Hasn't anyone seen their new commercial, where they talk about package tracking data flowing "over these lines", and the picture in the background is high tension power lines, because the marketing department couldn't find stock film footage of a fly-by of an underground fiber optic cable?
The plant ageing clock is the same as the human ageing clock; it's based on something called "the Hayflick limit", which is the limit on the number of times a cell can divide.
You can look it up on the web, but the short version is that each time a cell divides, it shortens the telomeres on the ends of its genes; when it runs out of telomeres, the cell dies (or becomes cancerous, or is subject to other age-related disease processes).
Baby humans and plants don't have this limit, since, in gametogenesis, the telomeres are lengthened by a chemical called "telomerase", effectively resetting the clock for the newly created entity.
Display rendering is very RAM intensive. It is particularly costly for these small devices.
For a 32,000 character Japanese or Chinese font, at 14 pixels square (about the smallest readable resolution), un compressed, you are talking about 800K.
On an 8M Palm, that ends up being 1/10th of your available memory.
Hebrew, Arabic, Tamil, Devengari, or other ligatured languages have much smaller fonts, but since the character rendering changes as a result of which characters are adjavent to each other, or the start or end of the line, you have similar memory constraints for the ligature rendering software, which could be considered "part of" the font.
That's just for display, and doesn't include input.
For something like Pinjin (Chineseh input) or Kanjihand (Japanese input), you are talking additional RAM taken up to allow both "chording", and translation of the pseudo "chords" (unless you have a keyboard) into the textual representation.
Storage for data is less of a problem; but most storage uses EUC or UTF or some other multibyte encoding. If it didn't, you couldn't shove it into 8-bit "files" on a PALM; if PALM supported 16-bit "files", this would be much easier.
But since it doesn't, you don't get the average 2.5:1 information density increase you would normally get from an ideogrammatic language (average English word length is 5 8-bit characters), and it drops down to about equal density (~1.2:1), so you don't win back your memory used on input and display processing that way.
So the net result is about the same as the original Macintosh: all the RAM is taken up by system processing, leaving nothing left for data or programs.
So what this boils down to is that the support has to be built into the OS area, instead of into the user area.
About the only PALM-like device I know that can do this is the Sharp Zarus. All the other vendors tend to fill their FLASH up with, well, pieces of PalmOS, not leaving any private-use areas for language add-on vendors.
PS: Yes, I know my font size of ~800K is uncompressed; the alternative is to compress it, and then include decompression code. That sort of works, but is compute intensive enough to make the system unpleasent to use, with the underpowered processors on most PDAs.
Once again: the MPAA is more concerned with a professional DVD pressing operation pressing 10,000 pirated copies, than they are about ordinary one-off copies by users. A professional operation has no problem writing the CSS tracks; they will modify the electronics if they have to do it.
Piracy is only a problem for them when they hit economies of scale. The ability to distribute large files quickly on the Internet to everyone with a Cable Modem or DSL line is what has them concerned. It's the fact that everyone can (potentially) get a copy quickly, not the fact that one person gets a copy, that has them upset.
Technically, you could actually accuse them of "thinking ahead" to "what do we do, now that the `last mile' problem has been solved?".
The "one person gets a copy" case happens all the time, in stores: it's called "shoplifting of consigned goods". The publisher eats the loss on that, or retail stores would be unwilling to carry the items, because their margin is ~6%, and one or two thefts could wipe out their entire profits from carrying them (e.g. Target, Walmart).
If the issue weren't economies of scale, MPAA would still be upset about VCRs (see prior posting).
In theory, your DVD player adds MacroVision(tm) to the signal.
But MPAA really do not care about analog redistribution, per se, because it's not an issue for them: it doesn't scale to 10,000 copies very easily, like dumping a DeCSS'ed data stream onto Internet II.
What they are more worried about is digital duplication of perfect copies over communications networks, both todays, and those just over the horizon.
At some point, the DVD-R/DVD-RW technology will get to where it can make a verbatim copy of a still-protected DVD, and the game will be up for them, since people will just copy verbatim images of the CSS'ed DVD's themselves.
In fact, you could just send the raw DVD data over a network to a remote site *now*, and burn a copy, and to heck with the idea of CSS anyway: a bit-for-bit copy is identical: let your DVD player DeCSS the contents for you with legal chips.
Then...the region codes supposedly stop that from happening, given that most of the large scale piracy actually occurs in China, and no one would ever think to bring a Region 1 or a region-free DVD player into China, and none of those DVD's could ever make it back to the U.S., right?
American Airlines has announced that the regular "Atlanata Shuttle" flights will now be monitored from Atlanta.
-- Terry
It doesn't mention the good legal reasons...
on
Sun Drops Linux Distro
·
· Score: 2, Interesting
It doesn't mention the good legal reasons for a software house to *not* have their own Linux distribution, and use a third party distribution instead: patents, and "SCO vs. IBM".
If you have a copy of Sun Linux 5, hold onto it.
For all of the Sun patents embodied in the GPL'ed portions of your copy of their distribution, you effectively have a royalty-free license to use those Sun patents, in perpetuity.
This is, BTW, the reason there is no "IBM Linux".
Sun was probably also at least a little afraid of the sabre-rattling of SCO vs. IBM; by discontinuing distribution, they move out of the area of having the SCO monkey on their backs.
All science fiction is about social commentary, if it is intended seriously at all.
And yes, we used to watch Science Fiction movies for product ideas, at IBM. Pick a movie, go in the conference room for the Thursday night brainstorming session, and then write down everything you see that you think you can implement, and everything that comes to mind as a result of that. Then everyone reads their list, making no comments, and people write down what they think of as a result of hearing the lists read.
Quite effective, actually.
-- Terry
The barrier to video on demand: lack of demand.
on
TiVo++ from India
·
· Score: 2, Interesting
The barrier to video on demand: lack of demand. The WiReD Magazine article from September, 1994, said it best.
Ray Noorda was the primary driving force behind the initial success of Novell. Novell was founded in 1979 as NDSI - Novell Data Systems, Inc.. It had a Motorola 68000 based network server box for MP/M and CP/M client machines, and sold everything as a high priced package.
In 1983, the VC forced a reincoporation as just "Novell", and forced Ray Noorda on the founders as "adult supervision" (the VC in question was Safeguard Scientifics).
Ray Noorda changed the business model, and the product line, targeting the newly created IBM PC as both server and client hardware.
Ray Noorda was almost singularly responsible for the success of Novell.
Ray Noorda personally intervened, after the purchase of USL, to get the USL/UCB lawsuit settled. I spent a lot of time talking to him and Mike DeFazio, then VP of the UNIX Systems Group, a legacy executive from AT&T who came with the USL purchase.
Ray Noorda encouraged an executive to move on, after he issued a statement that he didn't like, when that executive stated that Novell/USG was "de-emphasizing UNIX on the desktop". I asked "If not UnixWare, what _Novell_ Operating System will computer users be running on their desktops?" His answer was "None. They will run Windows.". Ray Noorda stormed from the room.
Ray Noorda was to Novell what Thomas Watson was to IBM. He was its strong leader, who forged a stunningly successful company from ashes and raw clay.
Novell was incredibly successful under Noorda. It's stock split 4 times from 1987 to 1992, reaching a high of almost $60 a share before the last split. The closest it's come to that after Noorda was almost $50, in the.COM run-up to the peak of January/February of 2000.
The one really big mistake he made was the purchase of Word Perfect; he did it because he believed that Microsoft was the enemy, and he needed to match product lines against them.
The mistake was in letting the Word Perfect founders know how he valued companies, when they were looking for an exit strategy after the incredible mistake of trying to turn technical support into a profit center. To maximize their "valuation", which Noorda based on PPE - Profit Per Employee - they threw all people not essential to the operation of their base business overboard. All the R&D people working on pen computing, all the human factors and other people who were working on ensuring the product was competitive with Microsoft Word, all of the people who worked on the VMS and UNIX versions of the product. How do you raise PPE? Increase "P" or reduce the number of "E"'s. And that's what they did.
What about funding Caldera? Caldera was funded by Canopy, a VC group answerable to The Noorda Family Trust, *AFTER* Noorda left Novell, *AFTER* Caldera was a going concern, *AFTER* some of the Novell/USG engineers, so fed up with the NIH of the USL side of things, started a "skunk works" project using Linux, and Mike DeFazio, VP of Novell/USG, and dyed-in-the-wool USL, got it shut down because it risked cannibalizing the UnixWare market. Rather than let the idea die, they left Novell and formed Caldera, funded out of the pockets of the two founders: Brian Sparks, to my knowledge, sold 50 acres of family land to fund it. Noorda came in after that, with additional funding from the NFT's Canopy venture fund.
Ray Noorda would not have approved of the cancellation of the Linux project inside Novell (while it was in house, we jokingly called it "LinuxWare").
Ray Noorda had a philosophy which Novell pays lip service to today, but which they no longer really follow: coopetition.
Coopetition is a word coined by Noorda as a combination of "cooperation" and "competition". It was realized in Novell by having 2 or 3 groups working on solving the same problem, and then letting the one that produced the best solution "win", and taking that product to market.
Having a "LinuxWare" project compete with UnixWare, and may the best product win, was the *very essence* of coopetition. Ray Noorda would have approved of it greatly.
When Noorda left as president, remaining on the Board, Novell ran on for a time on inertia, with an "office of the president". But the three people who were chosen for this task lacked sufficient vision, and couldn't carry off the duties of that office in keeping with the same philosophy and corporate culture. They were bean counters, which isn't bad in itself, but they didn't know the heart and soul of Novell.
Blame Caldera, if you must; I don't think that's exactly fair: they started with a good vision, and they got an incredibly bad rap when they initially didn't release source code for some things that they *couldn't* release source code on, because they were licensed from third parties. Yeah, this stuck to them, but I believe it stuck unfairly. I don't believe the people I knew who started the company would do this.
Blame SCO, if you must; I don't think that's exactly fair, either: my first job out of college was developing and porting communications software to around 140 different UNIX platforms, DOS, Windows, Mac, VMS, CP/M, etc., etc., and by far, SCO was always easy to work with, both as an OS, and as a company, and as people. I've had a number of very long talks with Doug Michaels, over the years; some one-on-one, some with one or two people, like Esther Dyson, present, and I hold him in *very* high regard. I don't believe the people I know at SCO would do this.
Blame USL, if you must: personally, that's my chief suspect. But SCO also has Microsoft investment, Microsoft code in their OS, and Microsoft board members. There are plenty of real villains to go around, and plenty of pseudo-villains who are likely just fighting for their jobs and their investments of money, time, and self.
But don't blame Ray Noorda.
PS: Novell, if you are reading this, you can have your soul back any time you want; it was never sold, only pawned.
PPS: IBM, if you are reading this, realize that, unrelated to this case, your soul is sitting on the pawn shop shelf next to Novell's; you can reclaim it any time you want, too, by internalizing your customer-facing philosophy.
Number of connections with SPECweb99: 250,000/500,000.
If you read the posting he referenced, you can see the calculations, and how you can get useful work done. It all boils down to transmit buffer usage (mbufs).
Remember that for most HTTP traffic, you have very small requests, and it's the responses that are larger, so the mbuf usage is asymmetric between inbound and outbound data.
The product this was for was a reverse proxy cache, and so if you didn't care about a lot of content, just getting it out fast, you could compromise between connections and cache size, and operate with 500,000 simultaneous client connections.
This was back in the days when there was an mbuf required per connection for the tcp_template structure. The thing that let me push it to 1.6M was I shrunk the size of that from 256b to 64b. But as of FreeBSD 4.5, the structure went away; a FreeBSD 4.5 based port of the same changes could probably gain another 150,000 connections, which would move the number up to 1.75M. The number of useful connections would (based on cache size) moved up to 300,000 (or 600,000) as a result.
Practically, the cache was a special case, because it was possible to share mbuf chains containing cached content between connections.
The UN is ineffective; the simple reason is that for major decisions, it requires unanimity of the security council to send UN troops anywhere.
Look at the current Iraq issue: any member of the security council can veto any resolution. So even if there is unanimity, minus one... no action.
Basically, this means that the UN can't even vote to censure a security council member's behaviour, because that member would veto the resolution. Thus the top level people effectively have carte blanche: even if everyone in the UN wanted to stand against them, they would have to do it as individual nations, without organization.
All in all, it's pretty toothless. Which has been both good and bad historically (e.g. no UN peacekeepers landing in Waco, Texas, or in Alabama after Brown vs. The Board Of Education, or in Berkeley, CA, etc., during the Vietnam War).
But for an organization which *has* to make decisions on protocols or assignments of address blocks, or dispute resolution, the ability for one member state to render the whole organization indecisive really can't be tolerated.
I was a senior software engineer at Whistle Communications, and later at IBM, for the Whistle InterJet/IBM Web Connections products. I did most of the last generation of email, user account management, mailing list, internal database, and other infrastructure services for the product.
This product has back doors. But they are all explicitly guarded.
From the front panel of the InterJet, you can enable remote management, for a short period of time. This allows a tier 1 support representative to help you configure/maintain your InterJet, while you are on the phone with them.
This required explicit customer consent for remote Web UI based administration.
From the Web UI, if you are logged in as "Admin", there are "secret URLs", which you can use to obtain raw access to the configuration database for much of the InterJet: all of the parts I personally wrote, and some of the rest of it, where the engineers used the standard APIs we had agreed upon for user interface and common configuration store code. This was done to work around the Web UI design, which failed to expose many useful features of the product, which we engineers knew would result in customers inability to use the product as it had been sold to them. It was likewise useful for tier 2 support, to avoid engineering escalations.
This required explicit customer consent for remote Web UI based administration.
Also from the front panel of the InterJet, you can enable "telnet mode". This was done by going to a particular configuration screen on the front panel, and entering a "T" (for "Telnet") on the front panel keypad at that screen. A time limited ability for a remote engineer to come in and manually access the system to diagnose and treat engineering escalations was thereby enabled.
This required explicit customer consent for remote shell based administration.
In addition, this mode only worked from a specific netblock of IP addresses.
Once in at the shell, it was possible for an engineer to force any of these protections. It was common practice for a persistant problem to leave the remote access for engineers open until the problem was verified to be resolved.
There was also a "magic" front panel sequence that would permit you to play "Pong" on the LCD display. I filed a sev-1 bug ("total loss of functionality") against the maintainer, because it did not support "Skunks" (scores of 7-0) as a victory condition. 8-).
All of them were under direct user control, in terms of outside access.
None of these are "proprietary" or "confidential", they just aren't useful to people without documentation.
Other than working around the Web UI designer's intent, with the second back door, none of these really qualifies as nefarious (I would argue that working around the Web UI designers intent qualifies as "routing around the damage").
A Fast File System for UNIX (1984) Marshall Kirk McKusick, William N. Joy, Samuel J. Leffler, Robert S. Fabry http://citeseer.nj.nec.com/mckusick84fast.h tml...otherwise known as BSD FFS.
Current BSD's have this capability, but it is generally disabled, because modern disk drives *lie* about their physical seek boundaries.
Theoretically, you can work around this with SCSI disks by reading the physical geometry off of mode page 2, and then taking it into account when laying out data, to avoid seeks. Maxtor also has a vendor private command for getting this information from some of the more modern ATA drives.
The BSD FFS code can't handle this information without work, though, because the code is very simple, and supports only the idea of uniform length tracks, and does simple math, rather than a table lookup (but it's not that hard to change).
Practically, you could expect a significant speedup, now that the relative spindle speed vs. seek speed makes seeks significant on 10,000 RPM drives; for most drives for the last 5-6 years, though, the seek is in the noise, and it's not that big a win (stepper moters vs. voice coils was the big change that made it matter much less, the first time).
Why is it the people who think Earth is screwed up that want to keep the rest of us down here to suffer with them?
Do they extend this philosophy to the rest of their lives, and insist that every time a dog takes a dump, everyone has to hurry over to stand in it with them?
No change in probability would result from there being even a significant amount of radiation.
Why?
If it's all ionizing, then it won't penetrate very deeply; not nearly enough to hit the theorized liquid water.
Even if we are talking heavy non-ionizing bombardment, which could penetrate fairly deeply, Europa is practically tidally locked in its orbit, just as Earth's moon is. I say practically, because even if the theorized rotation occurs:
the rate is 1 foot of motion every 17 years. It would not be hard for a large population of somethings to keep the bulk of the moon between itself an Jupiter, even if by accident.
-- Terry
Unsustainable business models...
on
Salon Asks for Help
·
· Score: 4, Interesting
Unsustainable business models are a dime a dozen these days.
Salon has spent $80M, and has 50,000 subscribers.
That's a customer acqusition cost of $1,600 per customer.
Say they get their doubled subscribership numbers; that drops the per customer acquisition cost down to $800 per.
Effectively, this means that they would have to get $67 a monthly issue in order to recoup costs, if acquisition was for a period of 1 year, which is normally how these things are measured.
Let's be incredibly generous, and call it 5 years of acquisition. Even so, we are still talking over $13/month/60 issues.
Does anyone really believe that this is going to happen?
These people obviously do not understand cost accounting or cash flow. They may or may not be good journalists, but they certainly are *not* good businessmen.
So, is it just me, or have they failed once again to put the ejection under software control so the pig can have it's driver's detached and the hardware powered down, FS's unmounted, etc., before the thing disappears out from under the OS?
Way to go... we're back at the same place PCMCIA was back in 1994, yet again. 8-(.
Configurability is inversely proportional to supportability.
Configurability damages portability of employee skills, and makes it harder to hire people who can "hit the ground running".
Configurability damages portability of employee skills within a company.
Try this exercise: think of the last non-trivial problem you had. Now think about what it would take to talk your mother through fixing it on the telephone. Now think about what it would take to talk your mother through it on the telephone if she had an entirely different "skin", "theme", or window manager.
Number one on this list is the US Embassy in Grosvenor Square. The area around that building is CCTV city, and has been for some time. Gee, I wonder why? Is it because the British goverment is obsessed with what the US Ambassador is having for lunch, or is it because it's a terrorist target?
I think the British government is obsessed, and given the state of British cuisine, I don't blame them. When was the last time you took a date out for a fine dinner at a British restaurant? Have you even ever seen a British restaurant?
I can't even imagine the conversation that contains "Hey Bob, let's go eat!" "Great idea, Sam! But not Chinese, we had that yesterday... how 'bout we go for some British food?".
On the other hand, I can well imagine this triggering the need for biometric information... after all, after actually eating British food, you'd want to be able to restore your metabolism from the backup copy on your card.
What you can do with a Geo Metro is often more than people give it credit for being able to do.
One of my fond memories from my mis-spent you was taking my Chevy Sprint (the forerunner to the Geo Metro, which was basically a brand rename for the Sprint) up over El Monte. El Monte is basically a "pass" over the mountain from North Fork, in Ogden Valley, UT, through to Porcupine, which is by Logan, UT. Basically, it's 1100 feet of vertical ascent over about 15 miles, over rutted dirt roads, followed by 800 vertical feet of descent (we were going fishing at Porcupine resevouir, and going that way saves 2 hours of travel).
At the very top of the mountain was a group of "he-man, danger-man" types, all there in their extremely jacked-up 4-wheel-drive trucks, drinking beer, and congradulating themselves on being able to "go wheelin'" and get up to the top of the mountain, away from the riff-raff. Each truck has a minimum cost of $15,000, compared to the less than $5,000 for my Sprint (1980's; the eqivalent price for the same truck these days is $45,000+).
The looks on their faces as I came over the crest of the last hill to the top of the mountain, and drove by them with four passengers (Mike, Kike, and Scott), invading their "four-wheel-drive-only" space in my Chevy Sprint was *priceless*.
About the only thing you couldn't do with it was tow a boat or a trailer full of snowmobiles.
Oh yeah; the thing had a carbuerator, not fuel injection, and it got 52 MPG (standard transmission, no air conditioning)...
Slashdot Mirror
UPS Uses It Now in the US!
Hasn't anyone seen their new commercial, where they talk about package tracking data flowing "over these lines", and the picture in the background is high tension power lines, because the marketing department couldn't find stock film footage of a fly-by of an underground fiber optic cable?
8-) 8-) 8-).
-- Terry
The plant ageing clock is the same as the human ageing clock; it's based on something called "the Hayflick limit", which is the limit on the number of times a cell can divide.
You can look it up on the web, but the short version is that each time a cell divides, it shortens the telomeres on the ends of its genes; when it runs out of telomeres, the cell dies (or becomes cancerous, or is subject to other age-related disease processes).
Baby humans and plants don't have this limit, since, in gametogenesis, the telomeres are lengthened by a chemical called "telomerase", effectively resetting the clock for the newly created entity.
-- Terry
Display rendering is very RAM intensive. It is particularly costly for these small devices.
For a 32,000 character Japanese or Chinese font, at 14 pixels square (about the smallest readable resolution), un compressed, you are talking about 800K.
On an 8M Palm, that ends up being 1/10th of your available memory.
Hebrew, Arabic, Tamil, Devengari, or other ligatured languages have much smaller fonts, but since the character rendering changes as a result of which characters are adjavent to each other, or the start or end of the line, you have similar memory constraints for the ligature rendering software, which could be considered "part of" the font.
That's just for display, and doesn't include input.
For something like Pinjin (Chineseh input) or Kanjihand (Japanese input), you are talking additional RAM taken up to allow both "chording", and translation of the pseudo "chords" (unless you have a keyboard) into the textual representation.
Storage for data is less of a problem; but most storage uses EUC or UTF or some other multibyte encoding. If it didn't, you couldn't shove it into 8-bit "files" on a PALM; if PALM supported 16-bit "files", this would be much easier.
But since it doesn't, you don't get the average 2.5:1 information density increase you would normally get from an ideogrammatic language (average English word length is 5 8-bit characters), and it drops down to about equal density (~1.2:1), so you don't win back your memory used on input and display processing that way.
So the net result is about the same as the original Macintosh: all the RAM is taken up by system processing, leaving nothing left for data or programs.
So what this boils down to is that the support has to be built into the OS area, instead of into the user area.
About the only PALM-like device I know that can do this is the Sharp Zarus. All the other vendors tend to fill their FLASH up with, well, pieces of PalmOS, not leaving any private-use areas for language add-on vendors.
PS: Yes, I know my font size of ~800K is uncompressed; the alternative is to compress it, and then include decompression code. That sort of works, but is compute intensive enough to make the system unpleasent to use, with the underpowered processors on most PDAs.
-- Terry
"Consumer DVD recorders cannot write CSS track"
Once again: the MPAA is more concerned with a professional DVD pressing operation pressing 10,000 pirated copies, than they are about ordinary one-off copies by users. A professional operation has no problem writing the CSS tracks; they will modify the electronics if they have to do it.
Piracy is only a problem for them when they hit economies of scale. The ability to distribute large files quickly on the Internet to everyone with a Cable Modem or DSL line is what has them concerned. It's the fact that everyone can (potentially) get a copy quickly, not the fact that one person gets a copy, that has them upset.
Technically, you could actually accuse them of "thinking ahead" to "what do we do, now that the `last mile' problem has been solved?".
The "one person gets a copy" case happens all the time, in stores: it's called "shoplifting of consigned goods". The publisher eats the loss on that, or retail stores would be unwilling to carry the items, because their margin is ~6%, and one or two thefts could wipe out their entire profits from carrying them (e.g. Target, Walmart).
If the issue weren't economies of scale, MPAA would still be upset about VCRs (see prior posting).
-- Terry
"What stops analog copying?"
...the region codes supposedly stop that from happening, given that most of the large scale piracy actually occurs in China, and no one would ever think to bring a Region 1 or a region-free DVD player into China, and none of those DVD's could ever make it back to the U.S., right?
In theory, your DVD player adds MacroVision(tm) to the signal.
But MPAA really do not care about analog redistribution, per se, because it's not an issue for them: it doesn't scale to 10,000 copies very easily, like dumping a DeCSS'ed data stream onto Internet II.
What they are more worried about is digital duplication of perfect copies over communications networks, both todays, and those just over the horizon.
At some point, the DVD-R/DVD-RW technology will get to where it can make a verbatim copy of a still-protected DVD, and the game will be up for them, since people will just copy verbatim images of the CSS'ed DVD's themselves.
In fact, you could just send the raw DVD data over a network to a remote site *now*, and burn a copy, and to heck with the idea of CSS anyway: a bit-for-bit copy is identical: let your DVD player DeCSS the contents for you with legal chips.
Then
-- Terry
In related news...
American Airlines has announced that the regular "Atlanata Shuttle" flights will now be monitored from Atlanta.
-- Terry
It doesn't mention the good legal reasons for a software house to *not* have their own Linux distribution, and use a third party distribution instead: patents, and "SCO vs. IBM".
If you have a copy of Sun Linux 5, hold onto it.
For all of the Sun patents embodied in the GPL'ed portions of your copy of their distribution, you effectively have a royalty-free license to use those Sun patents, in perpetuity.
This is, BTW, the reason there is no "IBM Linux".
Sun was probably also at least a little afraid of the sabre-rattling of SCO vs. IBM; by discontinuing distribution, they move out of the area of having the SCO monkey on their backs.
-- Terry
All science fiction is about social commentary, if it is intended seriously at all.
And yes, we used to watch Science Fiction movies for product ideas, at IBM. Pick a movie, go in the conference room for the Thursday night brainstorming session, and then write down everything you see that you think you can implement, and everything that comes to mind as a result of that. Then everyone reads their list, making no comments, and people write down what they think of as a result of hearing the lists read.
Quite effective, actually.
-- Terry
The barrier to video on demand: lack of demand. The WiReD Magazine article from September, 1994, said it best.
a bs .html
http://www.wired.com/wired/archive/2.09/cable.l
I supposed the lack of DVD support was calculated to ensure a built-in market for the VOD service offering.
I can't really see this device, or the service umbilical, going anywhere any time soon. It failed in 1994, it's fail today.
-- Terry
WHO IS RAY NOORDA? I'll tell you...
.COM run-up to the peak of January/February of 2000.
Ray Noorda was the primary driving force behind the initial success of Novell. Novell was founded in 1979 as NDSI - Novell Data Systems, Inc.. It had a Motorola 68000 based network server box for MP/M and CP/M client machines, and sold everything as a high priced package.
In 1983, the VC forced a reincoporation as just "Novell", and forced Ray Noorda on the founders as "adult supervision" (the VC in question was Safeguard Scientifics).
Ray Noorda changed the business model, and the product line, targeting the newly created IBM PC as both server and client hardware.
Ray Noorda was almost singularly responsible for the success of Novell.
Ray Noorda personally intervened, after the purchase of USL, to get the USL/UCB lawsuit settled. I spent a lot of time talking to him and Mike DeFazio, then VP of the UNIX Systems Group, a legacy executive from AT&T who came with the USL purchase.
Ray Noorda encouraged an executive to move on, after he issued a statement that he didn't like, when that executive stated that Novell/USG was "de-emphasizing UNIX on the desktop". I asked "If not UnixWare, what _Novell_ Operating System will computer users be running on their desktops?" His answer was "None. They will run Windows.". Ray Noorda stormed from the room.
Ray Noorda was to Novell what Thomas Watson was to IBM. He was its strong leader, who forged a stunningly successful company from ashes and raw clay.
Novell was incredibly successful under Noorda. It's stock split 4 times from 1987 to 1992, reaching a high of almost $60 a share before the last split. The closest it's come to that after Noorda was almost $50, in the
The one really big mistake he made was the purchase of Word Perfect; he did it because he believed that Microsoft was the enemy, and he needed to match product lines against them.
The mistake was in letting the Word Perfect founders know how he valued companies, when they were looking for an exit strategy after the incredible mistake of trying to turn technical support into a profit center. To maximize their "valuation", which Noorda based on PPE - Profit Per Employee - they threw all people not essential to the operation of their base business overboard. All the R&D people working on pen computing, all the human factors and other people who were working on ensuring the product was competitive with Microsoft Word, all of the people who worked on the VMS and UNIX versions of the product. How do you raise PPE? Increase "P" or reduce the number of "E"'s. And that's what they did.
What about funding Caldera? Caldera was funded by Canopy, a VC group answerable to The Noorda Family Trust, *AFTER* Noorda left Novell, *AFTER* Caldera was a going concern, *AFTER* some of the Novell/USG engineers, so fed up with the NIH of the USL side of things, started a "skunk works" project using Linux, and Mike DeFazio, VP of Novell/USG, and dyed-in-the-wool USL, got it shut down because it risked cannibalizing the UnixWare market. Rather than let the idea die, they left Novell and formed Caldera, funded out of the pockets of the two founders: Brian Sparks, to my knowledge, sold 50 acres of family land to fund it. Noorda came in after that, with additional funding from the NFT's Canopy venture fund.
Ray Noorda would not have approved of the cancellation of the Linux project inside Novell (while it was in house, we jokingly called it "LinuxWare").
Ray Noorda had a philosophy which Novell pays lip service to today, but which they no longer really follow: coopetition.
Coopetition is a word coined by Noorda as a combination of "cooperation" and "competition". It was realized in Novell by having 2 or 3 groups working on solving the same problem, and then letting the one that produced the best solution "win", and taking that product to market.
Having a "LinuxWare" project compete with UnixWare, and may the best product win, was the *very essence* of coopetition. Ray Noorda would have approved of it greatly.
When Noorda left as president, remaining on the Board, Novell ran on for a time on inertia, with an "office of the president". But the three people who were chosen for this task lacked sufficient vision, and couldn't carry off the duties of that office in keeping with the same philosophy and corporate culture. They were bean counters, which isn't bad in itself, but they didn't know the heart and soul of Novell.
Blame Caldera, if you must; I don't think that's exactly fair: they started with a good vision, and they got an incredibly bad rap when they initially didn't release source code for some things that they *couldn't* release source code on, because they were licensed from third parties. Yeah, this stuck to them, but I believe it stuck unfairly. I don't believe the people I knew who started the company would do this.
Blame SCO, if you must; I don't think that's exactly fair, either: my first job out of college was developing and porting communications software to around 140 different UNIX platforms, DOS, Windows, Mac, VMS, CP/M, etc., etc., and by far, SCO was always easy to work with, both as an OS, and as a company, and as people. I've had a number of very long talks with Doug Michaels, over the years; some one-on-one, some with one or two people, like Esther Dyson, present, and I hold him in *very* high regard. I don't believe the people I know at SCO would do this.
Blame USL, if you must: personally, that's my chief suspect. But SCO also has Microsoft investment, Microsoft code in their OS, and Microsoft board members. There are plenty of real villains to go around, and plenty of pseudo-villains who are likely just fighting for their jobs and their investments of money, time, and self.
But don't blame Ray Noorda.
PS: Novell, if you are reading this, you can have your soul back any time you want; it was never sold, only pawned.
PPS: IBM, if you are reading this, realize that, unrelated to this case, your soul is sitting on the pawn shop shelf next to Novell's; you can reclaim it any time you want, too, by internalizing your customer-facing philosophy.
-- Terry
Number of connections with SPECweb99: 250,000/500,000.
If you read the posting he referenced, you can see the calculations, and how you can get useful work done. It all boils down to transmit buffer usage (mbufs).
Remember that for most HTTP traffic, you have very small requests, and it's the responses that are larger, so the mbuf usage is asymmetric between inbound and outbound data.
The product this was for was a reverse proxy cache, and so if you didn't care about a lot of content, just getting it out fast, you could compromise between connections and cache size, and operate with 500,000 simultaneous client connections.
This was back in the days when there was an mbuf required per connection for the tcp_template structure. The thing that let me push it to 1.6M was I shrunk the size of that from 256b to 64b. But as of FreeBSD 4.5, the structure went away; a FreeBSD 4.5 based port of the same changes could probably gain another 150,000 connections, which would move the number up to 1.75M. The number of useful connections would (based on cache size) moved up to 300,000 (or 600,000) as a result.
Practically, the cache was a special case, because it was possible to share mbuf chains containing cached content between connections.
-- Terry
The UN is ineffective; the simple reason is that for major decisions, it requires unanimity of the security council to send UN troops anywhere.
Look at the current Iraq issue: any member of the security council can veto any resolution. So even if there is unanimity, minus one... no action.
Basically, this means that the UN can't even vote to censure a security council member's behaviour, because that member would veto the resolution. Thus the top level people effectively have carte blanche: even if everyone in the UN wanted to stand against them, they would have to do it as individual nations, without organization.
All in all, it's pretty toothless. Which has been both good and bad historically (e.g. no UN peacekeepers landing in Waco, Texas, or in Alabama after Brown vs. The Board Of Education, or in Berkeley, CA, etc., during the Vietnam War).
But for an organization which *has* to make decisions on protocols or assignments of address blocks, or dispute resolution, the ability for one member state to render the whole organization indecisive really can't be tolerated.
-- Terry
Not All Backdoors Are Nefarious.
I was a senior software engineer at Whistle Communications, and later at IBM, for the Whistle InterJet/IBM Web Connections products. I did most of the last generation of email, user account management, mailing list, internal database, and other infrastructure services for the product.
This product has back doors. But they are all explicitly guarded.
From the front panel of the InterJet, you can enable remote management, for a short period of time. This allows a tier 1 support representative to help you configure/maintain your InterJet, while you are on the phone with them.
This required explicit customer consent for remote Web UI based administration.
From the Web UI, if you are logged in as "Admin", there are "secret URLs", which you can use to obtain raw access to the configuration database for much of the InterJet: all of the parts I personally wrote, and some of the rest of it, where the engineers used the standard APIs we had agreed upon for user interface and common configuration store code. This was done to work around the Web UI design, which failed to expose many useful features of the product, which we engineers knew would result in customers inability to use the product as it had been sold to them. It was likewise useful for tier 2 support, to avoid engineering escalations.
This required explicit customer consent for remote Web UI based administration.
Also from the front panel of the InterJet, you can enable "telnet mode". This was done by going to a particular configuration screen on the front panel, and entering a "T" (for "Telnet") on the front panel keypad at that screen. A time limited ability for a remote engineer to come in and manually access the system to diagnose and treat engineering escalations was thereby enabled.
This required explicit customer consent for remote shell based administration.
In addition, this mode only worked from a specific netblock of IP addresses.
Once in at the shell, it was possible for an engineer to force any of these protections. It was common practice for a persistant problem to leave the remote access for engineers open until the problem was verified to be resolved.
There was also a "magic" front panel sequence that would permit you to play "Pong" on the LCD display. I filed a sev-1 bug ("total loss of functionality") against the maintainer, because it did not support "Skunks" (scores of 7-0) as a victory condition. 8-).
All of them were under direct user control, in terms of outside access.
None of these are "proprietary" or "confidential", they just aren't useful to people without documentation.
Other than working around the Web UI designer's intent, with the second back door, none of these really qualifies as nefarious (I would argue that working around the Web UI designers intent qualifies as "routing around the damage").
-- Terry
First use of this technology... 1984:
h tml ...otherwise known as BSD FFS.
A Fast File System for UNIX (1984)
Marshall Kirk McKusick, William N. Joy, Samuel J. Leffler, Robert S. Fabry
http://citeseer.nj.nec.com/mckusick84fast.
Current BSD's have this capability, but it is generally disabled, because modern disk drives *lie* about their physical seek boundaries.
Theoretically, you can work around this with SCSI disks by reading the physical geometry off of mode page 2, and then taking it into account when laying out data, to avoid seeks. Maxtor also has a vendor private command for getting this information from some of the more modern ATA drives.
The BSD FFS code can't handle this information without work, though, because the code is very simple, and supports only the idea of uniform length tracks, and does simple math, rather than a table lookup (but it's not that hard to change).
Practically, you could expect a significant speedup, now that the relative spindle speed vs. seek speed makes seeks significant on 10,000 RPM drives; for most drives for the last 5-6 years, though, the seek is in the noise, and it's not that big a win (stepper moters vs. voice coils was the big change that made it matter much less, the first time).
-- Terry
Why is it the people who think Earth is screwed up that want to keep the rest of us down here to suffer with them?
Do they extend this philosophy to the rest of their lives, and insist that every time a dog takes a dump, everyone has to hurry over to stand in it with them?
-- Terry
No change in probability would result from there being even a significant amount of radiation.
n s/ hoppa_rotation/
Why?
If it's all ionizing, then it won't penetrate very deeply; not nearly enough to hit the theorized liquid water.
Even if we are talking heavy non-ionizing bombardment, which could penetrate fairly deeply, Europa is practically tidally locked in its orbit, just as Earth's moon is. I say practically, because even if the theorized rotation occurs:
http://pirlwww.lpl.arizona.edu/HIIPS/Publicatio
the rate is 1 foot of motion every 17 years. It would not be hard for a large population of somethings to keep the bulk of the moon between itself an Jupiter, even if by accident.
-- Terry
Unsustainable business models are a dime a dozen these days.
Salon has spent $80M, and has 50,000 subscribers.
That's a customer acqusition cost of $1,600 per customer.
Say they get their doubled subscribership numbers; that drops the per customer acquisition cost down to $800 per.
Effectively, this means that they would have to get $67 a monthly issue in order to recoup costs, if acquisition was for a period of 1 year, which is normally how these things are measured.
Let's be incredibly generous, and call it 5 years of acquisition. Even so, we are still talking over $13/month/60 issues.
Does anyone really believe that this is going to happen?
These people obviously do not understand cost accounting or cash flow. They may or may not be good journalists, but they certainly are *not* good businessmen.
-- Terry
So, is it just me, or have they failed once again to put the ejection under software control so the pig can have it's driver's detached and the hardware powered down, FS's unmounted, etc., before the thing disappears out from under the OS?
Way to go... we're back at the same place PCMCIA was back in 1994, yet again. 8-(.
-- Terry
"Configurability Considered Harmful"
Configurability is inversely proportional to supportability.
Configurability damages portability of employee skills, and makes it harder to hire people who can "hit the ground running".
Configurability damages portability of employee skills within a company.
Try this exercise: think of the last non-trivial problem you had. Now think about what it would take to talk your mother through fixing it on the telephone. Now think about what it would take to talk your mother through it on the telephone if she had an entirely different "skin", "theme", or window manager.
-- Terry
Easy... no boot code.
-- Terry
I can't even imagine the conversation that contains "Hey Bob, let's go eat!" "Great idea, Sam! But not Chinese, we had that yesterday... how 'bout we go for some British food?".
On the other hand, I can well imagine this triggering the need for biometric information... after all, after actually eating British food, you'd want to be able to restore your metabolism from the backup copy on your card.
-- Terry
...and Ellen Feiss for Drug Czar!
Had to be said.
-- Terry
What you can do with a Geo Metro is often more than people give it credit for being able to do.
One of my fond memories from my mis-spent you was taking my Chevy Sprint (the forerunner to the Geo Metro, which was basically a brand rename for the Sprint) up over El Monte. El Monte is basically a "pass" over the mountain from North Fork, in Ogden Valley, UT, through to Porcupine, which is by Logan, UT. Basically, it's 1100 feet of vertical ascent over about 15 miles, over rutted dirt roads, followed by 800 vertical feet of descent (we were going fishing at Porcupine resevouir, and going that way saves 2 hours of travel).
At the very top of the mountain was a group of "he-man, danger-man" types, all there in their extremely jacked-up 4-wheel-drive trucks, drinking beer, and congradulating themselves on being able to "go wheelin'" and get up to the top of the mountain, away from the riff-raff. Each truck has a minimum cost of $15,000, compared to the less than $5,000 for my Sprint (1980's; the eqivalent price for the same truck these days is $45,000+).
The looks on their faces as I came over the crest of the last hill to the top of the mountain, and drove by them with four passengers (Mike, Kike, and Scott), invading their "four-wheel-drive-only" space in my Chevy Sprint was *priceless*.
About the only thing you couldn't do with it was tow a boat or a trailer full of snowmobiles.
Oh yeah; the thing had a carbuerator, not fuel injection, and it got 52 MPG (standard transmission, no air conditioning)...
-- Terry
..."works similarly to to nicotine or birth control patches"...
Oh yeah, there's a warehouse mixup waiting to happen.
-- Terry
If you think this is bad... ...Their first offer was to port the "dancing paperclip" to FreeBSD, Linux, and Solaris.
-- Terry