It would take a book to answer that, but here are the key ingredients:
sudo - Nobody should have root on any workstations. Sysadmins should have appropriately tailored sudo rights. If you have heavy-duty applications, the application administrators should be different from the platform admins. Naturally, the vast majority of boxes will never be logged into.
snmpd - Your central monitoring server can constantly sweep the workstations and collect memory, load and other stats. Proactively solve problems before a user notices.
ssh - With appropriate keys, it can be used by automated scripts to collect info or change configuration on all the workstations.
rsync - Rides on top of ssh to distribute new software.
remote re-imaging - there are several ways to automate this. If a workstation gets utterly hosed, it's not worth troubleshooting. Sorry if that sound un-unixy. You want to easily re-image without user intervention. Ideally, you want the hell-desk to be able to do this from a web page. This is a good reason to mount home directory over NFS, but I know you're asking about the un-nfs solutions.
strong central database - your 6000 machines won't be uniform. You'll have different ages, models, configurations, OS levels. The database keeps track of what build of every software package is on each machine, hw configurations, and history. Without this you're flying blind.
test pool and pilots - You need a lab with representative workstations to test new software on. Then roll out to a small subset of the population - 'pilot' users who are good at reporting problems. When that works, you can roll out to the whole machine population. That database keeps track of the rollout state so if the script is interrupted partway through, you know where to resume.
In real life, you don't get to plan installations like this and make them perfect. You inherit what's there and try to improve it.
SQL Server 7 does hold the highest scores for standardized database benchmarks. Nothing that runs on Linux can stand up to it, so don't even try:)
I guess you're referring to MS's impressive performance on TPC benchmarks. I don't mean to detract from Microsoft's accomplishment in setting new speed records, I think these scores have less linkage to reality than MS suggests. These results have little relevance to the average enterprise - I don't think the City of Virginia Beach needs a 24-way SMP machine to process their parking tickets or whatever. The basic reasons linux doesn't show up on these benchmarks are lack of scalable SMP and lack of funding to participate. These results have little relevance to the high end, because people who architect this stuff are quite conservative and prefer proven platforms like Sun E-series, HP's, Sequents. Neither Linux nor NT really has anything to offer this market. Lower cost per transaction? Great, until the first avoidable outage. Then the machine will eat all its savings in 30 seconds.
This goal of monopoly destruction is well worth a temporarily slightly less than optimal solution to a particular problem.
This seems to be a fundamental difference in thinking that pops up on slashdot in threads like these. The opposite of your viewpoint is the one that states "Use the right tool for the job." The split could be called idealistic vs. pragmatic. But this doesn't mean the idealist is impractical. The idealist wants what's best for him in the long run while the pragmatist seeks to maximize pleasure (or more likely minimize pain) today. Personally, I agree with you. Every time I see the "right tool" idea advocated, I want to respond, "where would you be today if people had always followed that advice?" Chamberlain did the pragmatic thing by appeasing Hitler, achieved "peace in our time" and was a temporary hero. Now his memory is a disgraceful one.
Please don't take the following as a description of how SDMI works. Rather, it's a skeletal frame to show that the 'trusted hardware' concept is feasible.
Bob downloads a protected song from Alice, the content owner. The song is encrypted and Bob doesn't have the key. To him, the song is just random bits. Bob would like to play the song on a software player. He can't.
Bob transfers the song to a hardware Player, which has been approved by Alice. The Player decrypts and plays the song when Bob tells it to.
The key storage, decryption, and rights management are performed on a single chip in the Player. Bob would like to extract the keys. He can't. Bob would like to intercept the decrypted audio before rights management decisions are made on it. He can't.
...there is no way SDMI can tell the difference between a hardware player and a software player.
The issue isn't really hardware vs software. It's 'trusted host' vs 'anyone else'. Alice trusts the Player because she thinks Bob can't take apart a chip.
What they are trying to do, decrypt something securely on someone else's system, can't be done without control of the hardware.
Given sufficient resources in Bob's hands, this is true. But there's only one hacker who can reverse engineer a VLSI chip for a great many who can reverse engineer a program. If tamper-resistance is built into the chip, as the NSA did with Clipper, reverse engineering could become enormously expensive and hard.
one of them adds up the financial report every month on his pocket calculator, because he doesn't believe Excel will add numbers right.
I worked on a billion-dollar project where all accounting was done by a huge interlinked maze of Excel spreadsheets. Discrepancies of several hundred dollars were common. The rule was, any gap less than $1k is not worth auditing.
Yes, BSD may have used this partition type before Thinkpads (although Thinkpads have had this kind of suspend feature for a long, long time).
Apparently, thinkpads used partition type 160 (a0) until now. FreeBSD used 165 (a5).
But messing around with its BIOS and suspend feature would be a major risk for IBM; I can understand that they don't want to do that.
Messing around is exactly what IBM did. Instead of sticking with the previously used partition type, they changed to one that conflicts with another known use.
In such cases, realistically, open source software is what can and probably should adapt.
They don't need to support or even know about every possible OS. What I'd like from them, and from every hardware maker, is a clean, open, documented interface. What we get is machine full of dirty secretive little hacks to make it just barely work with Windows. I doubt that IBM published the following:
On powerup, the computer looks for a partition of type 165, the recovery partition. We chose this number randomly and did not bother to see if it's being used for anything. If it finds an overly large partition of type 165, the computer hangs forever. The number 165 is hard-coded into the BIOS and can't be changed.
If they had to document the computer before selling it, I think they'd avoid nasty hacks like this out of sheer embarassment.
As before, computer-implemented inventions can be patented if they involve a new and inventive technical contribution to the state of the art. Technical solutions for use in data processing or for carrying out methods of doing business therefore remain patentable.
Oh, good. Technical solution for use in data processing: Access a database from a GUI across a network. (The substance of an obnoxious patent discussed here previously). Technical solution for carrying out methods of doing business: One-click ordering. What are we celebrating again?
Your logical chain is good until the last link. That's where I have a doubt:
And of course once we have translated it and released, everyone can have it and believe me, we will.
You're going to put an mp3 on the internet that is the cleanest possible encoding of the song given that it's already been through a different codec and the player's cheap D/A conversion. (No, I don't see the music industry allowing an unencrypted digital output!) I could play your mp3 on my soundblaster awe64 and probably be happy. There's a watermark in there, but I can't hear it and my hardware doesn't read it. But when SDMI-compliant soundcards become the norm, Joe Schmoe who bought his computer at Circuit City will find your mp3 impossible to play. Over time, this is meant to marginalize and eventually destroy mp3. Also, the strength of the current mp3 scene is that ripping/encoding is easy and doesn't require special equipment or skills. If the percentage of the "mp3 community" producing mp3's is drastically reduced, we'll have a lot less mp3s and it will be easier to demonize and shut down the remaining workers.
The only possible way to encrypt any sort of content that is intended for mass-distribution is by encrypting it on a per user basis.
What about DVD? They only need one public/private key pair per player manufacturer. Yes it was cracked, but a version of this scheme in tamper-resistant hardware would be hard to crack. The idea of one key per user is only necessary if users control their hardware. If the content cartel controls the hardware, they have no theoretical need for more than 1 key. However they'll create a bunch so that if one device is cracked they can stop including that key in future recordings.
I think the idea is to put a watermark that includes your name in all files that you purchase from the Big5.
I used to think that, but from reading SDMI's docs it seems they have a different plan, which doesn't involve customizing the data for each user. In SDMI, the watermark identifies the 'business rules' that apply to a recording. SDMI-compliant hardware won't perform an operation that violates the business rules. Non-compliant hardware won't be able to play the data at all.
My understanding is that SDMI will only trust hardware, not software. Hardware can be made very hard to reverse engineer; software can't. So, when SDMI sound cards become widespread, they'll probably release the specs so Open Source drivers can be written. It won't compromise SDMI because the access control decision will be made in the same chip that does the decryption and codec functions. If they license a single software player, their scheme will come crashing down immediately.
It seems that when people demand a "modern" look and feel, they're often looking for a MS look and feel. I'd much rather use a Motif app than something that looks like Bill Gates had a hand in making it. I'd like GUI apps to get rid of menu bars and icon bars. Use popup menus like xterm does. Don't waste pixels. MS Windows looks wrong, diseased, rancid. It's hard to analyze the impression, but part of it comes from the disgusting proportional font MS uses for most controls. Give me misc-fixed-medium. The Mac often has beautiful widgets. Unix has a cross-section of every possible approach. I love the scrollbars used on xterm and xfig - so clean and logical. What's the modern contribution? Reduce the contrast between "thumb" and background, making the eye work harder to resolve the thumb. Add little buttons at the top and bottom. Reduce 3-button functionality to 1-button. Add klassy "3d" look. I'm sad that there's so much momentum toward a desktop look that's "straight outa Redmond".
And good riddance to Illustrator and PageMaker too.
Adobe did not write Frame. They merely bought it and confusingly renamed it FrameMaker. Adobe does not understand Frame or the market that uses it. Don't lump Frame in with Adobe's efforts.
...open-source developers can have the Linux market to themselves.
The problem is that few people, in the linux world or anywhere else, understand what Frame is and why it's important. It's a powerful tool for creating structured documents, but it's confused with word processors, dtp programs, etc. Frame is enough of a paradigm shift that I wish Linux users could have it for a few years to learn the concepts - which I think would sit quite well with Unix hackers. Open Source is good at chasing tail-lights, but we only chase tail-lights we know about. Unfortunately, everyone knows about Windows/MS Office, so we'll see 10^6 clones of this crappy software. If Frame were as widely visible, we'd have seen some good clones of it already. I wish I had downloaded the Beta. Disclaimer: I know almost nothing about Frame.
Sounds like a good idea. I wonder if a relational database could be used for the back end. I believe Frame documents are basically collections of entities that reference each other through ID numbers. An RDBMS-based Frame would offer some benefits:
Multiple users could work on the same document concurrently without fear of corruption.
The API is wide open and well defined. The existing db monitor provides an excellent tool for inspecting document structure and debugging.
Of course, we'd need.mif import and export scripts. Please don't reinvent too many wheels. Desktop programmers seem to love spending time making inferior embedded databases that will never measure up to a real RDBMS for speed or clarity. I hate it when this PeeCee mentality creeps into Unix. So here's how I'd decompose it - sounds complex but it's meant to minimize the pain:
Client - C or C++ or even Java. Provides GUI and approximated rendering on the display.
Server - Perl or Python. Holds DB handles open and provides an abstraction layer for client db actions. Launches import/export/print/view programs per client direction.
Printing Engine - Perl or Python. Queries db and outputs Postscript. Or perhaps TeX - it might be possible to use an excellent layout engine instead of reinventing it badly.
This makes it extremely hard to do things like "extract page 10", because you have to run the program until it outputs 9 pages, somehow defeating the actual output, wait for it to request output for the 10th time, capture the raster memory, then kill the program.
Most postscript documents follow the Document Structure Conventions. This means a bunch of comments are inserted splitting the file into sections, including a chunk of code that's run for each new page, and the individual pages. This makes it easy to post-process the document.
Alice creates an e-commerce website. Like most websites, it has a copyright notice at the bottom of every page. Some or all of the pages are delivered via an access control mechanism. Specifically, these "protected pages" are viewed via a Java applet which effectively prevents the user from saving, printing, or copy-and-pasting the "protected pages". Any of the following can occur:
Alice advertises on TV: "Sign up now with my service for only $5 and get a free cordless phone!" Bob signs up and pays Alice $5. When he clicks on the link to get his free phone, he receives a lengthy "protected" page describing the terms of the offer - terms which render the offer effectively worthless. Carol sends email to Bob asking if Alice's web site is a good deal. To support his answer, Bob would like to copy a portion of the protected page. Carol can't access it directly because she's not a member. Bob can only report his subjective impression that Alice's website is not a good deal, without offering evidence. Next Bob contacts the police and accuses Alice of fraud. But he has no concrete evidence of Alice's refusal to furnish the free cordless phone, because he cannot print, save, or copy-and-paste the "protected" document. When Alice hears that Bob has gone to the police, she stops the advertising campaign and modifies the "protected" page so that it no longer violates the law. Although millions of people have read the document, she is able to instantly destroy all copies of it, leaving no evidence of her fraud.
Alice wants investors to buy stock in her.com startup, so she creates a "protected document" hyping the stock. The document breathlessly exaggerates the company's financial prospects and blatantly violates SEC rules. Bob, Carol, and Dave buy lots of stock. Bob would like to save a copy of the document but the access control mechanism prevents him. And any tool that would help circumvent the access control is illegal. A week later, Alice replaces the document with one conforming to SEC requirements.
I'm noticing some recurring confusion in the comments about this product.
Isn't this the same as a search engine? No, a search engine stores indexing information in a central repository and makes it available to users via a web interface. This product is more of a web client library, like Perl's LWP, intended to make HTTP requests to non-cooperating sites and organize the resulting hodge-podge of HTML into usable data.
Isn't "screen scraping" kind of lame when we could use XML/RDF/semantic markup/other buzzword? No. The problem is that you, the data gatherer do not control the web sites from which you're gathering data. You might wish that greedy.com made their data available in a friendly format, but what incentive do they have to do this? They're interested in getting eyeballs glued to their site, not in feeding your big data harvester. In fact, when high bandwidth becomes widely available, maybe commercial sites will start delivering whole pages as dynamically generated GIF's. There's a fundamental conflict of interest between the commercial web publisher and the data consumer. The publisher wants to dilute his teaspoon of info in a bucket of glitz and junk. The user wants to refine the messy end-product and extract the data, whether he uses his brain or software to do it.
I'll avoid the obvious, since you already know what a good calendaring app does (I hope.)
Synchronous, "real-time" architecture based on either TCP connections always open from client to server, or UDP packets to update status. What I see on my workstation should never be out of sync with the 'real calendar'. Microsoft's message-based architecture is the wrong way.
Windows client that supports both synchronous on-line operation and disconnected operation (for notebooks). Of course if you make an appointment while disconnected, it stays 'tentative' until you sync, and may get rejected then. I don't like Windows or notebooks, but the reality is that the heaviest calendaring users use both. It would be nice to have a similarly featureful Unix client, but realistically most technical people don't like or use calendaring much, while calendaring is the lifeblood of suit/PHB types. The app lives or dies by how well it works for the suits.
Outlook compatibility. For now. Ultimately I assume Microsoft will twist Outlook's protocols to make this impossible.
Scalability without server sync issues. The existing calendaring apps seem to create 'separate universes' ala Everquest for people on different servers. Therefore you might think you've scheduled a scarce resource, and then find out later when the servers sync up that it was already taken. I think the preferrable architecture is one big database server with a ring of application servers around it.
Open API for non-trusted users. There should be a way for people other than the calendar admins to easily script reports, etc. without compromising security. It could just be a simple web interface that doesn't change much. This could also be used to write better clients. It goes without saying that you'd provide an open API for the admins to script to.
Fast response. Sounds obvious, but some calendaring apps are sluggish at enterprise sizes.
Transactional integrity - two users should never see contradictory information, such as two people thinking they've booked the same room. The only ambiguity allowed is on disconnected laptops, where the ambiguous status is clearly shown to the user.
Web access. I guess that's obvious, but please don't use Java or JavaScript or unnecessary graphics. Concentrate on speed, simplicity, and clarity.
Again, scalability. It's really easy to make a calendaring system that works on the small scale and can't scale. Once you're supporting far more users than can connect to one machine, you start wishing you had a good architecture. Maybe you should develop the app on a cluster of 486's to force scalability from the beginning.
That's about it. There are lots of bells and whistles I could ask for, but all of them can be added if the API is open. Good luck!
I started reading Moorcock's piece, and already I have some complaints. Moorcock is starting with the assumption that it's virtuous to be a "leftist" and a "radical". Then he proceeds to scold his fellow "radicals" for reading books that don't hew to the party line.
there is Ayn Rand, the rabid opponent of trade unionism and the left, who, like many a reactionary before her, sees the problems of the world as a failure by capitalists to assume the responsibilities of 'good leadership'
OK, Rand's "rabid" because she disagrees with you. But if you're calling people crypto-fascists for opposing unions and the left, you're diluting the meaning of fascism a bit, aren't you? And only one of Rand's novels has much to do with capitalists.
there is Tolkein and that group of middle-class Christian fantasists who constantly sing the praises of bourgeois virtues and whose villains are thinly disguised working class agitators
I wonder what in the ring trilogy Moorcock can be thinking of. Is Saruman a working class agitator? Seems more like a twisted professor to me. And why use middle-class and bourgeouis as insults? Is Moorcock claiming membership in some higher or lower class? Of all the interpretations LOTR can be made to support, this has to be the least supportable. It seems that Moorcock is one of these over-politicized characters who imposes his Marxist grid on everything. However the essay has some interesting ideas.
I think it may be a little ridiculous to expect the registar to not be offended by anything. After all, they are human too.
No they're not. Legally they're corporations. Factually, they're networks of computers. If you still think registrars are human and capable of being offended, try registering nsieatsfeces.com. No objections there, because you're dealing with a pattern matcher, not a sentient being. For a human being, a broad-minded, tolerant outlook is something to strive for; intolerance is normal. For a computer, it's just the opposite. Aloof disinterest in the content of strings is normal - prejudice has to be programmed.
I downloaded one image of an 'indulgence' from the site. The page is beautiful. But I was very irritated to see at the bottom:
(C)2000 British Library Board
This is just wrong. The work of scanning the page is so much less than the work of creating it. In addition, the content was written by the Pope and meant to be widely known. It bothers me that the British Library Board would claim intellectual property rights in a document of that nature. Then I downloaded a page of the actual Bible and saw that instead of a copyright notice, it bears a simple notice of who scanned it and when. Much better. What if the Church had been able to use copyright law to prevent the publication of the Gutenberg Bible? Maybe they could have prevented the reformation. Maybe this is an argument that could show conservatives why intellectual property is not good. PS - I notice they're running IIS and have that odd microsoftish need to shorten file/directory names. The pioneer in question was, as near as I can render it in US-ASCII, Gutenberg, not "gutenbg". Are they saving keystrokes?
Plaintiffs allege that (the defendant) has covertly, without consent or authorization, planted 'cookies' upon Internet users' computer hard disk drives...
It's tempting to rejoice at these marketing bastards getting their comeuppance, but on balance I hope this suit fizzles. Here's the transaction they're really talking about:
Plaintiff requests an image from Defendant.
Defendant transmits the image, including in his response a 'Cookie' header, which is a valid header in HTTP, the language which Plaintiff and Defendant are speaking.
Plaintiff records this 'Cookie'.
Time goes by.
Plaintif requests a different image from Defendant. This time, Plaintiff includes the 'Cookie' header previously furnished by Defendant.
Defendant deduces that both requests came from the same "person".
I realize that most of this transaction was transparently performed by the plaintiff's web browser. But that's not the defendant's fault. I would really like the courts/legislature to validate the following principle:
If you send a request to party X via a well known Internet protocol, and party X sends you a response conforming to that protocol, you can't sue or prosecute party X for so responding.
Slashdot Mirror
- sudo - Nobody should have root on any workstations. Sysadmins should have appropriately tailored sudo rights. If you have heavy-duty applications, the application administrators should be different from the platform admins. Naturally, the vast majority of boxes will never be logged into.
- snmpd - Your central monitoring server can constantly sweep the workstations and collect memory, load and other stats. Proactively solve problems before a user notices.
- ssh - With appropriate keys, it can be used by automated scripts to collect info or change configuration on all the workstations.
- rsync - Rides on top of ssh to distribute new software.
- remote re-imaging - there are several ways to automate this. If a workstation gets utterly hosed, it's not worth troubleshooting. Sorry if that sound un-unixy. You want to easily re-image without user intervention. Ideally, you want the hell-desk to be able to do this from a web page. This is a good reason to mount home directory over NFS, but I know you're asking about the un-nfs solutions.
- strong central database - your 6000 machines won't be uniform. You'll have different ages, models, configurations, OS levels. The database keeps track of what build of every software package is on each machine, hw configurations, and history. Without this you're flying blind.
- test pool and pilots - You need a lab with representative workstations to test new software on. Then roll out to a small subset of the population - 'pilot' users who are good at reporting problems. When that works, you can roll out to the whole machine population. That database keeps track of the rollout state so if the script is interrupted partway through, you know where to resume.
In real life, you don't get to plan installations like this and make them perfect. You inherit what's there and try to improve it.I guess you're referring to MS's impressive performance on TPC benchmarks. I don't mean to detract from Microsoft's accomplishment in setting new speed records, I think these scores have less linkage to reality than MS suggests.
These results have little relevance to the average enterprise - I don't think the City of Virginia Beach needs a 24-way SMP machine to process their parking tickets or whatever. The basic reasons linux doesn't show up on these benchmarks are lack of scalable SMP and lack of funding to participate.
These results have little relevance to the high end, because people who architect this stuff are quite conservative and prefer proven platforms like Sun E-series, HP's, Sequents. Neither Linux nor NT really has anything to offer this market. Lower cost per transaction? Great, until the first avoidable outage. Then the machine will eat all its savings in 30 seconds.
This seems to be a fundamental difference in thinking that pops up on slashdot in threads like these. The opposite of your viewpoint is the one that states "Use the right tool for the job." The split could be called idealistic vs. pragmatic. But this doesn't mean the idealist is impractical. The idealist wants what's best for him in the long run while the pragmatist seeks to maximize pleasure (or more likely minimize pain) today.
Personally, I agree with you. Every time I see the "right tool" idea advocated, I want to respond, "where would you be today if people had always followed that advice?" Chamberlain did the pragmatic thing by appeasing Hitler, achieved "peace in our time" and was a temporary hero. Now his memory is a disgraceful one.
The issue isn't really hardware vs software. It's 'trusted host' vs 'anyone else'. Alice trusts the Player because she thinks Bob can't take apart a chip.
Given sufficient resources in Bob's hands, this is true. But there's only one hacker who can reverse engineer a VLSI chip for a great many who can reverse engineer a program. If tamper-resistance is built into the chip, as the NSA did with Clipper, reverse engineering could become enormously expensive and hard.
I worked on a billion-dollar project where all accounting was done by a huge interlinked maze of Excel spreadsheets. Discrepancies of several hundred dollars were common. The rule was, any gap less than $1k is not worth auditing.
Apparently, thinkpads used partition type 160 (a0) until now. FreeBSD used 165 (a5).
Messing around is exactly what IBM did. Instead of sticking with the previously used partition type, they changed to one that conflicts with another known use.
Or return the Thinkpads and don't buy more. Too bad IBM didn't bother to go to google and search for "partition list" and read the first hit.
Oh, good.
Technical solution for use in data processing: Access a database from a GUI across a network. (The substance of an obnoxious patent discussed here previously).
Technical solution for carrying out methods of doing business: One-click ordering.
What are we celebrating again?
You're going to put an mp3 on the internet that is the cleanest possible encoding of the song given that it's already been through a different codec and the player's cheap D/A conversion. (No, I don't see the music industry allowing an unencrypted digital output!) I could play your mp3 on my soundblaster awe64 and probably be happy. There's a watermark in there, but I can't hear it and my hardware doesn't read it.
But when SDMI-compliant soundcards become the norm, Joe Schmoe who bought his computer at Circuit City will find your mp3 impossible to play. Over time, this is meant to marginalize and eventually destroy mp3.
Also, the strength of the current mp3 scene is that ripping/encoding is easy and doesn't require special equipment or skills. If the percentage of the "mp3 community" producing mp3's is drastically reduced, we'll have a lot less mp3s and it will be easier to demonize and shut down the remaining workers.
My understanding is that SDMI will only trust hardware, not software. Hardware can be made very hard to reverse engineer; software can't. So, when SDMI sound cards become widespread, they'll probably release the specs so Open Source drivers can be written. It won't compromise SDMI because the access control decision will be made in the same chip that does the decryption and codec functions.
If they license a single software player, their scheme will come crashing down immediately.
It seems that when people demand a "modern" look and feel, they're often looking for a MS look and feel. I'd much rather use a Motif app than something that looks like Bill Gates had a hand in making it. I'd like GUI apps to get rid of menu bars and icon bars. Use popup menus like xterm does. Don't waste pixels.
MS Windows looks wrong, diseased, rancid. It's hard to analyze the impression, but part of it comes from the disgusting proportional font MS uses for most controls. Give me misc-fixed-medium.
The Mac often has beautiful widgets. Unix has a cross-section of every possible approach. I love the scrollbars used on xterm and xfig - so clean and logical. What's the modern contribution? Reduce the contrast between "thumb" and background, making the eye work harder to resolve the thumb. Add little buttons at the top and bottom. Reduce 3-button functionality to 1-button. Add klassy "3d" look.
I'm sad that there's so much momentum toward a desktop look that's "straight outa Redmond".
The problem is that few people, in the linux world or anywhere else, understand what Frame is and why it's important. It's a powerful tool for creating structured documents, but it's confused with word processors, dtp programs, etc. Frame is enough of a paradigm shift that I wish Linux users could have it for a few years to learn the concepts - which I think would sit quite well with Unix hackers. Open Source is good at chasing tail-lights, but we only chase tail-lights we know about.
Unfortunately, everyone knows about Windows/MS Office, so we'll see 10^6 clones of this crappy software. If Frame were as widely visible, we'd have seen some good clones of it already.
I wish I had downloaded the Beta.
Disclaimer: I know almost nothing about Frame.
- Multiple users could work on the same document concurrently without fear of corruption.
- The API is wide open and well defined. The existing db monitor provides an excellent tool for inspecting document structure and debugging.
Of course, we'd needPlease don't reinvent too many wheels. Desktop programmers seem to love spending time making inferior embedded databases that will never measure up to a real RDBMS for speed or clarity. I hate it when this PeeCee mentality creeps into Unix. So here's how I'd decompose it - sounds complex but it's meant to minimize the pain:
Carol sends email to Bob asking if Alice's web site is a good deal. To support his answer, Bob would like to copy a portion of the protected page. Carol can't access it directly because she's not a member. Bob can only report his subjective impression that Alice's website is not a good deal, without offering evidence.
Next Bob contacts the police and accuses Alice of fraud. But he has no concrete evidence of Alice's refusal to furnish the free cordless phone, because he cannot print, save, or copy-and-paste the "protected" document. When Alice hears that Bob has gone to the police, she stops the advertising campaign and modifies the "protected" page so that it no longer violates the law. Although millions of people have read the document, she is able to instantly destroy all copies of it, leaving no evidence of her fraud.
It seems that Moorcock is one of these over-politicized characters who imposes his Marxist grid on everything. However the essay has some interesting ideas.
For a human being, a broad-minded, tolerant outlook is something to strive for; intolerance is normal. For a computer, it's just the opposite. Aloof disinterest in the content of strings is normal - prejudice has to be programmed.
This is just wrong. The work of scanning the page is so much less than the work of creating it. In addition, the content was written by the Pope and meant to be widely known. It bothers me that the British Library Board would claim intellectual property rights in a document of that nature.
Then I downloaded a page of the actual Bible and saw that instead of a copyright notice, it bears a simple notice of who scanned it and when. Much better.
What if the Church had been able to use copyright law to prevent the publication of the Gutenberg Bible? Maybe they could have prevented the reformation. Maybe this is an argument that could show conservatives why intellectual property is not good.
PS - I notice they're running IIS and have that odd microsoftish need to shorten file/directory names. The pioneer in question was, as near as I can render it in US-ASCII, Gutenberg, not "gutenbg". Are they saving keystrokes?
- Plaintiff requests an image from Defendant.
- Defendant transmits the image, including in his response a 'Cookie' header, which is a valid header in HTTP, the language which Plaintiff and Defendant are speaking.
- Plaintiff records this 'Cookie'.
- Time goes by.
- Plaintif requests a different image from Defendant. This time, Plaintiff includes the 'Cookie' header previously furnished by Defendant.
- Defendant deduces that both requests came from the same "person".
I realize that most of this transaction was transparently performed by the plaintiff's web browser. But that's not the defendant's fault.I would really like the courts/legislature to validate the following principle: