1. A method of placing an order for an item comprising: under control of a client system, displaying information identifying the item; and in response to only a single action being performed, sending a request to order the item along with an identifier of a purchaser of the item to a server system; under control of a single-action ordering component of the server system, receiving the request; retrieving additional information previously stored for the purchaser identified by the identifier in the received request; and generating an order to purchase the requested item for the purchaser identified by the identifier in the received request using the retrieved additional information; and fulfilling the generated order to complete purchase of the item whereby the item is ordered without using a shopping cart ordering model.
Whew! Talk about a run-on sentence. Good thing legalese != English.
It'll be interesting to see how this works out. Normally in these kind of cases, it's a Big Boy going after someone without the resources to defend themselves. Barnes and Noble certainly has the wherewithall to bring this to court if their lawyers think they have a chance. However, I suspect that they'll just roll over and pay royalties to Amazon.
No company now would commit to a closed hardware strategy. It would cost them more than using commodity components. Just as importantly, it would commit them to a single source for support and parts.
As much as I'd like to believe that people don't buy closed hardware any more, I've actually seen a resurgence in closed hardware lately. It has been pushed by none other than our old friend, Sun Microsystems.
Examples:
Sun internal SCSI disks only work with Sun SCSI cards.
Only Sun external SCSI disks are "supported" by Sun. (I.e. if you use them and anything in the I/O system breaks, Sun can blame your disks and fail to offer support.
On many systems, only Sun memory may be installed despite complete compatibility with the DRAM types.
The deepest discounts on Sun equipment are only available if Sun becomes the official corporate supplier of all hardware.
I'm sure there are other examples, but the point is the same-- they want to lock you into a single hardware platform. This leaves very little leverage for negotiation when they drop the ball.
I totally agree that open hardware with well-defined, standard interfaces is the way to go from a customer's perspective. One of the reasons I have such high hopes for Linux is that the feature set is virtually the same across a variety of otherwise completely different hardware platforms.
What other examples of common closed hardware can you folks think of?
d) Open Source software becomes so prevalent that programmers merely download the source code for a package similar to what they want and adapt it to their needs rather than rebuilding the whole thing from scratch. Suits, er... Ironed T-Shirts everywhere hail a new era in programming where massive new applications can be built in a matter of days or weeks.
I'll admit to being a pessimist about most things, but from everything I've seen within the software/sysadmin industry for the past 10 years points to things continuing well into the forseeable future. (e.g. 5-10 years.)
Cases in point:
As development cycles have sped up, demands for more complex software have kept pace, nullifying any time savings. Feature/complexity demand is still on the rise and until this begins to stabilize, the supply of skilled developers will not catch up. (Developers just don't breed quickly enough.:-)
Complex software breaks in more ways than simple software, which increases the need for sysadmins even when normalized for expected software growth.
Now with that said, it would be foolish to think that the present trend will last forever. "Hope for the best, but plan for the worst" is a quote I heard recently which seems like exceptionally good advice for us all.
Things will be a LOT easier later on if we can each show some restraint and save some of the lucre we're getting these days. A good place to learn how is BYG Publishing's personal finance page. (I have no affiliation with them-- I just thought it was good, sound advice.) I'm sure a lot of other folks here will second the fact that we'll spend what we allow ourselves to spend irregardless of income.
Here's to hoping for more good times, and planning for bad!
Note that all of the above are hardware manufacturers, and the 99.9% uptime only applies to the hardware. If the hardware is up, and the OS is bluescreened, this still counts as "up" by the terms of these contracts.
Speaking of The Onion, once upon a time there was a "The Onion" channel on My Netscape linking back to The Onion site for individual stories. It was taken down at the request of The Onion.
This channel would have provided lots of traffic directed back to their site, more advertising dollars, etc. However, they didn't want their content linked in from anywhere else.
Does it provide a new method of producing high-quality software, or does it apply to only some programme categories?
I would argue that these are not necessarily exclusive. OSS certainly provides a new method of high-quality software-- there's no need to even argue this point with the audience here. However, it only applies to some types of development.
For example, an OS is ideally suited for OSS development since the design is well-defined (as opposed to bleeding-edge creative development), and stability is paramount. (Yes, I'd like everything to be as stable as an OS should be, but I have to concede that many development projects simply don't have the time to be so careful.)
As a counterexample, software which requires significant R&D for the development of new algorithms tends to be poorly suited to OSS. There would be no motivation to make the large up-front capital investment without assurances of getting a return on one's investment. These programs tend to be very expensive and with limited distribution. (E.g. some specialized sound/image processing software, Mathematica, etc.)
By far the bulk of software development could easily go OSS. How many of us have spent days/weeks/months developing custom in-house software to solve some specialized problem? These generally do not involve any leaps in technology-- they just involve putting the right pieces together. There wouldn't be any harm in releasing this kind of software to the world, yet most companies refrain from it, simply because they are used to everything being closed-source.
What does it mean for the economics of cyberspace?
I don't really see OSS significantly reducing the number of programmers needed. Right now the software market is so massively labor-limited that many useful leading-edge projects are stalled due to lack of people to build them. At the same time, there are hundreds of thousands of programmers happily reinventing the same wheel over and over again. How many of us have switched companies, just to find ourselves re-developing the same programs in a new environment? Valuable developer time is being spent building up infrastructure that should be readily available.
If instead, people were making their internal software available to anyone, much of the time spent reinventing the same thing would instead go towards important leading-edge projects. There are enough of these to keep the current software workforce busy well into the forseeable future. Developers: what percentage of your work is spent on the leading edge? What percentage seems likely to have already been built by someone, somewhere? (For me, it's about 20% leading, 80% redos and I think I get more leading edge stuff than usual...)
How can free software coexist with hard-nosed profit-maximising companies?
If there is no cost to distributing software, then it doesn't affect profits. Companies tend to be very conservative in spreading software, just because it's always been this way-- not because they've made any reasoned choice. Internal business applications take an unusually large amount of development time, and cost the company nothing to redistribute once they're done. (They had to build it anyway.) It could even turn into a revenue generator should they sell support for a product that they would not have released otherwise. The only concern would be possible liability, but all the OSS licenses I've seen cover this.
Flagship products will be the last to go Open Source, and in some cases will never go this way. However, these tend to be a small portion of the actual development taking place inside a company. I'd settle for seeing the other 80-90% Open-Sourced.
What is the maxium amount of damages that can be claimed in small claims court?
If you are in California, the maxium is $2,500. Other states are probably similar. Keep in mind that in small claims court, you represent youself and cannot have a lawyer present.
Thanks for the hint-- I'd forgotten about Concentric. Their pricing for DSL isn't too bad. Concentric charges $70 for 384K/128K ADSL versus $80 from PacBell for a similar level of service (4 IPs versus 5 IPs).
However, Concentric does not offer the low-cost single IP version. I was planning to go for at least 2 IPs anyhow-- one of them nicely firewalled for full-time web serving and one for my Windows box for game-playing.
Thanks for the info! I knew Slashdot wouldn't let me down!:-)
The Slashdot crowd remains one of the most technically-savvy discussion groups around, and this seems like an opportune time to ask since the discussion centers around high-speed connectivity.
Does anyone have experience with DSL in the Silicon Valley area? Some of the providers I've looked at have been:
Do any of you know of any web sites which have customer reviews of DSL in this area? Do any of you have personal experience with these (or other) ISPs? (I'm curious how well they are connected, and how well they tolerate those of us who use a real operating system.)
Although this was an interesting article, it was replete with factual errors:
DSL service is inherently more secure than using a cable modem because DSL provides a dedicated connection over your existing telephone line. A cable modem is more susceptible to hackers since it operates on a shared system, much like an old-fashioned party line
He used this in a context which was somewhat ambiguous. This was presented as an example of "noisy ammunition for their PR battle", but he never explicity stated whether he felt this was true or not. I would argue that the security problems one is likely to face from one's neighbors pale in comparison to those one faces from the world at large. Either way leaving one's computer connected to the Internet without a firewall is begging for trouble.
I can download Netscape Communicator 4.5-- which resides on a server in San Jose -- in less than 4 minutes.
I wonder if he checked the IP address he was downloading from... If Netscape's "tdns" system was working properly, the IP he got was for the East Coast Netscape colocation facility in Pennsauken, N.J. rather than Mountain View.
The company is fundamentally a West Coast ISP with delusions of grandeur. Specifically, Concentric doesn't actually have any employees in Boston. Instead, it contracts with a company called Covad to do all the grunt work.
I'm not certain, but I believe that Concentric uses Covad "to do all the grunt work" even on the West Coast.
The sleek "RISC" microprocessors from companies like Sun Microsystems and MIPS lost out to Intel's technologically inferior Pentium chips. Ultimately, this battle will be won and lost on mundane issues like price and quality of service.
I guess he missed the whole evolution of Merced from HP's PA-RISC chip.
Despite these problems, it was definately a thought-provoking article. I'm looking forward to the discussion here on Slashdot-- I bet that it provides better information than the original article!
This has been tried before-- and sometimes it works. Star Control II scoured the internet (such as it was in those days) for talented MOD programmers. As a result they had some of the most creative music of their time.
In more recent history, Total Annihilation recruited scores of people from the Internet to help design maps. Some of them are quite good.
It's not exactly Open Source, but it's still in the spirit of large-scale collaborative development-- and it often pays off.
The "claims" of a patent provide the legal definition of the scope of the patent. In order to infringe upon a patent, ALL of the claims must be violated. Needless to say, it's in the inventor's best interest to make the claims as vague as possible. However, the Patent and Trademark office has historically been good about denying overly broad patents.
If even one portion of the claims is not duplicated exactly as described there is no patent infringement.
The claims for this patent:
A method of human identification to facilitate electronic sale transactions comprising the steps of: providing identity information about a purchaser on a storage medium, providing skin marking invisible ink, applying said invisible ink to an appendage of said purchaser to form a tattoo on said purchaser, storing characteristics about said tattoo on said storage medium to form stored characteristics about said tattoo, and linking said identity information about said purchaser to said stored characteristics about said tattoo.
A method of human identification as in claim 1 further comprising the steps of: providing financial information about said purchaser on said storage medium, and linking said financial information about said purchaser to said identity information about said purchaser.
A method of human identification as in claim 1 further comprising the steps of: providing financial information about said purchaser on said storage medium, and linking said financial information about said purchaser to said stored characteristics about said tattoo.
A method of human identification as in claim 2 further comprising the steps of: scanning said tattoo on said purchaser with a scanning device to obtain scanned characteristics of said tattoo on said purchaser, comparing said scanned characteristics to characteristics about other tattoos found on said storage medium in order to determine if said scanned characteristics match said stored characteristics about said tattoo on said purchaser stored on said storage medium, and only if said scanned characteristics match said stored characteristics then providing said linked identity information in order to verify the identity of said purchaser.
A method of human identification as in claim 4 further comprising the step of: only if said scanned characteristics match said stored characteristics then providing said linked financial information in order to verify that said purchaser has sufficient funds to consummate a sales transaction.
A method of human identification as in claim 5 further comprising the step of: only upon verification that said purchaser has sufficient funds to consummate said sales transaction then modifying said stored financial information about said purchaser to reflect a consummated sales transaction.
A method of human identification as in claim 3 further comprising the steps of: scanning said tattoo on said purchaser with a scanning device to obtain scanned characteristics about said tattoo, comparing said scanned characteristics to characteristics about tattoos stored found on said storage medium in order to determine if said scanned characteristics match said stored characteristics of said tattoo on said purchaser stored on said storage medium, and only if said scanned characteristics match said stored characteristics then providing said linked financial information in order to verify that said purchaser has sufficient funds to consummate a sales transaction
A method of human identification as in claim 7 further comprising the step of: only upon verification that said purchaser has sufficient funds to consummate said sales transaction then modifying said stored financial information about said purchaser to reflect a consummated sales transaction.
A method of human identification as in claim 5 further comprising the steps of: providing financial information about a seller on said storage medium, and only upon verification that said purchaser has sufficient funds to consummate said sales transaction then modifying said stored financial information about said seller to reflect a consummated sales transaction.
A method of human identification as in claim 7 further comprising the steps of: providing financial information about a seller on said storage medium, and only upon verification that said purchaser has sufficient funds to consummate said sales transaction then modifying said stored financial information about said seller to reflect a consummated sales transaction.
So for example, if one were to use visible ink instead of invisible, it would not be patent infringement. Or, if one were not using financial information, then it would not infringe on this patent.
Of course, I am not a lawyer and cannot provide specific advice on patent issues. So the above may be patently false. (No pun intended.;->)
This is not a particularly surprising inclusion. Media companies put all kinds of language in their standard contracts assigning ownership to The Company.
For example, cartoon companies routinely have the creator sign away rights to the characters and style, allowing The Company to continue to publish the strip even if the creator quits. I'm sure some of us remember how bitter Bill Watterson, creator of Calvin and Hobbes was...
Unfortunately, people go ahead and sign these contracts since their only other alternative is to be forever consigned to oblivion. Without a publisher, these artists would never have been noticed.
With the advent of cheap (or even free) web publishing, this is all changing. Anyone can make their art readily available-- witness the popularity (for both consumer and supplier) of mp3.com, User Friendly, and others.
The Internet is starting to force media publishers back into the role of promoters rather than controllers, and they are understandably upset and/or scared. Consumers will still need publishers to help pick out what is good, but they no longer need be restricted to only what the publishers wish to promote, so the publisher will no longer have the same power over the artists they once had.
1.Put 5x (Numbers are Variables here: Don't complain about them!) as many moderators into the system.
2.When you have moderator access, only display moderate controls on say 20% of the comments.
I'm not convinced that moderator dilution is the answer here. I would expect that by spreading out the pool of moderators, Slashdot would see even more problems with badly moderated posts. What we want is instead a smaller pool of people. but people who moderate more thoughtfully.
But how to distinguish thoughtful moderators from irresponsible ones? I like the "karma" system since it provides more personal accountability. I don't think it has been in use long enough to really show its potential, but people who regularly post articles that other people like should have a chance to participate more in the moderation process
Here are some thoughts:
Create a fixed-size "pool" of moderation points (or perhaps make it a percentage of the total number of articles?)
Dole these out to a predetermined number of people with high-karma folks getting more moderation points. (Basically option 3 as mentioned in the article.)
Create a system where a very few of the very best people on slashdot can serve as uber-moderators. These people would have the power to permanently ban people from participating in moderation. (Probably best to choose these people by hand.)
With that said, I feel that the present system is working VERY well. It is extremely rare that I see a post moderated up which didn't really deserve it. Even better, by setting high thresholds, I can browse very quickly and still get some valuable insights on the articles.
Back in the Good Old Days (e.g. before about 1990), one could not patent software, according to the book "Patent It Yourself", third edition. I believe the logic was that software was simply too flexible to be reliably patented-- it was prohibitively difficult to verify that it was indeed a new unique process.
I'm not sure when the change was made to allow software patents, but I believe we can all now see the wisdom in disallowing them before. The Patent and Trademark office is clearly over its head when things like this are granted patent status. How hard is it for them to adjust their bias towards rejecting patents?
Keep in mind that the whole idea behind patents is fundamentally the same as Open Source Software. In exchange for a limited monopoly, people get full disclosure of how the patented process works. This knowledge can then be used to stimulate others' creativity, leading to new processes that might never have been created had the original process been kept secret.
Patents are a Good Thing, though it's sometimes hard not to lose sight of this when patents on widespread and overly generic processes are granted.
My experience with the RepKover binding has been decidedly mixed. My original Programmning Perl book separated from the cover after only three months, though admittedly I used it very often. My Practical C Programming book separated in the same way after a similar length of time. (The pages stay bound, but the cloth attached to the back cover came loose.)
On the other hand, the new Programming Perl book uses a standard binding, and it has held up for two years under extremely heavy use. The cover is scuffed and folded, and the lamination is peeling at the corners, but the binding is still solid.
Personally, I'd love to see some of the most frequently used O'Reilly books in hardcover. I know I'd buy Programming Perl in hardcover. I'd also buy Essential System Administration in hardcover. I'm willing to pay about 20% more for hardcover versus softcover-- about $10-15 for most O'Reilly books.)
Has anyone else had problems with the bindings on heavily used O'Reilly books? Who else is willing to pay for hardcover? What books would you all like to see in hardcover?
SGI makes a number of different Visual Workstations. Some of them were designed with Linux in mind, some were designed for NT but work well with Linux, and still others don't work with Linux.
For example, the 1400 systems were designed from the ground up for Linux. However, the 320 I tested Linux on had problems since its Qlogic SCSI cards were not supported at that time.
Check SGI's Linux on Intel collection for more information. (It looks like they fixed the Qlogic support problem-- see "BROKEN STUFF").
E*Trade is no longer accepting indications of interest for the IPO.
Anyone who feels left out may revel in the knowledge that if it goes anything like the 1800flowers IPO, it will be a much better bargain in 3-6 months anyhow.
SGI can help Linux out a great deal by providing dedicated hardware which supports some Linux nicities. First on my list would be support for full configuration via the serial port, similar to the SPARC systems from Sun.
Of course, it's also gratifying to see another vendor promising support for IA-64. This is enough of a radical departure from IA-32 that compiler support will be very tricky. IA-64 requires the compiler to properly order the instructions in order to gain any performance over IA-32. Without a very thorough optimizer, Linux on IA-64 would likely run slower than on IA-32. Since SGI is banking on high performance, it seems likely they'll contribute to EGCS or GCC.
I'm also looking forward to using XFS on linux. Imagine a logging filesystem with no 2GB file size limits...
Here's a gem that will help save you lots of needless image downloading. Check out JunkBuster. It's a simple HTTP proxy that filters out ads in a manner that you configure. Source code available under the General Public License.
Lots of people have "blockfiles" out there, and the JunkBuster FAQ has a link to an Altavista query to find one.
Of course, I would NEVER use this against Slashdot, even though those !@$*% Adfu ads never close their TCP connections.
PC hardware failure problem spots in my experience have been:
CD-ROMs (really flaky)
RAM
SCSI cards
Motherboards
Where to spend the extra money?
Buy Plextor CD-ROMs. Never seen one fail. I'm sure they do, but not real often. Nobody else comes close.
RAM. Buy the highest-quality name-brand DIMMS you can find. Fujitsu, Toshiba, and NEC all make good SIMMs. There is a LOT of crap floating around the market these days.
SCSI cards. These things just like to be flaky. Even the high-quality name brands.
Motherboards. ALWAYS buy a well-known brand. Asus, Tyan, Abit are all very popular.
Linux software RAID is not quite ready for the kind of reliability people want from RAID. They'll get there-- but they aren't there yet.
The documentation that comes with raidtools explictitly states that this is considered beta.
Most people prefer a slightly more robust environment for their Oracle databases.
'man mkraid'
BUGS Probably many.
With that said, I'm running RAID-5 and have had few problems with it. I strongly recommend that one become familiar with its quirks before trusting your data to it.
I suspect that Lucas wants people to see it in the theaters first. It will certainly look and sound a lot better there than on our pitiful little PCs.
Once Lucas releases the "real" trailer, I'm sure there will be a large number of well-equipped mirror sites to retrieve it from. Some of them will have pipes big enough to avoid getting Slashdotted.
And then we can tell countingdown.com to go take a flying leap. They really shouldn't tease us like that...
What benchmark software do you recommend? I've punished the living s**t out of a linux box (dual proc, 512MB RAM) with webstone 2.5 and the 100BaseT NIC (EEPro) was the bottleneck for static pages. Funneling half the traffic through a 3c905 (not 905B) boosted throughput quite a bit.
Here are the results from the best run versus Apache (performance goes down with more clients...)
WEBSTONE 2.5b3 results: Total number of clients: 200 Test time: 2 minutes Server connection rate: 684.13 connections/sec Server error rate: 0.00 err/sec Server thruput: 103.30 Mbit/sec Little's Load Factor: 198.46 Average response time: 0.290 sec Error Level: 0.00 % Average client thruput: 0.52 Mbit/sec Sum of client response times: 23815.70 sec Total number of pages read: 82096
Note that these were for the standard Webstone distribution for static file benchmarks. I have no dynamic results. I also don't have a complete benchmark disclosure ready-- this is just a quick test I ran a while back versus linux 2.2.0
You learned your lesson the same way I did. I always compile at least one monolithic (no modules) kernel and keep it in lilo.conf forever. That way my system will be bootable even if I hose up my modules. Again.;->
Slashdot Mirror
It is as bad as it seems.
Patent number 5,960,411-- you can read the entire patent document from this link.
Claim 1 (of 26):
1. A method of placing an order for an item comprising: under control of a client system, displaying information identifying the item; and in response to only a single action being performed, sending a request to order the item along with an identifier of a purchaser of the item to a server system; under control of a single-action ordering component of the server system, receiving the request; retrieving additional information previously stored for the purchaser identified by the identifier in the received request; and generating an order to purchase the requested item for the purchaser identified by the identifier in the received request using the retrieved additional information; and fulfilling the generated order to complete purchase of the item whereby the item is ordered without using a shopping cart ordering model.
Whew! Talk about a run-on sentence. Good thing legalese != English.
It'll be interesting to see how this works out. Normally in these kind of cases, it's a Big Boy going after someone without the resources to defend themselves. Barnes and Noble certainly has the wherewithall to bring this to court if their lawyers think they have a chance. However, I suspect that they'll just roll over and pay royalties to Amazon.
Very sad.
No company now would commit to a closed hardware strategy. It would cost them more than using commodity components. Just as importantly, it would commit them to a single source for support and parts.
As much as I'd like to believe that people don't buy closed hardware any more, I've actually seen a resurgence in closed hardware lately. It has been pushed by none other than our old friend, Sun Microsystems.
Examples:
I'm sure there are other examples, but the point is the same-- they want to lock you into a single hardware platform. This leaves very little leverage for negotiation when they drop the ball.
I totally agree that open hardware with well-defined, standard interfaces is the way to go from a customer's perspective. One of the reasons I have such high hopes for Linux is that the feature set is virtually the same across a variety of otherwise completely different hardware platforms.
What other examples of common closed hardware can you folks think of?
I'm personally hoping for:
d) Open Source software becomes so prevalent that programmers merely download the source code for a package similar to what they want and adapt it to their needs rather than rebuilding the whole thing from scratch. Suits, er... Ironed T-Shirts everywhere hail a new era in programming where massive new applications can be built in a matter of days or weeks.
I'll admit to being a pessimist about most things, but from everything I've seen within the software/sysadmin industry for the past 10 years points to things continuing well into the forseeable future. (e.g. 5-10 years.)
Cases in point:
As development cycles have sped up, demands for more complex software have kept pace, nullifying any time savings. Feature/complexity demand is still on the rise and until this begins to stabilize, the supply of skilled developers will not catch up. (Developers just don't breed quickly enough. :-)
Complex software breaks in more ways than simple software, which increases the need for sysadmins even when normalized for expected software growth.
Now with that said, it would be foolish to think that the present trend will last forever. "Hope for the best, but plan for the worst" is a quote I heard recently which seems like exceptionally good advice for us all.
Things will be a LOT easier later on if we can each show some restraint and save some of the lucre we're getting these days. A good place to learn how is BYG Publishing's personal finance page. (I have no affiliation with them-- I just thought it was good, sound advice.) I'm sure a lot of other folks here will second the fact that we'll spend what we allow ourselves to spend irregardless of income.
Here's to hoping for more good times, and planning for bad!
Note that all of the above are hardware manufacturers, and the 99.9% uptime only applies to the hardware. If the hardware is up, and the OS is bluescreened, this still counts as "up" by the terms of these contracts.
Speaking of The Onion, once upon a time there was a "The Onion" channel on My Netscape linking back to The Onion site for individual stories. It was taken down at the request of The Onion.
This channel would have provided lots of traffic directed back to their site, more advertising dollars, etc. However, they didn't want their content linked in from anywhere else.
They just don't get it.
Does it provide a new method of producing high-quality software, or does it apply to only some programme categories?
I would argue that these are not necessarily exclusive. OSS certainly provides a new method of high-quality software-- there's no need to even argue this point with the audience here. However, it only applies to some types of development.
For example, an OS is ideally suited for OSS development since the design is well-defined (as opposed to bleeding-edge creative development), and stability is paramount. (Yes, I'd like everything to be as stable as an OS should be, but I have to concede that many development projects simply don't have the time to be so careful.)
As a counterexample, software which requires significant R&D for the development of new algorithms tends to be poorly suited to OSS. There would be no motivation to make the large up-front capital investment without assurances of getting a return on one's investment. These programs tend to be very expensive and with limited distribution. (E.g. some specialized sound/image processing software, Mathematica, etc.)
By far the bulk of software development could easily go OSS. How many of us have spent days/weeks/months developing custom in-house software to solve some specialized problem? These generally do not involve any leaps in technology-- they just involve putting the right pieces together. There wouldn't be any harm in releasing this kind of software to the world, yet most companies refrain from it, simply because they are used to everything being closed-source.
What does it mean for the economics of cyberspace?
I don't really see OSS significantly reducing the number of programmers needed. Right now the software market is so massively labor-limited that many useful leading-edge projects are stalled due to lack of people to build them. At the same time, there are hundreds of thousands of programmers happily reinventing the same wheel over and over again. How many of us have switched companies, just to find ourselves re-developing the same programs in a new environment? Valuable developer time is being spent building up infrastructure that should be readily available.
If instead, people were making their internal software available to anyone, much of the time spent reinventing the same thing would instead go towards important leading-edge projects. There are enough of these to keep the current software workforce busy well into the forseeable future. Developers: what percentage of your work is spent on the leading edge? What percentage seems likely to have already been built by someone, somewhere? (For me, it's about 20% leading, 80% redos and I think I get more leading edge stuff than usual...)
How can free software coexist with hard-nosed profit-maximising companies?
If there is no cost to distributing software, then it doesn't affect profits. Companies tend to be very conservative in spreading software, just because it's always been this way-- not because they've made any reasoned choice. Internal business applications take an unusually large amount of development time, and cost the company nothing to redistribute once they're done. (They had to build it anyway.) It could even turn into a revenue generator should they sell support for a product that they would not have released otherwise. The only concern would be possible liability, but all the OSS licenses I've seen cover this.
Flagship products will be the last to go Open Source, and in some cases will never go this way. However, these tend to be a small portion of the actual development taking place inside a company. I'd settle for seeing the other 80-90% Open-Sourced.
What is the maxium amount of damages that can be claimed in small claims court?
If you are in California, the maxium is $2,500. Other states are probably similar. Keep in mind that in small claims court, you represent youself and cannot have a lawyer present.
Thanks for the hint-- I'd forgotten about Concentric. Their pricing for DSL isn't too bad. Concentric charges $70 for 384K/128K ADSL versus $80 from PacBell for a similar level of service (4 IPs versus 5 IPs).
However, Concentric does not offer the low-cost single IP version. I was planning to go for at least 2 IPs anyhow-- one of them nicely firewalled for full-time web serving and one for my Windows box for game-playing.
Thanks for the info! I knew Slashdot wouldn't let me down! :-)
The Slashdot crowd remains one of the most technically-savvy discussion groups around, and this seems like an opportune time to ask since the discussion centers around high-speed connectivity.
Does anyone have experience with DSL in the Silicon Valley area? Some of the providers I've looked at have been:
Do any of you know of any web sites which have customer reviews of DSL in this area? Do any of you have personal experience with these (or other) ISPs? (I'm curious how well they are connected, and how well they tolerate those of us who use a real operating system.)
Although this was an interesting article, it was replete with factual errors:
DSL service is inherently more secure than using a cable modem because DSL provides a dedicated connection over your existing telephone line. A cable modem is more susceptible to hackers since it operates on a shared system, much like an old-fashioned party line
He used this in a context which was somewhat ambiguous. This was presented as an example of "noisy ammunition for their PR battle", but he never explicity stated whether he felt this was true or not. I would argue that the security problems one is likely to face from one's neighbors pale in comparison to those one faces from the world at large. Either way leaving one's computer connected to the Internet without a firewall is begging for trouble.
I can download Netscape Communicator 4.5-- which resides on a server in San Jose -- in less than 4 minutes.
I wonder if he checked the IP address he was downloading from... If Netscape's "tdns" system was working properly, the IP he got was for the East Coast Netscape colocation facility in Pennsauken, N.J. rather than Mountain View.
The company is fundamentally a West Coast ISP with delusions of grandeur. Specifically, Concentric doesn't actually have any employees in Boston. Instead, it contracts with a company called Covad to do all the grunt work.
I'm not certain, but I believe that Concentric uses Covad "to do all the grunt work" even on the West Coast.
The sleek "RISC" microprocessors from companies like Sun Microsystems and MIPS lost out to Intel's technologically inferior Pentium chips. Ultimately, this battle will be won and lost on mundane issues like price and quality of service.
I guess he missed the whole evolution of Merced from HP's PA-RISC chip.
Despite these problems, it was definately a thought-provoking article. I'm looking forward to the discussion here on Slashdot-- I bet that it provides better information than the original article!
This has been tried before-- and sometimes it works. Star Control II scoured the internet (such as it was in those days) for talented MOD programmers. As a result they had some of the most creative music of their time.
In more recent history, Total Annihilation recruited scores of people from the Internet to help design maps. Some of them are quite good.
It's not exactly Open Source, but it's still in the spirit of large-scale collaborative development-- and it often pays off.
The "claims" of a patent provide the legal definition of the scope of the patent. In order to infringe upon a patent, ALL of the claims must be violated. Needless to say, it's in the inventor's best interest to make the claims as vague as possible. However, the Patent and Trademark office has historically been good about denying overly broad patents.
If even one portion of the claims is not duplicated exactly as described there is no patent infringement.
The claims for this patent:
providing identity information about a purchaser on a storage medium, providing skin marking invisible ink, applying said invisible ink to an appendage of said purchaser to form a tattoo on said purchaser, storing characteristics about said tattoo on said storage medium to form stored characteristics about said tattoo, and linking said identity information about said purchaser to said stored characteristics about said tattoo.
So for example, if one were to use visible ink instead of invisible, it would not be patent infringement. Or, if one were not using financial information, then it would not infringe on this patent.
Of course, I am not a lawyer and cannot provide specific advice on patent issues. So the above may be patently false. (No pun intended. ;->)
This is not a particularly surprising inclusion. Media companies put all kinds of language in their standard contracts assigning ownership to The Company.
For example, cartoon companies routinely have the creator sign away rights to the characters and style, allowing The Company to continue to publish the strip even if the creator quits. I'm sure some of us remember how bitter Bill Watterson, creator of Calvin and Hobbes was...
Unfortunately, people go ahead and sign these contracts since their only other alternative is to be forever consigned to oblivion. Without a publisher, these artists would never have been noticed.
With the advent of cheap (or even free) web publishing, this is all changing. Anyone can make their art readily available-- witness the popularity (for both consumer and supplier) of mp3.com, User Friendly, and others.
The Internet is starting to force media publishers back into the role of promoters rather than controllers, and they are understandably upset and/or scared. Consumers will still need publishers to help pick out what is good, but they no longer need be restricted to only what the publishers wish to promote, so the publisher will no longer have the same power over the artists they once had.
1.Put 5x (Numbers are Variables here: Don't complain about them!) as many moderators into the system.
2.When you have moderator access, only display moderate controls on say 20% of the comments.
I'm not convinced that moderator dilution is the answer here. I would expect that by spreading out the pool of moderators, Slashdot would see even more problems with badly moderated posts. What we want is instead a smaller pool of people. but people who moderate more thoughtfully.
But how to distinguish thoughtful moderators from irresponsible ones? I like the "karma" system since it provides more personal accountability. I don't think it has been in use long enough to really show its potential, but people who regularly post articles that other people like should have a chance to participate more in the moderation process
Here are some thoughts:
With that said, I feel that the present system is working VERY well. It is extremely rare that I see a post moderated up which didn't really deserve it. Even better, by setting high thresholds, I can browse very quickly and still get some valuable insights on the articles.
Good work, guys!
Back in the Good Old Days (e.g. before about 1990), one could not patent software, according to the book "Patent It Yourself", third edition. I believe the logic was that software was simply too flexible to be reliably patented-- it was prohibitively difficult to verify that it was indeed a new unique process.
I'm not sure when the change was made to allow software patents, but I believe we can all now see the wisdom in disallowing them before. The Patent and Trademark office is clearly over its head when things like this are granted patent status. How hard is it for them to adjust their bias towards rejecting patents?
Keep in mind that the whole idea behind patents is fundamentally the same as Open Source Software. In exchange for a limited monopoly, people get full disclosure of how the patented process works. This knowledge can then be used to stimulate others' creativity, leading to new processes that might never have been created had the original process been kept secret.
Patents are a Good Thing, though it's sometimes hard not to lose sight of this when patents on widespread and overly generic processes are granted.
My experience with the RepKover binding has been decidedly mixed. My original Programmning Perl book separated from the cover after only three months, though admittedly I used it very often. My Practical C Programming book separated in the same way after a similar length of time. (The pages stay bound, but the cloth attached to the back cover came loose.)
On the other hand, the new Programming Perl book uses a standard binding, and it has held up for two years under extremely heavy use. The cover is scuffed and folded, and the lamination is peeling at the corners, but the binding is still solid.
Personally, I'd love to see some of the most frequently used O'Reilly books in hardcover. I know I'd buy Programming Perl in hardcover. I'd also buy Essential System Administration in hardcover. I'm willing to pay about 20% more for hardcover versus softcover-- about $10-15 for most O'Reilly books.)
Has anyone else had problems with the bindings on heavily used O'Reilly books? Who else is willing to pay for hardcover? What books would you all like to see in hardcover?
SGI makes a number of different Visual Workstations. Some of them were designed with Linux in mind, some were designed for NT but work well with Linux, and still others don't work with Linux.
For example, the 1400 systems were designed from the ground up for Linux. However, the 320 I tested Linux on had problems since its Qlogic SCSI cards were not supported at that time.
Check SGI's Linux on Intel collection for more information. (It looks like they fixed the Qlogic support problem-- see "BROKEN STUFF").
E*Trade is no longer accepting indications of interest for the IPO.
Anyone who feels left out may revel in the knowledge that if it goes anything like the 1800flowers IPO, it will be a much better bargain in 3-6 months anyhow.
SGI can help Linux out a great deal by providing dedicated hardware which supports some Linux nicities. First on my list would be support for full configuration via the serial port, similar to the SPARC systems from Sun.
Of course, it's also gratifying to see another vendor promising support for IA-64. This is enough of a radical departure from IA-32 that compiler support will be very tricky. IA-64 requires the compiler to properly order the instructions in order to gain any performance over IA-32. Without a very thorough optimizer, Linux on IA-64 would likely run slower than on IA-32. Since SGI is banking on high performance, it seems likely they'll contribute to EGCS or GCC.
I'm also looking forward to using XFS on linux. Imagine a logging filesystem with no 2GB file size limits...
Here's a gem that will help save you lots of needless image downloading. Check out JunkBuster. It's a simple HTTP proxy that filters out ads in a manner that you configure. Source code available under the General Public License.
Lots of people have "blockfiles" out there, and the JunkBuster FAQ has a link to an Altavista query to find one.
Of course, I would NEVER use this against Slashdot, even though those !@$*% Adfu ads never close their TCP connections.
PC hardware failure problem spots in my experience have been:
Where to spend the extra money?
Linux software RAID is not quite ready for the kind of reliability people want from RAID. They'll get there-- but they aren't there yet.
The documentation that comes with raidtools explictitly states that this is considered beta.
Most people prefer a slightly more robust environment for their Oracle databases.
'man mkraid'
BUGS
Probably many.
With that said, I'm running RAID-5 and have had few problems with it. I strongly recommend that one become familiar with its quirks before trusting your data to it.
I suspect that Lucas wants people to see it in the theaters first. It will certainly look and sound a lot better there than on our pitiful little PCs.
Once Lucas releases the "real" trailer, I'm sure there will be a large number of well-equipped mirror sites to retrieve it from. Some of them will have pipes big enough to avoid getting Slashdotted.
And then we can tell countingdown.com to go take a flying leap. They really shouldn't tease us like that...
What benchmark software do you recommend? I've punished the living s**t out of a linux box (dual proc, 512MB RAM) with webstone 2.5 and the 100BaseT NIC (EEPro) was the bottleneck for static pages. Funneling half the traffic through a 3c905 (not 905B) boosted throughput quite a bit.
Here are the results from the best run versus Apache (performance goes down with more clients...)
WEBSTONE 2.5b3 results:
Total number of clients: 200
Test time: 2 minutes
Server connection rate: 684.13 connections/sec
Server error rate: 0.00 err/sec
Server thruput: 103.30 Mbit/sec
Little's Load Factor: 198.46
Average response time: 0.290 sec
Error Level: 0.00 %
Average client thruput: 0.52 Mbit/sec
Sum of client response times: 23815.70 sec
Total number of pages read: 82096
Note that these were for the standard Webstone distribution for static file benchmarks. I have no dynamic results. I also don't have a complete benchmark disclosure ready-- this is just a quick test I ran a while back versus linux 2.2.0
You learned your lesson the same way I did. I always compile at least one monolithic (no modules) kernel and keep it in lilo.conf forever. That way my system will be bootable even if I hose up my modules. Again. ;->