), and multiplexed heads would then again double the data transfer rate. Ta-da! 3ms or less access, and a theoretical potential to quadruple the sustained data transfer rate. All without increasing the platter speed any more.
With further research, a bit of time, and some new logic circuitry, I'm sure that the drive head carrier arms could later be ran asynchronously, therefore reducing even further the random access time.
All rights reserved, manufacturers are invited to contact me at mturley@mindspring.com for reasonable licensing terms.
The comments about reinvention of the wheel regarding Linux's position vis a vis Microsoft weren't too bright.
Windows, for all its merits, is fundamentally flawed, and in desperate need of repair. Programmers outside of Microsoft who do not want to wait for MS to eventually work out the problems CAN NOT repair the current "wheel" of Windows without access to the source code. Microsoft won't release it.
The current wheel, though useful for now, is inefficient. Repairs to the wheel are inefficient, late, and expensive.
Reinvention of the wheel, in this case, is the only option. Just as peer reviewed medical journals insure that scientific research is valid, peer review of open source code is making Linux a far better program for all uses far more quickly than Microsoft will be able to move the target of Windows. ("Target" meaning "standard of usefulness for the average user" rather than an objective to be reached or interface to be copied.)
It's only in this time, when MS has a significant head start, that Linux may look like a fool's venture. But now that momentum has built, it will soon become very evident that secret source code operating systems are well along the same path that the dinosaurs followed years ago.
Windows compromises security and reliability for a pretty, easy user interface. Linux makes no such compromises, and is rapidly moving toward a pretty, easy user interface.
While I use both Windows and Linux today, I can already easily see the time when I'll no longer have use for the Windows compromise that the columnist seems so willing to condemn us to for all time.
The article says "The free OS comes in a number of relatively compatible flavours known as "distributions". Debian and Independence (www.independence.seul.org) are two popular distributions."
This kind of baffled me, because Debian was the last of the "Big Four" that I heard of, and this is the first I've ever heard of Independence, and I just researched the heck out of distributions about 9 months ago. (When I finally stopped dinking with other people's machines and set up my own.)
I won't guess which one is the most popular, but wonder why the author chose to put these at the front of the article, as if these were the ones that people who are unfamiliar with Linux should choose.
Michael.
(And no, I won't call it GNU/Linux. Circular logic [GNU's not Unix] in the name, even if technically proper, would only create confusion among the pointy-haired and promote an image of dorkiness for Linux and those who recommend it.)
I wouldn't be the least surprised if Micro$haft and CC/Divx were in cahoots. Anyone know where you could get info like that? Has M$ dumped money into CC/Divx? Or have they just given FUD classes to CC drones?
Just so you know, DIVX was invented by and is owned by Circuit City (66%) and an L.A. law firm (34%), I can't remember the name of the firm, but one of the partners' last name starts with a "Z". Circuit City is blowing huge wads trying to get this crap accepted (over $100 million last year) and has publicly announced intentions to spend over $400 million more over the coming few years. They're also trying to find another partner to invest in and promote the format.
DIVX is invented/owned by the slimiest of electronics retailers and a group of L.A. lawyers.
Gee, there's a real news flash.
This reminds me of first year WebTV sales. It actually would have been cheaper for WebTV to wrap the boxes in ten dollar bills and give them away than to promote and sell them like they did. I expect CC will find that millions in promotion cannot make a stupid idea like DIVX popular.
I don't think 3Com can compete over the long term with dozens of other companies trying to create their own WinCE PDAs.
Note items 3 and 4.
MS's handheld tactics so far:
Write crappy, bug ridden OS while hyping the vaporware until the gullible press is in a frenzy.
Spend millions convincing hardware companies to invest even more millions developing hardware for it.
Dictate exact hardware requirements to manufacturers, preventing them from innovating on their own, so all the resulting units are essentially identical, generic commodities.
Make as much money as possible licensing the OS while the competition between hardware vendors cuts their margins to pennies per unit.
Revise crappy OS and hardware requirements, try to steal the name and market of the very innovator that created the market.
Don't bother to think about what works and why it works, just more and more questionable "features" in there until the hardware screams for another upgrade.
Repeat as necessary, using monopoly profits and the resulting desperation of the hardware vendors, the press, and MS toadies to whip them in line.
Notes:
MS doesn't care about the hardware competition, and the vendors can't afford to cut prices because they have too much of a vested interest right now.
I tracked the inventory of the WinCE 1 units we received. We had to sell them each an average of SIX times before the damned things stayed sold.
My own tracking of the WinCE 2 units reveals that they aren't much better at "staying out the door."
Hey! Who the hell's gonna clean up all this monkey shit?
To hell with that, let us /.ers in on the venture capital financing round! Rob gets new hardware now and all of us (including Rob) get rich later!
VC is where you start. It's also where it's at if you know you've got a success on your hands. When the IPO lemming buyers get their hands on it, the investment grows an order of magnitude or three.
Think about it Rob, a small farm of multiprocessor Alpha boxes, one of those cool new SGIs, a Thinkpad, a Palm V, and 50,000 BTUs of pre-paid heat!
No more space heaters for him, we'll get him the BFH-9000.
Look into the new NLX or LPX format motherboards and cases. They're available now, and they're great for just this type of use. You may have to add your own NIC, though, but I've found usable NICs for this purpose for as little as $10.
This is the route I'm going with for my automotive MP3 player, assuming I don't run across a cheapo used integrated board soon. I don't know why I haven't had much luck finding one yet, they used to be dirt common, and nearly as cheap.
You can find them on the internet shopping agent sites, but be sure that the chips used for the integrated features are supported by whatever OS you use.
I've been extremely happy with IC-direct at www.ic-direct.com, but their server is dead right now. The case/mobo combo will add up to about $140-200 plus shipping.
First, the new Zip is 250, not 200.
According to Iomega's web page (Doesn't anyone check these anymore?) the minimum operating temperature for a Zip drive is even higher than than a hard drive. The specs say that the Zip drive's minimum operating temperature is 10C, which would limit its use to 2 1/2 seasons for most US users, worse than the 3 seasons a hard drive affords.
Any other ideas? I'd like to hear them, as I'm in the process of building a 3 season MPEG player for my car that I'd like to use in the winter, too. I'll probably just run the HDD below it's operating specs and see how long it takes to toast itself, after all, decent size drives can be had for $99 in case of a "learning experience."
The problem is not the LCD, car stereos have those already. The problem is that the MP3 files are read from a hard drive.
I have yet to see a hard drive that doesn't list 5C or thereabouts as the minimum operating temperature. Furthermore, it would be useless to take it inside every day because in addition to the 5C minimum operating temperature, hard drive specs also state a limit of maximum temperature change per hour. The maximum I have found is 10C temperature change per hour, so going from a warm house to a cold car is begging for failure.
So without some other memory technology to store the MP3 files, there is no practical way for this to work in the winter in areas where it gets really cold.
Is it worth the price to use it 3 seasons of the year?
Best regards,
Michael.